# Robot Navigation

Robot navigation is the ability of a robot to determine its position and plan a path to a goal, using representations such as maps and techniques including computer vision algorithms.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## PGX: Physically Grounded Execution for Long-Horizon Humanoid Loco-Manipulation

DevFeed: [PGX: Physically Grounded Execution for Long-Horizon Humanoid Loco-Manipulation](<https://devfeed.tech/articles/pgx-physically-grounded-execution-for-long-horizon-humanoid-loco-manipulation-145692.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.26596>)

Author: Kai Lu, Chenyang Ma, Ioannis Havoutis

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Humanoid Robots](<https://devfeed.tech/topics/humanoid-robots.md>), [Robot Manipulation](<https://devfeed.tech/topics/robot-manipulation.md>), [Robotics](<https://devfeed.tech/topics/robotics.md>), [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>)

Tags: [actions](<https://devfeed.tech/tags/actions.md>), [base](<https://devfeed.tech/tags/base.md>), [based](<https://devfeed.tech/tags/based.md>), [components](<https://devfeed.tech/tags/components.md>), [consistency](<https://devfeed.tech/tags/consistency.md>), [control](<https://devfeed.tech/tags/control.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [execution](<https://devfeed.tech/tags/execution.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [flat](<https://devfeed.tech/tags/flat.md>), [focus](<https://devfeed.tech/tags/focus.md>), [framework](<https://devfeed.tech/tags/framework.md>), [gate](<https://devfeed.tech/tags/gate.md>), [grounded](<https://devfeed.tech/tags/grounded.md>), [horizon](<https://devfeed.tech/tags/horizon.md>), [humanoid](<https://devfeed.tech/tags/humanoid.md>), [humanoid-robot](<https://devfeed.tech/tags/humanoid-robot.md>), [improvements](<https://devfeed.tech/tags/improvements.md>), [language](<https://devfeed.tech/tags/language.md>), [manipulation](<https://devfeed.tech/tags/manipulation.md>)

### AI overview

PGX addresses long-horizon humanoid loco-manipulation by separating task progression and spatial precision from VLA skill execution. A predicate-based state machine gates phase changes, triggers recovery, and coordinates navigation and waiting; shared predicate semantics also link demonstration segmentation, execution, and stage-wise evaluation. Within each active manipulation phase, flat or stage-trained VLA models focus on relevant actions. Training-free world-coordinate locking maintains spatial consistency through base station keeping, object-relative positioning, and world-anchored contact targets. Experiments on three simulated tasks and one real-world task with a Unitree G1 report improved stage progression and task success rates; the excerpt provides no quantitative results.

### Source excerpt

arXiv:2611.26596v1 Announce Type: cross Abstract: Robotic vision-language-action (VLA) models can execute diverse individual manipulation skills, yet they still struggle with long-horizon loco-manipulation tasks, where temporal structure, skills, and precision are tightly coupled within a monolithic model. We propose Physically Grounded Execution (PGX), a framework that offloads temporal and precision control to two classical training-free components, allowing the VLA to focus on stage-specific skill dynamics. PGX decomposes long-horizon loco-manipulation into three layers: (i) a predicate-based state machine governs task progression, using physical predicates to gate phase transitions, trigger recovery, and coordinate navigation and waiting. Shared predicate semantics connect offline demonstration segmentation, online execution, and stage-wise evaluation; (ii) VLA skills operate within active manipulation phases, with flat or stage-trained models concentrating on task-relevant actions; and (iii) world-coordinate locking (WCL) maintains spatial consistency through base station keeping, object-relative positioning, and world-anchored contact targets. We demonstrate the effectiveness of PGX with extensive experiments conducted on three simulated tasks and one real-world task using a Unitree G1 humanoid robot for long-horizon loco-manipulation, showing improvements in stage progression and task success rates. Project page is available at: https://kl-research.github.io/pgx

## Bridging vision-language reasoning and learned control for legged navigation in the wild

DevFeed: [Bridging vision-language reasoning and learned control for legged navigation in the wild](<https://devfeed.tech/articles/bridging-vision-language-reasoning-and-learned-control-for-legged-navigation-in-the-wild-145484.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25215>)

Author: Takahiro Miki, Robin Schmid, Manthan Patel, Turcan Tuna, Julia Richter, Julian Nubert, Joonho Lee, Matthias M\\"uller, Marco Hutter

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>), [Robotics](<https://devfeed.tech/topics/robotics.md>), [vlm](<https://devfeed.tech/topics/vlm.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>)

Tags: [autonomous-navigation](<https://devfeed.tech/tags/autonomous-navigation.md>), [control](<https://devfeed.tech/tags/control.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [legged-robots](<https://devfeed.tech/tags/legged-robots.md>), [navigation](<https://devfeed.tech/tags/navigation.md>), [planning](<https://devfeed.tech/tags/planning.md>), [real-time](<https://devfeed.tech/tags/real-time.md>), [real-world](<https://devfeed.tech/tags/real-world.md>), [reasoning](<https://devfeed.tech/tags/reasoning.md>), [reinforcement-learning](<https://devfeed.tech/tags/reinforcement-learning.md>), [reliability](<https://devfeed.tech/tags/reliability.md>), [risk](<https://devfeed.tech/tags/risk.md>), [robotics](<https://devfeed.tech/tags/robotics.md>), [spatial-reasoning](<https://devfeed.tech/tags/spatial-reasoning.md>), [vision](<https://devfeed.tech/tags/vision.md>)

### AI overview

ROAM combines a vision-language model (VLM) with two reinforcement-learning policies to help legged robots navigate complex alpine and urban environments. The VLM interprets scenes, assesses risks, and assigns semantic navigation costs; open-vocabulary segmentation turns these labels into a real-time risk map that conditions the policies for locomotion and local navigation. Reported real-world trials include kilometer-long hikes with minimal intervention, interpreting signs to avoid danger, and identifying safer places to step. The abstract also flags VLM reliability and spatial reasoning as unresolved challenges, so foundational models are useful components but not a complete solution for navigation in the wild.

### Source excerpt

arXiv:2611.25215v1 Announce Type: cross Abstract: Autonomous navigation in diverse real-world environments remains one of the greatest challenges in robotics. It demands both robust low-level motion control to handle diverse terrain and high-level reasoning to interpret environmental cues. We present ROAM (Robust Open-world Autonomous Mobility), a system that enables legged robots to perform kilometer-scale navigation in complex alpine and urban settings. ROAM combines the world knowledge of a Vision-Language Model (VLM), pre-trained on internet-scale data, with robust locomotion and local planning policies learned in simulation via reinforcement learning (RL). The VLM provides common-sense reasoning and semantic interpretation of the scene, assessing risks, interpreting signs, and producing semantic labels with associated navigation costs. An open-vocabulary segmentation module converts these labels into a dense, real-time risk map, which conditions two pre-trained RL policies responsible for agile locomotion and local navigation, effectively coupling high-level VLM reasoning with low-level physical control. Real-world experiments showed the system's ability to autonomously navigate a wide range of complex real-world environments. The robot completes kilometer-long hikes with minimal intervention, interprets signs to avoid danger, and evaluates its environment to identify safe areas on which to step. This work represents a step toward the reliable deployment of systems in unstructured environments. However, it also underscores the challenges that remain in VLM reliability and spatial reasoning, indicating that foundational models are a crucial component but not yet a complete solution for navigation in the wild.

## Plan your Self-Adaptation! Ontology-Mediated Task and Architecture Co-adaptation Planning for Robots

DevFeed: [Plan your Self-Adaptation! Ontology-Mediated Task and Architecture Co-adaptation Planning for Robots](<https://devfeed.tech/articles/plan-your-self-adaptation-ontology-mediated-task-and-architecture-co-adaptation-planning-for-robots-145490.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25254>)

Author: Gustavo Rezende Silva, Tobias John, Francisco Mart\\'in, EinarR Broch Johnsen

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Robotics](<https://devfeed.tech/topics/robotics.md>), [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>)

Tags: [architecture](<https://devfeed.tech/tags/architecture.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [dependencies](<https://devfeed.tech/tags/dependencies.md>), [execution](<https://devfeed.tech/tags/execution.md>), [functional](<https://devfeed.tech/tags/functional.md>), [goals](<https://devfeed.tech/tags/goals.md>), [inspection](<https://devfeed.tech/tags/inspection.md>), [language](<https://devfeed.tech/tags/language.md>), [memory](<https://devfeed.tech/tags/memory.md>), [mission](<https://devfeed.tech/tags/mission.md>), [navigation](<https://devfeed.tech/tags/navigation.md>), [ontology](<https://devfeed.tech/tags/ontology.md>), [planning](<https://devfeed.tech/tags/planning.md>), [requirements](<https://devfeed.tech/tags/requirements.md>), [results](<https://devfeed.tech/tags/results.md>), [robot](<https://devfeed.tech/tags/robot.md>), [robots](<https://devfeed.tech/tags/robots.md>), [runtime](<https://devfeed.tech/tags/runtime.md>), [time](<https://devfeed.tech/tags/time.md>)

### AI overview

PLANTA jointly plans a robot's mission tasks and architectural reconfigurations because each constrains the other: architecture determines available actions, while the mission determines needed capabilities. It represents architectural options and functional and quality requirements in a Web Ontology Language ontology, and tasks and goals in Planning Domain Definition Language; automatic translation and integration enable joint planning. PLANTA uses satisficing planning and replans when execution changes invalidate the current plan. Evaluations covered underwater inspection and ground navigation. In the navigation scenario, it used less planning time and memory than optimal-policy generation while producing plans of similar quality. The supplied text is an abstract and does not provide further evaluation details.

### Source excerpt

arXiv:2611.25254v1 Announce Type: cross Abstract: Autonomous robots may need to adapt both their task plans and architectural configurations as operating conditions change. These decisions are interdependent: architectural configurations determine which actions a robot can perform, while task plans determine which capabilities are required. This paper presents PLANTA, an ontology-mediated approach to task and architecture co-adaptation that considers these dependencies during mission planning. PLANTA uses an ontology expressed in the Web Ontology Language to represent architectural alternatives and their associated functional and quality requirements, while the Planning Domain Definition Language defines task actions and mission goals. By automatically translating and integrating these models, PLANTA enables joint planning of task and reconfiguration actions. The approach uses satisficing planning and replans when changes during execution invalidate the current plan. PLANTA is evaluated in two robotic scenarios involving underwater inspection and ground navigation. The results support the feasibility of runtime co-adaptation and show lower planning time and memory consumption compared with optimal-policy generation in the evaluated navigation scenario, while producing plans of similar quality.

## CoachSI: Self-Improvement of Robot Policies with a Language Coach

DevFeed: [CoachSI: Self-Improvement of Robot Policies with a Language Coach](<https://devfeed.tech/articles/coachsi-self-improvement-of-robot-policies-with-a-language-coach-145529.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25529>)

Author: Jiachen Tao, Gengyu Zhang, Haoxuan Wang, Bing Liu, Yan Yan

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>), [vlm](<https://devfeed.tech/topics/vlm.md>), [post-training](<https://devfeed.tech/topics/post-training.md>)

Tags: [camera](<https://devfeed.tech/tags/camera.md>), [coaching](<https://devfeed.tech/tags/coaching.md>), [completion](<https://devfeed.tech/tags/completion.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [fine-tuning](<https://devfeed.tech/tags/fine-tuning.md>), [horizon](<https://devfeed.tech/tags/horizon.md>), [improvement](<https://devfeed.tech/tags/improvement.md>), [instructions](<https://devfeed.tech/tags/instructions.md>), [language](<https://devfeed.tech/tags/language.md>), [language-model](<https://devfeed.tech/tags/language-model.md>), [model](<https://devfeed.tech/tags/model.md>), [naming](<https://devfeed.tech/tags/naming.md>), [next](<https://devfeed.tech/tags/next.md>), [policies](<https://devfeed.tech/tags/policies.md>), [policy](<https://devfeed.tech/tags/policy.md>), [robot](<https://devfeed.tech/tags/robot.md>), [self-improvement](<https://devfeed.tech/tags/self-improvement.md>), [vision](<https://devfeed.tech/tags/vision.md>), [vision-language-action](<https://devfeed.tech/tags/vision-language-action.md>), [vision-language-model](<https://devfeed.tech/tags/vision-language-model.md>), [vlm](<https://devfeed.tech/tags/vlm.md>)

### AI overview

CoachSI improves long-horizon robot task performance by having a frozen vision-language model watch the robot and name the next step in the policy's atomic-instruction format; it never supplies actions. Successful coached runs are relabeled with the original task instruction and used to fine-tune the policy, which is then deployed without a coach. The action backbone remains frozen. On stalled states, next-step coaching raises subgoal completion 4.6-fold over the unchanged task instruction. Unassisted success in unseen scenes rises from 47.4% to 59.3%, exceeding human-demonstration fine-tuning by 9.8 percentage points and equal-budget self-collected-success fine-tuning by 7.3 points. With a coach limited to camera images, success reaches 56.8%, above both baselines.

### Source excerpt

arXiv:2611.25529v1 Announce Type: cross Abstract: Generalist robot policies execute atomic skills reliably but fail when the same skills are composed into long-horizon tasks. Self-improvement from the policy's own experience does not close this gap: experience collected by the policy alone stays within its own distribution and reproduces what it can already do. We show that what the policy lacks is not a skill but the next step. We present CoachSI, a self-improvement method in which a frozen, off-the-shelf vision-language model (VLM) coaches the policy in language: it watches the robot and names the next step in the format of the policy's atomic instructions, and never outputs an action. Successful coached rollouts are relabeled in hindsight with the original task instruction and used to fine-tune the policy. The fine-tuned policy is then deployed on its own. On states where a World Action Model and a Vision-Language-Action model have stalled, naming the next step raises subgoal completion by 4.6x over the unchanged task instruction, and the frozen VLM supplies it from the same camera images the policy sees. After fine-tuning, unassisted success on long-horizon tasks in unseen scenes rises from 47.4% to 59.3%, 9.8 points above fine-tuning on human demonstrations and 7.3 points above fine-tuning on self-collected successes at an equal data budget. With a coach that sees only camera images, it reaches 56.8%, above both baselines. The gain is obtained with the action backbone frozen: language coaching teaches the policy what should happen next, and the skills it already has do the rest.

## Collision-Free Is Not Enough: Passage-Conditioned Navigation for Wheeled-Legged Robots

DevFeed: [Collision-Free Is Not Enough: Passage-Conditioned Navigation for Wheeled-Legged Robots](<https://devfeed.tech/articles/collision-free-is-not-enough-passage-conditioned-navigation-for-wheeled-legged-robots-145587.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25788>)

Author: Wai Ka Sun, Waiyiu Pong, Bowen Wang

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>), [navigation](<https://devfeed.tech/topics/navigation.md>), [few-shot](<https://devfeed.tech/topics/few-shot.md>), [Reinforcement learning](<https://devfeed.tech/topics/reinforcement-learning.md>)

Tags: [competition](<https://devfeed.tech/tags/competition.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [legged-robots](<https://devfeed.tech/tags/legged-robots.md>), [navigation](<https://devfeed.tech/tags/navigation.md>), [robots](<https://devfeed.tech/tags/robots.md>)

### AI overview

Passage-conditioned navigation records a successful taught crossing for each stair or ledge and, after every local replan, uses an entry gate to make the robot turn or back up and retry when its approach differs by more than 10 cm or 10°; it requires neither a robot model nor additional training. With path, gait schedule, and sensing unchanged, the method increased disturbed ledge crossings within those limits from 24.1% to 88.6%. In contact simulation, robots pushed off course before stairs completed 180 climbs without falling, compared with 9 falls when the stored direction was ignored; after a 1.2 m displacement, they rejoined the route and ended an average 0.14 m from it, versus 1.10 m without rejoining. A real wheeled-legged robot completed six competition runs across 271 m of multi-obstacle terrain and 7.0 m of elevation, finishing fully autonomously.

### Source excerpt

arXiv:2611.25788v1 Announce Type: cross Abstract: A collision-free detour can still bring a legged robot to the next stair or ledge a few centimetres or degrees off, and a fixed locomotion skill, factory or learned, can stall or fall. Existing methods prevent this with a robot model or extensive trial and error, but on a new site with unknown skills neither is available, so they cannot be used. We find that a single 27-minute teaching session already supplies what these methods have to derive or learn: for every stair and ledge, one crossing the skill has completed. Our method, passage-conditioned navigation, makes the route remember how each transition was entered, not only where the path went. It stores each taught crossing as an event and, without training or a robot model, enforces it after every local replan: before each ledge, an entry gate makes the robot turn or back up and retry if it is more than 10 cm or 10{\deg} off the taught crossing. Under localization noise, with path, gait schedule, and sensing unchanged, this raises the share of disturbed ledge crossings that stay within these limits from 24.1% to 88.6%. In contact simulation with our learned policies, a robot pushed off course before stairs never falls in 180 climbs, against 9 falls when the stored direction is ignored; pushed 1.2 m away, it rejoins the route, ending on average 0.14 m from it (1.10 m without rejoining), and finishes as often as an undisturbed robot. This suits any robot with fixed skills that repeats a route over terrain transitions, as in inspection and patrol, where one wrong entry can end the task. We deployed it in competition on a real wheeled-legged robot: over 271 m of multi-obstacle terrain spanning 7.0 m of elevation, it completed all six runs, won third place at the 2026 Global Open-source AI Challenge, which drew 2,999 submissions, and was the only finalist in its task to finish fully autonomously.

## ZeroVLN: Learning to Navigate without Supervised Fine-Tuning of the VLM Brain

DevFeed: [ZeroVLN: Learning to Navigate without Supervised Fine-Tuning of the VLM Brain](<https://devfeed.tech/articles/zerovln-learning-to-navigate-without-supervised-fine-tuning-of-the-vlm-brain-145668.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.26414>)

Author: Zixuan Huang, Yang Zhou, Guli Zhang, Sunzhu Li, Hongyan Xie, Zehao Chen, Fuzhen Zhuang, Kaixuan Wang, Xiaozhi Chen, Yikun Ban, Deqing Wang

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>), [vlm](<https://devfeed.tech/topics/vlm.md>), [Reinforcement learning](<https://devfeed.tech/topics/reinforcement-learning.md>), [Fine-tuning](<https://devfeed.tech/topics/fine-tuning.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>)

Tags: [agentic](<https://devfeed.tech/tags/agentic.md>), [capabilities](<https://devfeed.tech/tags/capabilities.md>), [continuous](<https://devfeed.tech/tags/continuous.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [model](<https://devfeed.tech/tags/model.md>), [natural-language](<https://devfeed.tech/tags/natural-language.md>), [navigation](<https://devfeed.tech/tags/navigation.md>), [reinforcement-learning](<https://devfeed.tech/tags/reinforcement-learning.md>), [sft](<https://devfeed.tech/tags/sft.md>), [tool-calling](<https://devfeed.tech/tags/tool-calling.md>), [vision-language-model](<https://devfeed.tech/tags/vision-language-model.md>)

### AI overview

ZeroVLN targets a tradeoff in vision-language navigation: fine-tuning the VLM brain on navigation data can change its weights and weaken general visual capabilities. Instead, the framework leaves the brain without navigation-domain supervised fine-tuning and gives it a semantic tool-calling interface to send natural-language movement requests to an action expert. The expert is first adapted to execute those requests; closed-loop reinforcement learning then jointly improves the brain's movement decisions and the expert's continuous control. On standard VLN-CE benchmarks, using the same navigation episodes, the authors report substantial gains over representative SFT-based hierarchical methods, with only minor changes to the brain's weights and largely preserved general capabilities.

### Source excerpt

arXiv:2611.26414v1 Announce Type: new Abstract: Representative vision-language navigation (VLN) methods typically couple a vision-language model (VLM) brain with a learnable action head and fine-tune the VLM brain on navigation-domain data to predict geometric goals. However, such adaptation can substantially alter base VLM weights and degrade general visual capabilities. We introduce ZeroVLN, a slow-fast navigation framework that achieves effective navigation with zero navigation-domain SFT of the VLM brain. Rather than teaching the brain to predict geometric goals, we equip it with a semantic tool-calling interface for conveying movement decisions to an action expert in natural language. We first adapt the action expert to carry out these movement requests, then jointly improve the brain's movement decisions and the action expert's continuous execution through closed-loop reinforcement learning (RL). Trained on the same navigation episodes, ZeroVLN substantially outperforms representative SFT-based hierarchical VLN methods on standard VLN-CE (continuous environments) benchmarks, while inducing only minor changes to the brain's weights and largely preserving its general capabilities. To our knowledge, ZeroVLN is the first slow-fast VLN framework with zero navigation-domain SFT of the VLM brain, and the first to jointly optimize its slow and fast policies through closed-loop agentic RL in VLN-CE.

## RePlan-VLA: Progress-Aware Replanning for Long-Horizon Vision-Language-Action Models

DevFeed: [RePlan-VLA: Progress-Aware Replanning for Long-Horizon Vision-Language-Action Models](<https://devfeed.tech/articles/replan-vla-progress-aware-replanning-for-long-horizon-vision-language-action-models-142709.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23843>)

Author: Apoorva Vashisth, Manav Kulshrestha, Damon Conover, Zachary Kingston, Aniket Bera

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Amazon robotics research](<https://devfeed.tech/topics/amazon-robotics-research.md>), [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>)

Tags: [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [execution](<https://devfeed.tech/tags/execution.md>), [horizon](<https://devfeed.tech/tags/horizon.md>), [progress](<https://devfeed.tech/tags/progress.md>), [real-world](<https://devfeed.tech/tags/real-world.md>), [robotic-manipulation](<https://devfeed.tech/tags/robotic-manipulation.md>), [vision-language-action](<https://devfeed.tech/tags/vision-language-action.md>)

### AI overview

RePlan-VLA aims to improve long-horizon robotic manipulation by using a VLA model's short-horizon skills within an execution-aware framework. It monitors robot motion to assess progress and revises the future plan based on observed outcomes, while leaving the underlying VLA policy unchanged and requiring no fine-tuning. The abstract reports success-rate gains over the next strongest baseline of 21.9% on RoboCasa365 and 10.4% on CALVIN, and validation on a Franka Research 3 robot.

### Source excerpt

arXiv:2611.23843v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models have emerged as a promising paradigm for general-purpose robotic manipulation, yet their performance often degrades on long-horizon tasks that require multiple interdependent manipulations. Instead of burdening the VLA policies with long-horizon reasoning, memory, action generation, progress tracking, and error recovery, we argue for leveraging their short-horizon manipulation capabilities within an execution-aware framework. To this end, we introduce \textnormal{\textbf{RePlan-VLA}}, a VLA-agnostic framework that translates a VLA's short-horizon capabilities into reliable long-horizon execution. RePlan-VLA explicitly monitors the robot's motion for progress evaluation, and adaptively revises the future plan based on observed outcomes. We preserve the underlying VLA policy entirely, requiring no additional fine-tuning. Experiments on RoboCasa365 and CALVIN benchmarks show an improvement of 21.9% and 10.4% in success rate over the next strongest baseline. We also validate our approach for real-world deployment on a Franka Research 3 robot.

## AeroRSI: Recursive Self-Improvement for Aerial Vision-Language Navigation

DevFeed: [AeroRSI: Recursive Self-Improvement for Aerial Vision-Language Navigation](<https://devfeed.tech/articles/aerorsi-recursive-self-improvement-for-aerial-vision-language-navigation-142752.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.24077>)

Author: Shiwei Wu, Yuwen Pan, Zhi Li, Yaowen Li, Yan Zhou, Xuemiao Xu, Yu Liu

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Computer vision](<https://devfeed.tech/topics/computer-vision.md>)

Tags: [behavior](<https://devfeed.tech/tags/behavior.md>), [code](<https://devfeed.tech/tags/code.md>), [continuation](<https://devfeed.tech/tags/continuation.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [distilling](<https://devfeed.tech/tags/distilling.md>), [failure](<https://devfeed.tech/tags/failure.md>), [inference](<https://devfeed.tech/tags/inference.md>), [natural-language](<https://devfeed.tech/tags/natural-language.md>), [navigation](<https://devfeed.tech/tags/navigation.md>), [policy](<https://devfeed.tech/tags/policy.md>), [raise](<https://devfeed.tech/tags/raise.md>), [recursive-self-improvement](<https://devfeed.tech/tags/recursive-self-improvement.md>), [reduce](<https://devfeed.tech/tags/reduce.md>), [self-improvement](<https://devfeed.tech/tags/self-improvement.md>), [supervision](<https://devfeed.tech/tags/supervision.md>), [uavs](<https://devfeed.tech/tags/uavs.md>), [vision](<https://devfeed.tech/tags/vision.md>), [vision-language-action](<https://devfeed.tech/tags/vision-language-action.md>), [vla](<https://devfeed.tech/tags/vla.md>)

### AI overview

AeroRSI improves aerial vision-language navigation by having a policy learn from its own flight failures. In each round, it anchors supervision to failures reached by the current policy, tests candidate repairs by continuing the flight from them, then distills verified repairs while preserving behavior elsewhere. This lets even sampled actions and rule-based proposals that fail when directly imitated provide useful supervision. Repeating the process also reveals failures previously hidden because episodes stop at the first failure. On TravelUAV, three rounds increased a supervised 2B-parameter policy's success rate from 46.4% to 67.5% and halved collisions; the paper reports a 10.6-point gain over prior state of the art from the same initialization, without extra human labels or inference cost.

### Source excerpt

arXiv:2611.24077v1 Announce Type: cross Abstract: Vision-language-action (VLA) models enable unmanned aerial vehicles (UAVs) to follow natural-language instructions, yet still fail frequently in closed-loop flight. Correcting these failures requires repairs, i.e., actions that avoid them at the states the policy itself reaches, which human pilots cannot label at scale and rule-based planners cannot supply reliably. In this paper, we present AeroRSI, a recursive self-improvement framework in which an aerial navigation policy iteratively learns from its own failures. Our key insight is that a failing policy should judge its own repairs by flying on from them. Judged in this way, its sampled actions and even rule-based proposals that hurt it when imitated directly become reliable supervision. Moreover, an episode terminates at its first failure, concealing subsequent failures until the earlier ones are repaired. AeroRSI thus renews its supervision in every round, anchoring the failures of the current policy, verifying repairs by its continuation, and distilling them while keeping its behavior elsewhere. On TravelUAV, three rounds raise the success rate of a supervised 2B-parameter policy from 46.4% to 67.5% and reduce collisions by half, surpassing the previous state of the art by 10.6 points from the same initialization without extra human labels or inference cost. Code and models are available at https://github.com/starsholic/AeroRSI.

## RoboBuilder: Agentic Robot Co-Design with Real-world Video Feedback in the Loop

DevFeed: [RoboBuilder: Agentic Robot Co-Design with Real-world Video Feedback in the Loop](<https://devfeed.tech/articles/robobuilder-agentic-robot-co-design-with-real-world-video-feedback-in-the-loop-142711.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23855>)

Author: Ghadi Nehme, Dahyun Nam, Akshay Govind Srinivasan, Faez Ahmed

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Physical AI](<https://devfeed.tech/topics/physical-ai.md>), [3D](<https://devfeed.tech/topics/3d.md>), [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>), [Computer vision](<https://devfeed.tech/topics/computer-vision.md>)

Tags: [3d](<https://devfeed.tech/tags/3d.md>), [3d-printed](<https://devfeed.tech/tags/3d-printed.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [co-design](<https://devfeed.tech/tags/co-design.md>), [coding](<https://devfeed.tech/tags/coding.md>), [coding-agents](<https://devfeed.tech/tags/coding-agents.md>), [course](<https://devfeed.tech/tags/course.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [custom](<https://devfeed.tech/tags/custom.md>), [data](<https://devfeed.tech/tags/data.md>), [design](<https://devfeed.tech/tags/design.md>), [feedback](<https://devfeed.tech/tags/feedback.md>), [in-the-loop](<https://devfeed.tech/tags/in-the-loop.md>), [real-world](<https://devfeed.tech/tags/real-world.md>), [robot](<https://devfeed.tech/tags/robot.md>), [video](<https://devfeed.tech/tags/video.md>)

### AI overview

RoboBuilder evaluates coding agents that co-design robot hardware and programs, build designs that pass simulation, and test them on a physical obstacle course. In the reported experiments, none of 15 initial simulation-passing designs crossed the real course, motivating agents to learn simulator behavior from requested measurements and videos. Across four agents, ten obstacles, and three difficulty levels, agents solved nearly all easy obstacles and produced mechanisms not reached by the search baselines, but struggled with harder levels and the task of one robot crossing every obstacle. Only 66% of simulated passes held in reality across levels. Providing video of a real failure led agents to redesign robots, sometimes correcting the simulator, and removed 72% of level-1 and level-2 failures.

### Source excerpt

arXiv:2611.23855v1 Announce Type: new Abstract: Can AI agents design robots that work in the real world? We introduce RoboBuilder, a benchmark in which coding agents co-design a robot's hardware and program from standard parts, motors and a custom 3D-printed part, and every design that passes in simulation is built and tested on a physical course. Simulation alone is not enough: under public part data, none of 15 initial passing designs crossed the real course, so an agent learns the simulator from measurements it requests and videos of real runs. Across four frontier coding agents, ten obstacles at three difficulty levels, 16,427 simulations of 12,774 distinct robots and 135 physical robots built in an estimated 240 hours, agents solve nearly all easy obstacles and invent mechanisms our search baselines do not reach, but falter at harder levels, when one robot must cross every obstacle, and in the real world, where 66% of simulated passes hold across all levels. Closing the loop helps: shown a video of its real failure, an agent redesigns the robot, sometimes correcting its simulator, and removes 72% of level-1 and level-2 failures. RoboBuilder measures progress toward agents that engineer real machines. Project Page: https://ghadinehme.com/robobuilder.github.io/

## Reinforcement Learning for Large-Scale Lifelong Multi-Agent Path Finding with Shared Spatial Fields

DevFeed: [Reinforcement Learning for Large-Scale Lifelong Multi-Agent Path Finding with Shared Spatial Fields](<https://devfeed.tech/articles/reinforcement-learning-for-large-scale-lifelong-multi-agent-path-finding-with-shared-spatial-fields-143332.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.18186>)

Author: Jean Seong Bjorn Choe

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Reinforcement learning](<https://devfeed.tech/topics/reinforcement-learning.md>), [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>)

Tags: [000](<https://devfeed.tech/tags/000.md>), [abstract](<https://devfeed.tech/tags/abstract.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [applications](<https://devfeed.tech/tags/applications.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [call](<https://devfeed.tech/tags/call.md>), [class](<https://devfeed.tech/tags/class.md>), [complete](<https://devfeed.tech/tags/complete.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [deploy](<https://devfeed.tech/tags/deploy.md>), [execution](<https://devfeed.tech/tags/execution.md>), [faster](<https://devfeed.tech/tags/faster.md>), [fields](<https://devfeed.tech/tags/fields.md>), [large](<https://devfeed.tech/tags/large.md>), [learning](<https://devfeed.tech/tags/learning.md>), [multi-agent](<https://devfeed.tech/tags/multi-agent.md>), [reinforcement-learning](<https://devfeed.tech/tags/reinforcement-learning.md>), [scale](<https://devfeed.tech/tags/scale.md>), [spatial](<https://devfeed.tech/tags/spatial.md>), [with](<https://devfeed.tech/tags/with.md>)

### AI overview

The method handles lifelong multi-agent path finding by computing one shared spatial field across the map per timestep, then using a compact goal-conditioned operator to produce each agent's movement preferences. It trains the complete policy end to end with average-reward reinforcement learning, using throughput as the reward, without planner demonstrations or a separate online path planner. One policy per layout class is trained on small maps and deployed without retraining on larger unseen layouts and agent populations. Across six layouts with 10,000 agents, reported throughput was comparable to or higher than the strongest evaluated baselines; average execution time was below 10 ms per step, over an order of magnitude faster than several established planners.

### Source excerpt

arXiv:2610.18186v2 Announce Type: replace Abstract: Large-scale applications of lifelong multi-agent path finding (LMAPF) call for computationally inexpensive movement decisions under changing traffic conditions and across varied layouts. We propose a navigation policy in which a convolutional network computes a shared spatial field over the map once per timestep and a compact goal-conditioned operator derives agents' movement preferences. The complete policy is trained end-to-end using average-reward reinforcement learning with throughput as the reward, without planner demonstrations or a separate online path planner. For structured and unstructured layouts, we train one policy per class on pools of small maps and deploy each without retraining on larger, unseen layouts and agent populations. Across six layouts with 10,000 agents, our policies achieve throughput comparable to or higher than the strongest evaluated baselines. The proposed method has an average execution time below 10 ms per step, more than an order of magnitude faster than several established planners.

## Towards Pose-Robust Few-Shot Vision-Language-Action Models

DevFeed: [Towards Pose-Robust Few-Shot Vision-Language-Action Models](<https://devfeed.tech/articles/towards-pose-robust-few-shot-vision-language-action-models-142785.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.24360>)

Author: Wenhao Li, Zhongyi Han, Yuhao Wei, Liang Han, Yilong Yin, Liqiang Nie

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>), [few-shot](<https://devfeed.tech/topics/few-shot.md>), [vlm](<https://devfeed.tech/topics/vlm.md>), [Fine-tuning](<https://devfeed.tech/topics/fine-tuning.md>)

Tags: [anchor](<https://devfeed.tech/tags/anchor.md>), [arm](<https://devfeed.tech/tags/arm.md>), [base](<https://devfeed.tech/tags/base.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [camera](<https://devfeed.tech/tags/camera.md>), [canonicalization](<https://devfeed.tech/tags/canonicalization.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [data](<https://devfeed.tech/tags/data.md>), [data-collection](<https://devfeed.tech/tags/data-collection.md>), [estimator](<https://devfeed.tech/tags/estimator.md>), [few-shot](<https://devfeed.tech/tags/few-shot.md>), [fine-tuning](<https://devfeed.tech/tags/fine-tuning.md>), [frame](<https://devfeed.tech/tags/frame.md>), [framework](<https://devfeed.tech/tags/framework.md>), [image](<https://devfeed.tech/tags/image.md>), [language](<https://devfeed.tech/tags/language.md>), [models](<https://devfeed.tech/tags/models.md>), [robust](<https://devfeed.tech/tags/robust.md>), [vision](<https://devfeed.tech/tags/vision.md>), [vision-language-action](<https://devfeed.tech/tags/vision-language-action.md>), [vision-language-action-models](<https://devfeed.tech/tags/vision-language-action-models.md>)

### AI overview

WACO addresses camera and robot-base shifts that disrupt few-shot vision-language-action policies by treating both as changes in the camera-to-base relationship. From one RGB-D frame, it uses the robot arm's forward-kinematic pose as a metric calibration target and the static scene as a world anchor, then canonicalizes the image, state, action, and start pose to the original calibration frames. It requires neither ground-truth extrinsics nor extra data, and fine-tunes the unmodified policy using artifacts from its own estimator. The authors introduce LIBERO-PoseShift, a benchmark with graded camera, base, and combined shifts. Across that benchmark, LIBERO-Plus's camera track, and four UR5e real-world tasks, WACO reportedly improves shifted success by 13.2 points in simulation and 36.7 points on the robot, recovering 90% and 87% of the respective gaps to a ground-truth oracle, without test-time calibration or perturbed training data.

### Source excerpt

arXiv:2611.24360v1 Announce Type: cross Abstract: Fine-tuning a vision-language-action (VLA) model on a few dozen demonstrations is the standard way to put a manipulation policy on a robot, yet the resulting policy silently assumes that the camera and the robot base stay exactly where they were during data collection: a bumped tripod or a base re-docked a few centimetres away makes it reach for the wrong place. We show that both failures are one phenomenon, a change of the camera-to-base extrinsic that the policy has memorised, and that the robot carries everything needed to undo it. We introduce WACO, a world-anchored canonicalization framework for VLA models: the arm, placed by forward kinematics at the measured joint angles, serves as a metric calibration target, and the static scene serves as the world anchor that tells a moved camera from a moved robot. From a single RGB-D frame, WACO estimates both shifts without ground-truth extrinsics or extra data, canonicalises the image to the calibration viewpoint and the state, action and start pose to the calibration base frame, and fine-tunes the unmodified policy under the artefacts of its own estimator. We also introduce LIBERO-PoseShift, the first benchmark for RGB VLAs with graded camera, base and compound shifts. Across LIBERO-PoseShift, the LIBERO-Plus camera track and four real-world tasks on a UR5e robot, WACO consistently outperforms the standard few-shot recipe and prior viewpoint-robust methods, raising shifted success by 13.2 points in simulation and by 36.7 points on the real robot, recovering 90% of the simulation gap to a ground-truth oracle and 87% of the real-robot gap, and establishing a new state of the art among methods that use no perturbed training data, with no test-time calibration. Project page: https://peacelwh.github.io/WACO/.

## RoboDx: Repair and Diagnosis of Faults in Robot Software

DevFeed: [RoboDx: Repair and Diagnosis of Faults in Robot Software](<https://devfeed.tech/articles/robodx-repair-and-diagnosis-of-faults-in-robot-software-142197.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23816>)

Author: Soham Dan

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Software Engineering](<https://devfeed.tech/sources/arxiv-software-engineering.md>)

Topics: [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>), [patches](<https://devfeed.tech/topics/patches.md>), [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>), [Test-driven development](<https://devfeed.tech/topics/tdd.md>)

Tags: [bug-fix](<https://devfeed.tech/tags/bug-fix.md>), [claude-opus-5-5](<https://devfeed.tech/tags/claude-opus-5-5.md>), [configuration](<https://devfeed.tech/tags/configuration.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [cs-se](<https://devfeed.tech/tags/cs-se.md>), [diagnosis](<https://devfeed.tech/tags/diagnosis.md>), [documentation](<https://devfeed.tech/tags/documentation.md>), [navigation](<https://devfeed.tech/tags/navigation.md>), [opus-5-5](<https://devfeed.tech/tags/opus-5-5.md>), [repair](<https://devfeed.tech/tags/repair.md>), [robot](<https://devfeed.tech/tags/robot.md>), [simulator](<https://devfeed.tech/tags/simulator.md>), [software](<https://devfeed.tech/tags/software.md>), [test](<https://devfeed.tech/tags/test.md>)

### AI overview

RoboDx evaluates whether agents can repair and diagnose faults in simulated robot navigation software. Its hidden verifier checks driving in unseen scenarios, but a verified repair does not guarantee every fault was removed: on 50 two-fault tasks, only 17 of Claude Opus 5.5's 49 verified repairs removed both injected edits, often leaving undocumented configuration constants. Without code, Opus identified both faults on 10 of those 50 tasks. In diagnosis, four models named the cause on at most 38 of 94 tasks using robot measurements; access to simulator outputs for each hypothesis yielded near-universal correctness, while models' own output predictions showed no detectable gain. Results on 97 real bug-fix pull requests did not order the three stronger models, and Haiku 5.5 scored lowest.

### Source excerpt

arXiv:2611.23816v1 Announce Type: cross Abstract: A patch to a robot's navigation software can pass every unit test and still drive the robot off its path. Checking how the repaired robot drives catches such patches. It does not show that the agent found everything that was wrong. We introduce RoboDx, a generator of repair tasks for simulated robot navigation software with known injected faults. A hidden verifier marks a repair as verified only if the robot then drives correctly in hidden scenarios, which the agent never sees. A second track hides the code and asks the agent to pick the cause from a list of hypotheses. A task with two faults can contain a quiet fault, a fault that makes no hidden scenario fail unless the other fault is also present. On our set of 100 tasks, RoboDx-Hard-100, almost every repair by Claude Opus 5.5 with standard coding tools is verified. We analyse 94 of the tasks, 44 with one fault and 50 with two. On the 50 two-fault tasks, only 17 of Opus's 49 verified repairs remove both injected edits. Most faults left behind are configuration constants whose correct value the task's documentation does not state. Opus names both faults on 10/50 of these tasks when it cannot see the code. We then give four current models the same measurements of each faulty robot, picked so that, in the simulator, only the true cause produces all of them. The models name the cause on at most 38 of 94 tasks. They are right on almost all when they also get the simulator's output for every hypothesis. Their own predictions of that output give no detectable gain. On 97 tasks from real bug-fix pull requests, Claude Haiku 5.5, the weakest model on the generated tasks, scores lowest, and the other three models are not ordered.

## AERR-Nav: Adaptive Exploration-Recovery-Reminiscing Strategy for Zero-Shot Object Navigation

DevFeed: [AERR-Nav: Adaptive Exploration-Recovery-Reminiscing Strategy for Zero-Shot Object Navigation](<https://devfeed.tech/articles/aerr-nav-adaptive-exploration-recovery-reminiscing-strategy-for-zero-shot-object-navigation-143243.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2603.17712>)

Author: Jingzhi Huang, Junkai Huang, Haoyang Yang, Haoang Li, Yi Wang

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>)

Tags: [adaptive](<https://devfeed.tech/tags/adaptive.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [exploration](<https://devfeed.tech/tags/exploration.md>), [navigation](<https://devfeed.tech/tags/navigation.md>), [object](<https://devfeed.tech/tags/object.md>), [recovery](<https://devfeed.tech/tags/recovery.md>), [strategy](<https://devfeed.tech/tags/strategy.md>), [zero-shot](<https://devfeed.tech/tags/zero-shot.md>)

### AI overview

AERR-Nav addresses zero-shot object navigation in unfamiliar multi-floor environments by adapting its behavior to the robot's surroundings. It transitions among exploration, recovery, and reminiscing states, and uses fast- and slow-thinking modes during exploration to balance exploration, exploitation, and higher-level reasoning. The abstract reports state-of-the-art performance among zero-shot methods on HM3D and MP3D, with ablation studies supporting the proposed strategy and modules; it gives no quantitative results or implementation details.

### Source excerpt

arXiv:2603.17712v2 Announce Type: replace-cross Abstract: Zero-Shot Object Navigation (ZSON) in unknown multi-floor environments presents a significant challenge. Recent methods, mostly based on semantic value greedy waypoint selection, spatial topology-enhanced memory, and Multimodal Large Language Model (MLLM) as a decision-making framework, have led to improvements. However, these architectures struggle to balance exploration and exploitation for ZSON when encountering unseen environments, especially in multi-floor settings, such as robots getting stuck at narrow intersections, endlessly wandering, or failing to find stair entrances. To overcome these challenges, we propose AERR-Nav, a Zero-Shot Object Navigation framework that dynamically adjusts its state based on the robot's environment. Specifically, AERR-Nav has the following two key advantages: (1) An Adaptive Exploration-Recovery-Reminiscing Strategy, enables robots to dynamically transition between three states, facilitating specialized responses to diverse navigation scenarios. (2) An Adaptive Exploration State featuring Fast and Slow-Thinking modes helps robots better balance exploration, exploitation, and higher-level reasoning based on evolving environmental information. Extensive experiments on the HM3D and MP3D benchmarks demonstrate that our AERR-Nav achieves state-of-the-art performance among zero-shot methods. Comprehensive ablation studies further validate the efficacy of our proposed strategy and modules.

## Spatiotemporal and Multi-Robot Motion Planning on Graphs of Space-Time Convex Sets

DevFeed: [Spatiotemporal and Multi-Robot Motion Planning on Graphs of Space-Time Convex Sets](<https://devfeed.tech/articles/spatiotemporal-and-multi-robot-motion-planning-on-graphs-of-space-time-convex-sets-139786.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2607.00444>)

Author: Jingtao Tang, Zining Mao, Lufan Yang, Hang Ma

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>), [Combinatorial optimization](<https://devfeed.tech/topics/combinatorial-optimization.md>), [graph theory](<https://devfeed.tech/topics/graph-theory.md>)

Tags: [continuous](<https://devfeed.tech/tags/continuous.md>), [convex](<https://devfeed.tech/tags/convex.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [graphs](<https://devfeed.tech/tags/graphs.md>), [motion](<https://devfeed.tech/tags/motion.md>), [motion-planning](<https://devfeed.tech/tags/motion-planning.md>), [planning](<https://devfeed.tech/tags/planning.md>), [pruning](<https://devfeed.tech/tags/pruning.md>), [quality](<https://devfeed.tech/tags/quality.md>), [robot](<https://devfeed.tech/tags/robot.md>), [robots](<https://devfeed.tech/tags/robots.md>), [route](<https://devfeed.tech/tags/route.md>), [scalability](<https://devfeed.tech/tags/scalability.md>), [search](<https://devfeed.tech/tags/search.md>), [simulations](<https://devfeed.tech/tags/simulations.md>), [space](<https://devfeed.tech/tags/space.md>), [temporal](<https://devfeed.tech/tags/temporal.md>), [time](<https://devfeed.tech/tags/time.md>), [timing](<https://devfeed.tech/tags/timing.md>), [velocity](<https://devfeed.tech/tags/velocity.md>)

### AI overview

The framework plans continuous robot motion in space and time by representing collision-free regions as convex sets, jointly optimizing the route and piecewise-linear trajectory timing under velocity limits. Exact Convex Decomposition removes swept space occupied by moving obstacles and previously planned robots while preserving a convex decomposition for later queries. A best-first solver uses temporal reachability for dominance pruning and admissible heuristics; heuristic bounds can be reused after reservation updates. The authors combine these methods with prioritized planning and dynamically adjusted coordination windows to limit graph growth. Experiments report improved scalability and competitive solution quality against optimization-based and sampling-based planners; simulations coordinate up to 100 robots within a minute, with real-robot demonstrations also reported.

### Source excerpt

arXiv:2607.00444v2 Announce Type: replace-cross Abstract: Decoupled multi-robot motion planning avoids the high-dimensional joint configuration space but requires repeated single-robot planning among moving obstacles, where route choice and timing are coupled. We develop a framework based on graphs of space-time convex sets (ST-GCSs) for continuous spatiotemporal planning and multi-robot coordination. An ST-GCS represents collision-free space-time by convex sets, allowing the sequence of regions and the geometry and timing of piecewise-linear trajectories to be optimized jointly under velocity limits. To incorporate moving obstacles and previously planned robots, we develop Exact Convex Decomposition (ECD), which removes their swept occupancies while maintaining a convex decomposition for subsequent planning queries. Building on path-based GCS search, we develop a best-first solver tailored to the ST-GCS motion model, exploiting temporal reachability for dominance pruning and using admissible heuristics with separate or joint search-key evaluation. Precomputed heuristic bounds can also be reused across reservation-updated graphs. We integrate ST-GCS planning and ECD into prioritized multi-robot planning and introduce windowed coordination with dynamic window adjustment to limit reservation-induced graph growth. Experiments demonstrate improved scalability and competitive solution quality against optimization-based and sampling-based planners on challenging spatiotemporal and multi-robot problems. Large-scale simulations coordinate up to $100$ robots within a minute, while real-robot demonstrations showcase the practical applicability of the framework. Project homepage: https://sites.google.com/view/stgcs.

## Coordination of Heterogeneous Teams with Uncertain Task Requirements

DevFeed: [Coordination of Heterogeneous Teams with Uncertain Task Requirements](<https://devfeed.tech/articles/coordination-of-heterogeneous-teams-with-uncertain-task-requirements-139915.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.21828>)

Author: Jongann Lee, Melkior Ornik

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>), [Neural Network](<https://devfeed.tech/topics/neural-network.md>)

Tags: [actions](<https://devfeed.tech/tags/actions.md>), [agents](<https://devfeed.tech/tags/agents.md>), [algorithm](<https://devfeed.tech/tags/algorithm.md>), [capabilities](<https://devfeed.tech/tags/capabilities.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-ma](<https://devfeed.tech/tags/cs-ma.md>), [discovery](<https://devfeed.tech/tags/discovery.md>), [engagement](<https://devfeed.tech/tags/engagement.md>), [execution](<https://devfeed.tech/tags/execution.md>), [graph](<https://devfeed.tech/tags/graph.md>), [location](<https://devfeed.tech/tags/location.md>), [map](<https://devfeed.tech/tags/map.md>), [multi-agent](<https://devfeed.tech/tags/multi-agent.md>), [network](<https://devfeed.tech/tags/network.md>), [plan](<https://devfeed.tech/tags/plan.md>), [planner](<https://devfeed.tech/tags/planner.md>), [planning](<https://devfeed.tech/tags/planning.md>), [requirements](<https://devfeed.tech/tags/requirements.md>), [shortest-path](<https://devfeed.tech/tags/shortest-path.md>), [staging](<https://devfeed.tech/tags/staging.md>), [tasks](<https://devfeed.tech/tags/tasks.md>), [teams](<https://devfeed.tech/tags/teams.md>)

### AI overview

The paper presents a hierarchical planner for heterogeneous teams facing tasks whose requirements are initially unknown. A graph neural network policy assigns actions to agents using a CAAT graph of agents, tasks, capabilities, and actions; a shortest-path algorithm then routes each agent to its assigned action. The authors report that jointly coordinating requirement discovery and task completion is faster than baselines that optimize discovery and engagement independently. They also report zero-shot transfer from an outdoor training map to an indoor setting with more agents, tasks, and capabilities, without fine-tuning. These findings are reported in the abstract; it gives no quantitative results or further experimental details.

### Source excerpt

arXiv:2611.21828v1 Announce Type: cross Abstract: Coordinating a team of heterogeneous agents to complete tasks with unknown requirements is a challenging multi-agent planning problem. Agents differ in their ability to discover task requirements and in their ability to engage tasks, so the planner must coordinate task discovery, execution, and staging. This paper introduces a hierarchical planner that uses a graph neural network policy to assign actions to agents, then finds a path to the action's location using a shortest-path algorithm. The central part of our approach is the CAAT graph, which is a graph representation of the planning context consisting of the agents, tasks, capabilities, and actions. We demonstrate that our learned planner can jointly coordinate a heterogeneous team to simultaneously discover task requirements and complete tasks in a shorter time compared to baselines that optimize task discovery and task engagement independently. Furthermore, we show that our policy can zero-shot plan on out-of-distribution settings by training only on an outdoor map then transferring without finetuning to an indoor map with more agents, tasks, and capabilities than seen during training.

## Explore, Map, Remember, Decide: A Benchmark for VLM Readiness in Safety-Critical Scenarios

DevFeed: [Explore, Map, Remember, Decide: A Benchmark for VLM Readiness in Safety-Critical Scenarios](<https://devfeed.tech/articles/explore-map-remember-decide-a-benchmark-for-vlm-readiness-in-safety-critical-scenarios-139800.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2608.08077>)

Author: Gabriele La Malfa, Nitay Alon, Emanuele La Malfa, Reuth Mirsky, Stefan Sarkadi

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>), [test](<https://devfeed.tech/topics/test.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [choice](<https://devfeed.tech/tags/choice.md>), [controllers](<https://devfeed.tech/tags/controllers.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ma](<https://devfeed.tech/tags/cs-ma.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [environment](<https://devfeed.tech/tags/environment.md>), [error](<https://devfeed.tech/tags/error.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [explore](<https://devfeed.tech/tags/explore.md>), [external](<https://devfeed.tech/tags/external.md>), [framework](<https://devfeed.tech/tags/framework.md>), [gathering](<https://devfeed.tech/tags/gathering.md>), [horizon](<https://devfeed.tech/tags/horizon.md>), [limits](<https://devfeed.tech/tags/limits.md>), [map](<https://devfeed.tech/tags/map.md>), [maps](<https://devfeed.tech/tags/maps.md>), [memory](<https://devfeed.tech/tags/memory.md>), [metrics](<https://devfeed.tech/tags/metrics.md>), [models](<https://devfeed.tech/tags/models.md>), [safety](<https://devfeed.tech/tags/safety.md>), [vlm](<https://devfeed.tech/tags/vlm.md>)

### AI overview

EMRD evaluates VLM spatial reasoning in four stages--Explore, Map, Remember, and Decide--using 18 metrics across 16 buildings and three environmental stresses. Open-weight models left rooms unvisited in 95% of episodes and detected at most 51% of objects; closed-weight models explored more, though two of three using a notepad stopped prematurely. Without external memory, most models retained fewer observed objects, especially in larger buildings. Starting location constrained exploration, and open-weight models starting in different rooms agreed on the final choice no more often than chance. Room-label swaps often changed that choice, indicating sensitivity to semantic priors. Motion noise raised object-position error by 12-60% for closed-weight models; strong darkness cut map recall by 28-31 percentage points, and dense smoke reduced objects seen by 25-39 points. Among rooms entered, environmental stress did not significantly change the final choice.

### Source excerpt

arXiv:2608.08077v2 Announce Type: replace Abstract: Vision-Language Models (VLMs) are often deployed as embodied controllers for robotics, yet it remains unclear how well they can explore an environment, maintain an internal map, and ground safety-critical decisions in long-horizon sequential trajectories. In this work, we introduce EMRD, an evaluation framework that decouples spatial reasoning into four stages: Explore (what the agent observes), Map (what it records and where), Remember (what it retains), and Decide (whether its final choice is justified), scored across 18 stage-specific metrics. We use EMRD to evaluate eleven open- and closed-weight VLMs across 16 buildings under nominal conditions and three environmental stresses (darkness, smoke, and motion noise), where an embodied VLM explores unknown rooms, maps observed objects, and selects the safest gathering place. Open-weight models leave rooms unvisited in 95% of episodes and detect at most 51% of objects; closed-weight models explore more, though with an external memory (notepad), two of the three stop prematurely. Without external memory, most VLMs keep fewer of the objects they saw, especially in larger buildings. The start limits which rooms are explored, and open-weight models started in different rooms agree on the final room not significantly more often than chance. Moreover, these choices are biased by semantic priors: swapping a room's label is often enough to change the final choice. Finally, motion noise increases object position error (by 12-60% for closed-weight models), strong darkness lowers map recall by 28-31 percentage points, and dense smoke reduces seen objects by 25-39 points, while the choice among the rooms entered does not change significantly.

## WoRD: Diffusion-Based Posterior Inference for Online Goal Recognition

DevFeed: [WoRD: Diffusion-Based Posterior Inference for Online Goal Recognition](<https://devfeed.tech/articles/word-diffusion-based-posterior-inference-for-online-goal-recognition-140129.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23192>)

Author: Almog Anschel, Sarah Keren

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>), [Calibrated uncertainty](<https://devfeed.tech/topics/calibrated-uncertainty.md>), [diffusion-transformers](<https://devfeed.tech/topics/diffusion-transformers.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [bayes](<https://devfeed.tech/tags/bayes.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [denoising](<https://devfeed.tech/tags/denoising.md>), [diffusion](<https://devfeed.tech/tags/diffusion.md>), [inference](<https://devfeed.tech/tags/inference.md>), [motion](<https://devfeed.tech/tags/motion.md>), [multimodal](<https://devfeed.tech/tags/multimodal.md>), [online](<https://devfeed.tech/tags/online.md>), [recognition](<https://devfeed.tech/tags/recognition.md>), [robot](<https://devfeed.tech/tags/robot.md>), [vehicle](<https://devfeed.tech/tags/vehicle.md>)

### AI overview

WoRD infers an agent's goal from partial continuous trajectories by using a learned trajectory prior rather than requiring an explicit dynamics model. For each candidate goal, it runs a warm-started denoising chain, steers samples toward observations with a closed-form log-likelihood gradient, then applies Bayes' rule to form a calibrated goal posterior. In evaluations on simulated robots, pedestrians, and vehicles, the method reportedly preserved multiple plausible goals through ambiguous prefixes, while filter- and classifier-based observers sometimes collapsed to one mode or became overconfident. The supplied text is an abstract and does not provide metric names or detailed results.

### Source excerpt

arXiv:2611.23192v1 Announce Type: new Abstract: Goal recognition is the task of inferring an agent's intended goal from its observed behavior. Most existing work focuses on high-level symbolic recognition, abstracting away the continuous motion through which agents reach their goal. Methods that do reason over continuous trajectories rely on an explicit dynamics model that is hard to acquire and rarely transfers across settings. Neither approach explicitly learns how agents move toward each candidate goal, limiting accuracy on ambiguous trajectory prefixes where such knowledge is most needed. We present WoRD (Waypoint Recognition with Diffusion), which uses a learned trajectory prior to infer the agent's goal from partial observations. At each step, WoRD runs one warm-started denoising chain per candidate goal, steers each chain toward the observed trajectory through a closed-form log-likelihood gradient, and reweights the resulting samples by Bayes' rule into a calibrated posterior over goals. We evaluate WoRD on simulated robot settings, pedestrian trajectories, and vehicle trajectories using four metrics targeting distinct failure modes. Across all three settings, the same inference procedure preserves calibrated multimodal beliefs through ambiguous prefixes where filter- and classifier-based observers collapse to a single mode or grow overconfident.

## Better Calibrated, Better Planned: Uncertainty-Aware Multi-ASV Informative Path Planning for Oil Spill Monitoring

DevFeed: [Better Calibrated, Better Planned: Uncertainty-Aware Multi-ASV Informative Path Planning for Oil Spill Monitoring](<https://devfeed.tech/articles/better-calibrated-better-planned-uncertainty-aware-multi-asv-informative-path-planning-for-oil-spill-monitoring-141675.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.21783>)

Author: Alejandro Casado-P\\'erez, Marija Popovi\\'c, Daniel Guti\\'errez-Reina, Sergio L. Toral., Samuel Yanes

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Calibrated uncertainty](<https://devfeed.tech/topics/calibrated-uncertainty.md>), [Neural Network](<https://devfeed.tech/topics/neural-network.md>), [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>)

Tags: [autonomous](<https://devfeed.tech/tags/autonomous.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [cells](<https://devfeed.tech/tags/cells.md>), [coupling](<https://devfeed.tech/tags/coupling.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-ma](<https://devfeed.tech/tags/cs-ma.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [data](<https://devfeed.tech/tags/data.md>), [deep](<https://devfeed.tech/tags/deep.md>), [deep-learning](<https://devfeed.tech/tags/deep-learning.md>), [energy](<https://devfeed.tech/tags/energy.md>), [error](<https://devfeed.tech/tags/error.md>), [field](<https://devfeed.tech/tags/field.md>), [fleet](<https://devfeed.tech/tags/fleet.md>), [gaussian](<https://devfeed.tech/tags/gaussian.md>), [information](<https://devfeed.tech/tags/information.md>), [ipp](<https://devfeed.tech/tags/ipp.md>), [learning](<https://devfeed.tech/tags/learning.md>), [monitoring](<https://devfeed.tech/tags/monitoring.md>), [oil](<https://devfeed.tech/tags/oil.md>), [path-planning](<https://devfeed.tech/tags/path-planning.md>), [planning](<https://devfeed.tech/tags/planning.md>), [uncertainty](<https://devfeed.tech/tags/uncertainty.md>)

### AI overview

In physics-based oil spill simulations across 60 maps, the uncertainty model mattered far more than planner choice among deep learning approaches, explaining 80% versus 1% of the variation in final reconstruction error. Evidential Deep Learning (EDL) had the lowest error and highest IoU under all five planners, was best calibrated, and ran in 3.9 ms per step. Using 50 cells per vehicle, EDL outperformed Monte Carlo Dropout and Deep Ensembles using 150. Ensembles grew more confident as their errors increased, while EDL was unaffected. The results suggest that uncertainty-aware multi-ASV planning requires an information model robust to the data generated by its own planner.

### Source excerpt

arXiv:2611.21783v1 Announce Type: cross Abstract: Monitoring oil spills with fleets of Autonomous Surface Vehicles (ASVs) requires mapping a heterogeneous pollutant field from sparse, noisy measurements under tight time and energy budgets. Informative Path Planning (IPP) addresses this problem by coupling a reconstruction model, which infers the field and its uncertainty, with a planner that decides where to sample next. Prior work has focused on planning strategies and has largely relied on Gaussian Processes (GPs), leaving the calibration of the model's uncertainty underexplored. We study how uncertainty calibration conditions multi-vehicle IPP with a fleet of ASVs that coordinates online through a shared uncertainty map. We introduce Evidential Deep Learning (EDL) as an uncertainty-calibrated information model. We benchmark EDL, Monte Carlo Dropout and Deep Ensembles against a GP in closed loop with five planners on physics-based oil spill simulations. Across 60 maps, the choice of model dominates the outcome, explaining 80\% of the variance of the final reconstruction error, against 1\% for the planner among the deep models. EDL achieves the lowest error and the highest IoU under every planner, is the best calibrated model and runs at 3.9 ms per step. With 50 cells per vehicle it reconstructs the spill better than MC Dropout and Deep Ensembles with 150. Deep Ensembles become more confident as their error grows, whereas EDL is unaffected. Robustness of the information model to the data its own planner collects is therefore a requirement of uncertainty-aware IPP.

## Skypart : Weather-Robust Cross-View Geo-Localization via Prototype-Based Semantic Part Discovery

DevFeed: [Skypart : Weather-Robust Cross-View Geo-Localization via Prototype-Based Semantic Part Discovery](<https://devfeed.tech/articles/skypart-weather-robust-cross-view-geo-localization-via-prototype-based-semantic-part-discovery-139768.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2605.11654>)

Author: Chi-Nguyen Tran, Dao Sy Duy Minh, Huynh Trung Kiet, Nguyen Lam Phu Quy, Phu-Hoa Pham, Long Tran-Thanh

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Neural Network](<https://devfeed.tech/topics/neural-network.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Embeddings](<https://devfeed.tech/topics/embeddings.md>), [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>)

Tags: [attention](<https://devfeed.tech/tags/attention.md>), [cross](<https://devfeed.tech/tags/cross.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [drone](<https://devfeed.tech/tags/drone.md>), [geo](<https://devfeed.tech/tags/geo.md>), [gnss](<https://devfeed.tech/tags/gnss.md>), [lightweight](<https://devfeed.tech/tags/lightweight.md>), [linear](<https://devfeed.tech/tags/linear.md>), [localization](<https://devfeed.tech/tags/localization.md>), [prototype](<https://devfeed.tech/tags/prototype.md>), [retrieval](<https://devfeed.tech/tags/retrieval.md>), [robust](<https://devfeed.tech/tags/robust.md>), [satellite](<https://devfeed.tech/tags/satellite.md>), [semantic](<https://devfeed.tech/tags/semantic.md>), [training](<https://devfeed.tech/tags/training.md>), [transformers](<https://devfeed.tech/tags/transformers.md>), [uncertainty](<https://devfeed.tech/tags/uncertainty.md>), [weather](<https://devfeed.tech/tags/weather.md>)

### AI overview

SKYPART improves cross-view drone-to-satellite localization by grouping image patches into semantic parts with competing learnable prototypes, then using graph attention to read out active parts. During training, altitude-conditioned modulation helps account for scale changes; the resulting retrieval embeddings omit altitude at inference. Kendall uncertainty weighting learns the relative weights of multiple training objectives. The authors report 26.95M parameters and 22.14 GFLOPs, state-of-the-art results on SUES-200, University-1652, and DenseUAV with single-pass retrieval and no reranking or test-time augmentation, and a larger accuracy advantage over the strongest baseline under ten WeatherPrompt corruption conditions. These are abstract-level claims; the supplied text gives no numerical accuracy results or implementation details.

### Source excerpt

arXiv:2605.11654v3 Announce Type: replace-cross Abstract: The low-altitude economy is growing rapidly, increasing demand for reliable drone navigation. Common tasks such as delivery, inspection, and search-and-rescue rely on Global Navigation Satellite Systems (GNSS), which can be jammed, spoofed, or unavailable. Cross-view geo-localization (CVGL), which matches an oblique drone view to a geo-referenced satellite tile, offers an alternative for autonomous localization when GNSS fails. However, current CVGL methods have three limitations: (1) global descriptors compress patch grids into a single vector, without separating cross-view-shared layout from view-specific texture; (2) altitude-related scale variation remains in the learned embedding instead of being marginalized as a nuisance; and (3) multi-objective training uses hand-tuned loss weights despite incompatible gradient scales. We propose SKYPART, a lightweight, swappable head for patch-based vision transformers (ViTs) that explicitly groups patches into parts. It combines (i) learnable prototypes that compete for patch tokens through single-pass cosine assignment; (ii) altitude-conditioned linear modulation used only during training, yielding altitude-free retrieval embeddings at inference; (iii) graph-attention readout over active prototypes; and (iv) Kendall uncertainty-weighted multi-objective learning that learns loss-group weights end-to-end. With 26.95M parameters and 22.14 GFLOPs, SKYPART is the smallest top-performing method in our comparison and achieves state-of-the-art results on SUES-200, University-1652, and DenseUAV using single-pass retrieval without reranking or test-time augmentation (TTA). Its accuracy advantage over the strongest baseline further increases under ten WeatherPrompt corruption conditions. Code: https://github.com/chisngyen/SkyPart.

## Escaping Latent Traps by Remembering: Training-Free Episodic Kernel Repulsion for World-Model Planning

DevFeed: [Escaping Latent Traps by Remembering: Training-Free Episodic Kernel Repulsion for World-Model Planning](<https://devfeed.tech/articles/escaping-latent-traps-by-remembering-training-free-episodic-kernel-repulsion-for-world-model-planning-141780.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.22470>)

Author: Yican Zhao, Bolin Yang, Ruizhe Wang, Haiyang Guo, Fuji Yang, Bingo Wing-Kuen Ling, Zelin Zang

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [World models](<https://devfeed.tech/topics/world-models.md>), [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [gaussian](<https://devfeed.tech/tags/gaussian.md>), [kernel](<https://devfeed.tech/tags/kernel.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [memory](<https://devfeed.tech/tags/memory.md>), [navigation](<https://devfeed.tech/tags/navigation.md>), [planning](<https://devfeed.tech/tags/planning.md>), [training](<https://devfeed.tech/tags/training.md>), [world-model](<https://devfeed.tech/tags/world-model.md>)

### AI overview

Episodic Kernel Repulsion (EKR) improves goal-directed planning in latent world models by adding a training-free cost that repels candidate plans from a novelty-gated memory of visited latents. The authors report average success rising from 31.8% to 46.7% for LeWM and 54.7% to 64.0% for H-JEPA across four benchmarks, with the same model, planner, and budget. Counterfactuals indicate that plans move away from visited states in most settings; stuck episodes and revisits decrease. In seven of eight settings, EKR matches or exceeds using four times as many samples at 0.8-1.5 times the compute. The abstract also reports untuned transfer to DINO-WM with CEM, including gains on Wall, PointMaze, and Push-T tasks.

### Source excerpt

arXiv:2611.22470v1 Announce Type: new Abstract: Latent world models plan towards a goal image by minimising the distance between predicted and goal latents. This distance is not geodesic: it has local minima, latent traps, behind walls, at corners and in dead ends, and memoryless receding-horizon replanning keeps the agent in them, revisiting states far from the goal. We propose Episodic Kernel Repulsion (EKR), a training-free cost term that penalises candidate plans by a top-$k$ Gaussian kernel to a novelty-gated memory of the latents the agent has actually visited, rescaled by the current base cost so that its strength is comparable across models. Across two JEPA world models and four benchmarks, EKR raises average success from 31.8% to 46.7% (LeWM) and from 54.7% to 64.0% (H-JEPA) with the same model, planner and budget. Same-memory counterfactuals show that, in most settings, EKR pushes plans away from visited states, in navigation, often at a higher base cost; stuck episodes and revisits drop accordingly. In seven of eight settings it matches or exceeds $4\times$ samples at $0.8$--$1.5\times$ compute. Without tuning, it transfers to DINO-WM with CEM: $+6.7$ and $+19.3$ points on official and longer-goal Wall tasks, and from 44.7% to 100.0% and from 21.0% to 32.0% on long-horizon PointMaze and Push-T.

## Latent Goal Prediction from Language for Model-Based Planning

DevFeed: [Latent Goal Prediction from Language for Model-Based Planning](<https://devfeed.tech/articles/latent-goal-prediction-from-language-for-model-based-planning-139783.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2606.20627>)

Author: Samuel Barbeau, Simon Roy, Giovanni Beltrame, Christian Desrosiers, Nicolas Thome

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [World models](<https://devfeed.tech/topics/world-models.md>), [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>)

Tags: [alignment](<https://devfeed.tech/tags/alignment.md>), [behavior-cloning](<https://devfeed.tech/tags/behavior-cloning.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [grounding](<https://devfeed.tech/tags/grounding.md>), [horizon](<https://devfeed.tech/tags/horizon.md>), [image](<https://devfeed.tech/tags/image.md>), [instructions](<https://devfeed.tech/tags/instructions.md>), [intermediate](<https://devfeed.tech/tags/intermediate.md>), [language](<https://devfeed.tech/tags/language.md>), [manipulation](<https://devfeed.tech/tags/manipulation.md>), [model](<https://devfeed.tech/tags/model.md>), [navigation](<https://devfeed.tech/tags/navigation.md>), [objectives](<https://devfeed.tech/tags/objectives.md>), [plan](<https://devfeed.tech/tags/plan.md>), [planning](<https://devfeed.tech/tags/planning.md>), [prediction](<https://devfeed.tech/tags/prediction.md>), [reward](<https://devfeed.tech/tags/reward.md>), [sequences](<https://devfeed.tech/tags/sequences.md>), [space](<https://devfeed.tech/tags/space.md>), [world](<https://devfeed.tech/tags/world.md>)

### AI overview

LAGO uses one predictor to forecast action-conditioned dynamics and translate language instructions into intermediate goals in the same latent space. At each planning step, it predicts subgoals and optimizes actions with a soft-minimum alignment cost that rewards proximity without requiring a fixed path; it predicts new subgoals as the agent moves. The abstract reports better performance than flat and hierarchical image-goal planners across navigation and manipulation, with the largest gains on long-horizon navigation through curved, hazard-constrained paths. There, LAGO more than doubles the success rate of a goal-conditioned behavior-cloning policy trained on the same demonstrations. It also outperforms vision-language reward models, including one using ground-truth dynamics.

### Source excerpt

arXiv:2606.20627v4 Announce Type: replace Abstract: Joint-Embedding Predictive Architectures (JEPAs) enable agents to plan in latent space by imagining the outcomes of candidate actions, yet task specification remains a bottleneck. Visual targets provide precise local gradients but poor distant guidance, while language is flexible yet limited by noisy cross-modal alignment or dependence on distinct large generative models. We introduce LAGO (Latent Goal Prediction from Language), a hierarchical world model in which a single predictor both forecasts action-conditioned dynamics and grounds language instructions as sequences of intermediate latent subgoals, training both modes with a single regression objective over a shared latent space. At each planning step, LAGO predicts a sequence of latent subgoals from a language instruction and optimizes an action sequence using a soft-minimum alignment cost that rewards subgoal proximity without enforcing a rigid path. Subgoals are repredicted as the agent's state evolves, turning a single long-horizon instruction into a sequence of locally tractable objectives. Across environments spanning navigation and manipulation, LAGO outperforms flat and hierarchical image-goal planners, with the largest gains on long-horizon navigation with curved, hazard-constrained paths, where it more than doubles the success rate of a goal-conditioned behavior cloning policy trained on the same demonstrations. Grounding language in the world model's own latent space also outperforms vision-language reward models, including one that plans with ground-truth dynamics.

## AgenticJevNav: Agentic Object Search and Navigation via Event-Driven Verification

DevFeed: [AgenticJevNav: Agentic Object Search and Navigation via Event-Driven Verification](<https://devfeed.tech/articles/agenticjevnav-agentic-object-search-and-navigation-via-event-driven-verification-137512.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.19912>)

Author: Shuning Zhang, Shuli Zhang, Yunheng Wang, Wei Lu, Jiaxi Zhang, Yihang Kang, Tao Wang, Xiangyu Chen, Yixiao Feng, Kai Sheng, Renjing Xu

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>), [AI Agent](<https://devfeed.tech/topics/ai-agent.md>), [Multi-step AI processes](<https://devfeed.tech/topics/multi-step-ai-processes.md>), [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>)

Tags: [abstract](<https://devfeed.tech/tags/abstract.md>), [actions](<https://devfeed.tech/tags/actions.md>), [after](<https://devfeed.tech/tags/after.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [aggregate](<https://devfeed.tech/tags/aggregate.md>), [application](<https://devfeed.tech/tags/application.md>), [approach](<https://devfeed.tech/tags/approach.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [best](<https://devfeed.tech/tags/best.md>), [bounded](<https://devfeed.tech/tags/bounded.md>), [control](<https://devfeed.tech/tags/control.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [decision](<https://devfeed.tech/tags/decision.md>), [event](<https://devfeed.tech/tags/event.md>), [event-driven](<https://devfeed.tech/tags/event-driven.md>), [navigation](<https://devfeed.tech/tags/navigation.md>), [object](<https://devfeed.tech/tags/object.md>), [search](<https://devfeed.tech/tags/search.md>), [verification](<https://devfeed.tech/tags/verification.md>)

### AI overview

AgenticJevNav combines goal-directed planning, evidence memory, event-triggered verification, and bounded execution to handle incomplete observations and action failures. The Agent plans from RGB, odometry, memory, and feedback; Jev reviews evidence after blocked or failed execution or two consecutive low-progress segments. Version, freshness, and once-only checks govern applying decisions, while a single executor and independent RGB reflex constrain motion. Goal revisions revoke obsolete requests and actions before replanning. In six simulated task families, it achieved 73.33% success over 60 trials, 21.67 percentage points above the next-best Agent-only aggregate control; the comparison used differing historical protocols. A GENISOM L1 quadruped achieved 70.0% over 30 real-world trials, including bounded search reporting. The authors identify grounding, approach planning, and runtime as areas for improvement.

### Source excerpt

arXiv:2611.19912v1 Announce Type: cross Abstract: Agentic navigation must maintain goals and adapt actions from incomplete evidence despite execution failures and delayed inference. We present AgenticJevNav, which links goal-directed planning, evidence memory, event-triggered verification, and bounded execution. The Agent plans from RGB, odometry, memory, and feedback, retaining goal ownership while authorizing a bounded action set. Jev reviews structured evidence after blocked/failed execution or two consecutive low-progress segments. Version, freshness, and once-only checks govern decision application. A single executor and independent RGB reflex constrain motion. Goal revisions revoke obsolete requests and actions before replanning. Across six simulated task families, AgenticJevNav achieves 73.33% SR over 60 trials, 21.67 percentage points above the next-best Agent-only aggregate control under differing historical protocols. A real-world GENISOM L1 quadruped deployment achieves 70.0% success over 30 trials, including bounded search reporting. Together, the simulation and physical-robot results demonstrate the effectiveness of the proposed architecture, while remaining failures highlight opportunities to improve grounding, approach planning, and runtime.

## ARC-VLN: Adaptive Recursive Computing-based Visual-Language Navigation

DevFeed: [ARC-VLN: Adaptive Recursive Computing-based Visual-Language Navigation](<https://devfeed.tech/articles/arc-vln-adaptive-recursive-computing-based-visual-language-navigation-137327.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.18696>)

Author: Xiang Zhang, Xiangyu Dong, Yang Lu, Xiaoguang Ma

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>), [Transformers](<https://devfeed.tech/topics/transformers.md>), [navigation](<https://devfeed.tech/topics/navigation.md>)

Tags: [adaptive](<https://devfeed.tech/tags/adaptive.md>), [compaction](<https://devfeed.tech/tags/compaction.md>), [computing](<https://devfeed.tech/tags/computing.md>), [costs](<https://devfeed.tech/tags/costs.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [decision](<https://devfeed.tech/tags/decision.md>), [exit](<https://devfeed.tech/tags/exit.md>), [floating-point](<https://devfeed.tech/tags/floating-point.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [gpu-memory](<https://devfeed.tech/tags/gpu-memory.md>), [inference](<https://devfeed.tech/tags/inference.md>), [inference-costs](<https://devfeed.tech/tags/inference-costs.md>), [learning](<https://devfeed.tech/tags/learning.md>), [lightweight](<https://devfeed.tech/tags/lightweight.md>), [model](<https://devfeed.tech/tags/model.md>), [model-inference](<https://devfeed.tech/tags/model-inference.md>), [navigation](<https://devfeed.tech/tags/navigation.md>), [routing](<https://devfeed.tech/tags/routing.md>), [sequences](<https://devfeed.tech/tags/sequences.md>), [storage](<https://devfeed.tech/tags/storage.md>), [visual-language](<https://devfeed.tech/tags/visual-language.md>)

### AI overview

ARC-VLN reduces aerial visual-language navigation compute by applying shared recursive language-backbone blocks at full depth to text and action tokens while adaptively exiting visual tokens at selected depths. It retains early-exit visual representations and compacts active tokens to shorten later sequences; a learning strategy encourages routing visual tokens across multiple exit depths. On OpenFly Test-Seen and Test-Unseen, the abstract reports the lowest navigation error and highest success rate and success weighted by path length among compared methods, alongside lower average decision steps, total FLOPs, inference time per trajectory, and GPU memory use. The excerpt gives no numerical results or implementation details.

### Source excerpt

arXiv:2611.18696v1 Announce Type: cross Abstract: Efficient aerial visual-language navigation (VLN) requires reducing model storage and inference costs while maintaining instruction-conditioned action prediction. We present ARC-VLN, a token-type-aware recursive computation framework built on a language backbone with shared parameters. ARC-VLN retains full-depth computation for text and action tokens while adaptively selecting exit depths for visual tokens. Representations of early-exiting visual tokens are retained, and active-token compaction shortens the sequences processed by subsequent recursive blocks. To mitigate routing collapse to a single exit depth, we design a dedicated learning strategy that encourages the model to distribute visual tokens across multiple exit depths.On OpenFly's Test-Seen and Test-Unseen subsets, ARC-VLN achieves the lowest navigation error and the highest success rate and success weighted by path length among all compared methods. Meanwhile, it reduces the average number of decision steps per trajectory, total floating-point operations, mean model inference time per trajectory, and average GPU memory usage during inference, indicating its great potential as a VLN backbone for lightweight applications.

## HospitalBench: Diagnosing Where and Why Embodied Agents Fail on Long Hospital Task Chains

DevFeed: [HospitalBench: Diagnosing Where and Why Embodied Agents Fail on Long Hospital Task Chains](<https://devfeed.tech/articles/hospitalbench-diagnosing-where-and-why-embodied-agents-fail-on-long-hospital-task-chains-137424.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.19274>)

Author: Gaytri Jena, Kapil Wanaskar, Aman Chadha, Vasu Sharma, Vinija Jain, Amitava Das

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Isaac Sim](<https://devfeed.tech/topics/isaac-sim.md>), [Agent Harness](<https://devfeed.tech/topics/agent-harness.md>), [Multi-step AI processes](<https://devfeed.tech/topics/multi-step-ai-processes.md>), [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>), [Isaac](<https://devfeed.tech/topics/isaac.md>)

Tags: [abstract](<https://devfeed.tech/tags/abstract.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [call](<https://devfeed.tech/tags/call.md>), [camera](<https://devfeed.tech/tags/camera.md>), [continuous](<https://devfeed.tech/tags/continuous.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [diagnostic](<https://devfeed.tech/tags/diagnostic.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [evaluation-harness](<https://devfeed.tech/tags/evaluation-harness.md>), [fail](<https://devfeed.tech/tags/fail.md>), [failed](<https://devfeed.tech/tags/failed.md>), [first](<https://devfeed.tech/tags/first.md>), [full](<https://devfeed.tech/tags/full.md>), [horizon](<https://devfeed.tech/tags/horizon.md>), [hospital](<https://devfeed.tech/tags/hospital.md>), [isaac](<https://devfeed.tech/tags/isaac.md>), [isaac-sim](<https://devfeed.tech/tags/isaac-sim.md>), [language](<https://devfeed.tech/tags/language.md>)

### AI overview

HospitalBench tests whether vision-language agents can complete causal chains of hospital chores in one continuous NVIDIA Isaac Sim environment, exposing failures that end-task success alone can miss. Each chore has two unscored warm-up steps and five scored sub-tasks; agents act through skill-level tools using camera images. A scripted robot restricted to the same calls checks whether a chore is solvable. The benchmark tracks where chains break, responses to failed calls, and drift between estimated and true position. Preliminary results cover 15 of 60 planned runs across three models and three chores, with scripted-robot proofs for three of ten chores: two runs completed all five sub-tasks. Of 189 failed calls, 168 were followed by a different call and 111 by a call that did not fail. These results are not a full benchmark.

### Source excerpt

arXiv:2611.19274v1 Announce Type: cross Abstract: Long-horizon embodied tasks fail in ways that end-task success cannot see. HospitalBench is a diagnostic benchmark of ten hospital chores in NVIDIA Isaac Sim, each two unscored warm-up steps followed by a causal chain of five scored sub-tasks in one continuous simulation: the state one step leaves (an object pose, a choice, a fact seen once) determines what the next step must do, unlike benchmarks that place sub-tasks side by side. Vision-language agents act through skill-level tools from camera images. A scripted robot that makes only the calls a model gets is the solvability check of a chore. From each trajectory the benchmark reads where a chain breaks, what the agent did after a failed call, and how far its own position estimate drifts from the true one. This version reports first results only: 15 of 60 planned runs (three models at high effort, one run each on chores T1 and T10 from both start states and T3 from one), with scripted-robot proofs on three of the ten chores. Of the 15 runs, 2 passed all five sub-tasks. Runs still going after sub-tasks 1 to 5 were 15, 8, 4, 2, 2. Of 189 failed calls, 168 were followed by a different call and 111 by a call that did not fail. These are first results and not a full benchmark. We will release the scenes, task chains, scripted robot, evaluation harness and trajectories.

[Next page](<https://devfeed.tech/topics/robot-navigation.md?cursor=WyIyMDI2LTExLTE3VDA1OjAwOjAwKzAwOjAwIiwgIjk2MGJkZDZjLTM0MzctNGYyYy04NDQxLTUwZmY2YjExOWUwZCJd>)