# Robotics Simulation

Robotics simulation uses physics-based virtual environments to design, test, and train robotic systems before physical deployment.

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

## A four-level executable validation protocol for evaluating LLM-generated topology specifications of physics-based HVAC digital twins

DevFeed: [A four-level executable validation protocol for evaluating LLM-generated topology specifications of physics-based HVAC digital twins](<https://devfeed.tech/articles/a-four-level-executable-validation-protocol-for-evaluating-llm-generated-topology-specifications-of-physics-based-hvac-digital-twins-145061.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25471>)

Author: Younghwan Joo, Jeonghoon Han, Sanghyun Oh, Soosik Bang, Sung-il Kim

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

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

Topics: [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>)

Tags: [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-se](<https://devfeed.tech/tags/cs-se.md>), [cs-sy](<https://devfeed.tech/tags/cs-sy.md>), [digital-twins](<https://devfeed.tech/tags/digital-twins.md>), [eess-sy](<https://devfeed.tech/tags/eess-sy.md>), [llm](<https://devfeed.tech/tags/llm.md>), [physics](<https://devfeed.tech/tags/physics.md>), [protocol](<https://devfeed.tech/tags/protocol.md>), [specifications](<https://devfeed.tech/tags/specifications.md>), [topology](<https://devfeed.tech/tags/topology.md>), [validation](<https://devfeed.tech/tags/validation.md>)

### AI overview

The four-level protocol distinguishes executable HVAC digital twins from physically valid ones by checking static validity, model build, simulation convergence, and behavioral equivalence to a hidden reference within bands calibrated using ±20% sizing perturbations. Across eight LLMs and fourteen topologies, seven models built in 98.6-100% of trials after up to three repairs and simulations converged in at least 80%, but behavioral equivalence ranged from 50.0-98.6%. About half of failed trials still converged to a steady state different from the reference, mostly for desiccant rotor systems. Pooled pass rates differed by under five percentage points across description authors. The abstract argues that difficult domain physics should be represented explicitly in validation levels; it proposes adaptation to other fields that generate executable digital twins.

### Source excerpt

arXiv:2611.25471v1 Announce Type: cross Abstract: Physics-based digital twins of air conditioning, a principal energy consumer in manufacturing and buildings, can be generated automatically from a declarative specification of its equipment, connections, and sizing. Large language models (LLMs) now write such specifications from plain-language descriptions. Evaluations of these specifications stop at parse success, completed runs, or textual similarity, and cannot reject a model that converges in simulation but is physically wrong. In this study, a four-level protocol is proposed that tests whether an LLM-written specification not only executes but also yields a physically valid digital twin. Each specification is executed through static validity, model build, simulation convergence, and behavioral equivalence with a hidden reference model within bands calibrated by $\pm$20% sizing perturbations. The protocol was applied to eight LLMs and fourteen HVAC topologies, five describing an existing low-humidity test chamber. To test for authorship bias, engineers and a second model family also described six topologies. Within three repairs, seven of eight models built in 98.6 to 100% of trials, with simulations converging in at least 80.0%, but reached behavioral equivalence in only 50.0 to 98.6%. About half of the failed trials converged, but not to the reference steady state, mostly in desiccant rotor systems. Pooled pass rates varied by under five percentage points across description authors. These results imply that the physics hardest to model in each field, here the desiccant rotor, must enter the generation framework as validation levels. The protocol can be adapted to other fields that generate executable digital twins from declarative specifications.

## Mind2Act World: A Diagnostic Benchmark for Embodied Mind, Physical Action, and Mind-Action Composition in Robotic Manipulation

DevFeed: [Mind2Act World: A Diagnostic Benchmark for Embodied Mind, Physical Action, and Mind-Action Composition in Robotic Manipulation](<https://devfeed.tech/articles/mind2act-world-a-diagnostic-benchmark-for-embodied-mind-physical-action-and-mind-action-composition-in-robotic-manipulation-145673.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.26444>)

Author: Yuzhi Huang, Jie Wu, Fanding Huang, Junchao Lin, Weichen Zhang, Ziyi Xiong, Yang Li, Zhengwei Xie, Zhimin Wang, Zhen Liu, Jiazhen Huang, Yuzheng Zhang, Yuche Liu, Hongyi Yang, Chen Tang, Shixiang Tang

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>), [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>), [Physical AI](<https://devfeed.tech/topics/physical-ai.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [contact-rich manipulation](<https://devfeed.tech/topics/contact-rich-manipulation.md>)

Tags: [act](<https://devfeed.tech/tags/act.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [composition](<https://devfeed.tech/tags/composition.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>), [diagnostic](<https://devfeed.tech/tags/diagnostic.md>), [manipulation](<https://devfeed.tech/tags/manipulation.md>), [mind](<https://devfeed.tech/tags/mind.md>), [robotic](<https://devfeed.tech/tags/robotic.md>), [robotic-manipulation](<https://devfeed.tech/tags/robotic-manipulation.md>), [world](<https://devfeed.tech/tags/world.md>)

### AI overview

Mind2Act World is a benchmark for diagnosing whether robotic manipulation failures stem from reasoning, physical execution, or combining the two through environmental feedback. It contains 30 simulated tasks across five application domains, with three difficulty levels and Clean or Random scenes (180 configurations). Its Mind view tests decisions from past observations, Act tests precise and timely execution, and Mind2Act tests closed-loop tasks where action outcomes inform later decisions; a diagnosis harness attributes failures to these capabilities or their composition. In evaluations of multimodal agents, vision-language-action models, and world-action models, success in individual views did not transfer to combined tasks: the strongest agent succeeded in 52.7% of Mind episodes and 39.0% of Act episodes, but only 12.0% of Mind2Act episodes. The abstract reports that this agent retained task state but struggled with sustained contact, while end-to-end policies performed practiced motions but retained little task state.

### Source excerpt

arXiv:2611.26444v1 Announce Type: cross Abstract: Embodied intelligence aims to let robots perform complex physical tasks with human-like adaptability and reliability, which requires three complementary capabilities: history-based cognitive reasoning, precise and timely physical execution, and the composition of reasoning and execution through environmental feedback. Yet existing benchmarks either evaluate reasoning and execution within the same tasks or assess them separately without their composition, so they reveal little about whether a failure comes from reasoning, execution, or their composition. We introduce Mind2Act World, a diagnostic benchmark of 30 simulated manipulation tasks from five real-world application domains. Each task has three difficulty levels and two scene settings (Clean and Random), yielding 180 configurations. The tasks form three views: Mind tests decisions based on past observations, Act tests precise and timely execution, and Mind2Act combines both in closed-loop tasks where action outcomes guide later decisions. Comparing the three views shows whether a system lacks a capability or fails to compose capabilities it already has. A diagnosis harness further attributes the failed episodes of three representative systems to Mind, Act, or their composition and summarizes each system in a profile card. Our evaluation of multimodal agents, vision-language-action models (VLAs), and world-action models (WAMs) shows that competence in a single view does not carry over to Mind2Act tasks. Even the strongest agent succeeds in 52.7% of Mind and 39.0% of Act episodes but in only 12.0% of Mind2Act episodes. The strongest agent retains task state but struggles with sustained contact, whereas end-to-end policies execute practiced motions but retain little task state. These findings point to composing reasoning and execution through environmental feedback as the main open problem.

## WorldComposer: Generative High-Fidelity Simulation with Digital Cousins for Generalizable Robot Learning and Evaluation

DevFeed: [WorldComposer: Generative High-Fidelity Simulation with Digital Cousins for Generalizable Robot Learning and Evaluation](<https://devfeed.tech/articles/worldcomposer-generative-high-fidelity-simulation-with-digital-cousins-for-generalizable-robot-learning-and-evaluation-142517.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2604.15805>)

Author: Jasper Lu, Zhenhao Shen, Yuanfei Wang, Shugao Liu, Shengqiang Xu, Shawn Xie, Jingkai Xu, Feng Jiang, Jade Yang, Chen Xie, Ruihai Wu

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

Content type: article

Language: en

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

Topics: [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>), [Robot Manipulation](<https://devfeed.tech/topics/robot-manipulation.md>), [Dataset development](<https://devfeed.tech/topics/dataset-development.md>)

Tags: [automated](<https://devfeed.tech/tags/automated.md>), [correlation](<https://devfeed.tech/tags/correlation.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [data-augmentation](<https://devfeed.tech/tags/data-augmentation.md>), [engines](<https://devfeed.tech/tags/engines.md>), [environments](<https://devfeed.tech/tags/environments.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [framework](<https://devfeed.tech/tags/framework.md>), [generation](<https://devfeed.tech/tags/generation.md>), [generative](<https://devfeed.tech/tags/generative.md>), [interactive](<https://devfeed.tech/tags/interactive.md>), [learning](<https://devfeed.tech/tags/learning.md>), [manipulation](<https://devfeed.tech/tags/manipulation.md>), [maps](<https://devfeed.tech/tags/maps.md>), [object](<https://devfeed.tech/tags/object.md>), [physics](<https://devfeed.tech/tags/physics.md>), [real-world](<https://devfeed.tech/tags/real-world.md>), [robot](<https://devfeed.tech/tags/robot.md>), [sim-to-real](<https://devfeed.tech/tags/sim-to-real.md>), [simulation](<https://devfeed.tech/tags/simulation.md>), [tasks](<https://devfeed.tech/tags/tasks.md>)

### AI overview

WorldComposer maps real-world panoramas into task-ready simulations and generates "Digital Cousins" by varying scenes and objects, aiming to expand robot-learning data without repeatedly acquiring physical assets or reconfiguring environments. The framework combines realistic assets and physics engines for manipulation of rigid, articulated, and deformable objects, and stitches rooms into larger environments for long-horizon tasks. The abstract reports strong sim-to-real correlation and improved generalization to unseen scene and object variations as generated data scales up.

### Source excerpt

arXiv:2604.15805v2 Announce Type: replace-cross Abstract: Learning robust robot policies in real-world environments requires diverse data augmentation, yet scaling real-world data collection is costly due to the need for acquiring physical assets and reconfiguring environments. Therefore, augmenting real-world scenes into simulation has become a practical augmentation for efficient learning and evaluation. We present WorldComposer, an automated generative Real2Sim2Real framework that maps real-world panoramas into high-fidelity, task-ready simulation environments and further synthesizes diverse Digital Cousins through scene and object variations. Combined with high-quality physics engines and realistic assets, WorldComposer supports interactive manipulation across rigid, articulated, and deformable objects. Additionally, we incorporate multi-room stitching to construct consistent large-scale environments for long-horizon tasks. Experiments demonstrate a strong sim-to-real correlation validating our platform's fidelity, and show that extensively scaling up data generation leads to significantly better generalization to unseen scene and object variations, demonstrating the effectiveness of Digital Cousins for generalizable robot learning and evaluation.

## Auditing a Learned Simulator for Legged Robots: The Robotic World Model on ANYmal D

DevFeed: [Auditing a Learned Simulator for Legged Robots: The Robotic World Model on ANYmal D](<https://devfeed.tech/articles/auditing-a-learned-simulator-for-legged-robots-the-robotic-world-model-on-anymal-d-141845.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23118>)

Author: Joyjeet Singh

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>), [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>), [audit](<https://devfeed.tech/topics/audit.md>), [Neural Network](<https://devfeed.tech/topics/neural-network.md>), [Physical AI](<https://devfeed.tech/topics/physical-ai.md>)

Tags: [anymal](<https://devfeed.tech/tags/anymal.md>), [auditing](<https://devfeed.tech/tags/auditing.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [legged-robots](<https://devfeed.tech/tags/legged-robots.md>), [model](<https://devfeed.tech/tags/model.md>), [mujoco](<https://devfeed.tech/tags/mujoco.md>), [policies](<https://devfeed.tech/tags/policies.md>), [policy](<https://devfeed.tech/tags/policy.md>), [pruning](<https://devfeed.tech/tags/pruning.md>), [simulator](<https://devfeed.tech/tags/simulator.md>), [training](<https://devfeed.tech/tags/training.md>), [uncertainty](<https://devfeed.tech/tags/uncertainty.md>), [variance](<https://devfeed.tech/tags/variance.md>), [world-model](<https://devfeed.tech/tags/world-model.md>)

### AI overview

The audit finds that one-step accuracy is a poor guide to long rollouts and policy performance for the ANYmal D Robotic World Model. Training on the model's own rollouts performed better at 368 steps across ten held-out episodes, though one-step training was better one step ahead; pruning 30% of weights retained 90% of one-step skill but 39% at two seconds. Ensemble disagreement tracked rollout error well (Spearman +0.90 at 100 steps), but its one-sigma coverage was only 4.6%, versus 68% for a calibrated estimate, because the networks' predicted variances collapsed under the implemented loss. Across 18 fine-tuned copies, all improved one- and eight-step predictions over the released model, yet 13 produced failing policies. Registered tests of whether the proposed checks rank simulators by policy performance remained inconclusive; running the authors' policy inside each simulator ranked them well in a secondary test. The audit covers one robot and gait, simulated data only, and no hardware.

### Source excerpt

arXiv:2611.23118v1 Announce Type: cross Abstract: A legged robot can learn to walk inside a learned simulator: a neural network that predicts the robot's next state and is run forward on its own predictions. They are usually trained and checked one step ahead, although a policy uses them for a hundred steps or more. We audit one published system end to end, without a GPU: the dynamics model of the Robotic World Model (RWM) for the ANYmal D quadruped, and the uncertainty estimate of its follow-up, RWM-U. We rebuilt the model to match the released code's outputs, losses and gradients exactly, measured 101 trained simulators, 200 compressed copies and 38 simulators of three other architectures, and trained policies inside simulators that we tested in a hand-built MuJoCo model of the robot. Training on the model's own rollouts beat one-step training at 368 steps on all ten held-out episodes (registered), although one-step training was better one step ahead. One-step checks therefore misjudge design choices: pruning 30% of the weights kept 90% of the model's skill one step ahead and 39% at 2 s. Ensemble disagreement ranked rollout error well (Spearman +0.90 at 100 steps), yet covered 4.6% of outcomes at one sigma, against 68% for a calibrated estimate, as each network's own variance collapses under the implemented loss. Two checks that need no recorded future ranked some pools of simulators well, but neither was shown better than 8-step error in a primary test. At the robot level, accuracy has not predicted walking: all 18 fine-tuned copies we tested predicted better than the released model one and eight steps ahead, yet 13 gave failing policies, and registered tests of whether any check ranks simulators by their policies were not settled. In a secondary test, running the authors' policy inside each simulator ranked them well. We also report nine of our own claims that we withdrew. Scope: one robot, one gait, simulated data, no hardware.

## Magflow: Differentiable Simulation of Industrial-Grade 6D Magnetic Levitation

DevFeed: [Magflow: Differentiable Simulation of Industrial-Grade 6D Magnetic Levitation](<https://devfeed.tech/articles/magflow-differentiable-simulation-of-industrial-grade-6d-magnetic-levitation-141836.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23050>)

Author: Philipp Hartmann, Jannick Strangh\\"oner, Klaus Neumann

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

Content type: news

Language: en

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

Topics: [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>), [Simulation](<https://devfeed.tech/topics/simulation.md>), [Physical AI](<https://devfeed.tech/topics/physical-ai.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Robotics](<https://devfeed.tech/topics/robotics.md>), [GPU](<https://devfeed.tech/topics/gpu.md>)

Tags: [co-design](<https://devfeed.tech/tags/co-design.md>), [control](<https://devfeed.tech/tags/control.md>), [controller](<https://devfeed.tech/tags/controller.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [cs-sy](<https://devfeed.tech/tags/cs-sy.md>), [eess-sy](<https://devfeed.tech/tags/eess-sy.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [industrial](<https://devfeed.tech/tags/industrial.md>), [learning](<https://devfeed.tech/tags/learning.md>), [magnetic](<https://devfeed.tech/tags/magnetic.md>), [neural](<https://devfeed.tech/tags/neural.md>), [physics](<https://devfeed.tech/tags/physics.md>), [real-time](<https://devfeed.tech/tags/real-time.md>), [simulation](<https://devfeed.tech/tags/simulation.md>), [simulator](<https://devfeed.tech/tags/simulator.md>), [topology](<https://devfeed.tech/tags/topology.md>), [zero-shot](<https://devfeed.tech/tags/zero-shot.md>)

### AI overview

Magflow uses a fully differentiable, GPU-parallel simulator to model 6D magnetic levitation directly, avoiding hand-crafted electromagnetic force approximations used in industrial systems. Those approximations help make kilohertz control tractable but constrain the operating workspace and supported mover topologies. The authors report training an industrial-grade neural controller purely in simulation that transfers zero-shot to real hardware, meets 4 kHz hard real-time constraints, exhibits new behaviors, and levitates a mover topology lacking a hand-crafted solution. These results suggest hardware-software co-design is not physically necessary for this control task.

### Source excerpt

arXiv:2611.23050v1 Announce Type: cross Abstract: Magnetic levitation with six-degrees-of-freedom enables frictionless transport and micrometer-scale positioning in principle, but low-level control is inherently challenging due to its complex and unstable dynamics. Industrial systems typically rely on deep hardware-software co-design, where mover and stator geometries are jointly designed to enable a compact, hand-crafted approximation of the electromagnetic force computation, thereby keeping control computationally tractable at kilohertz rates. However, this approach imposes overly conservative constraints on the operating workspace and the range of supported mover topologies. Instead of reducing the underlying physics to a hand-crafted approximation, we compute the physics directly. We introduce Magflow, a fully differentiable, GPU-parallel simulator for 6D magnetic levitation. It is used to train the first neural controller for an industrial-grade 6D magnetic levitation system that is learned purely from simulated physics. We show that the learned neural controller transfers zero-shot from simulation to real hardware while satisfying hard real-time constraints at 4 kHz. The learned controller achieves qualitatively new behaviors and even levitates a mover topology for which no hand-crafted solution exists. These results demonstrate that co-design is not a physical necessity for 6D magnetic levitation control. More broadly, control learning from simulated physics can find solutions beyond the reach of control engineering in demanding industrial domains. Source code and demonstration videos are publicly available at https://sites.google.com/view/magflow.

## When the World Is the Test Suite: Agentic Intelligence for the Physical World

DevFeed: [When the World Is the Test Suite: Agentic Intelligence for the Physical World](<https://devfeed.tech/articles/when-the-world-is-the-test-suite-agentic-intelligence-for-the-physical-world-139601.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.22291>)

Author: Kaixin Li, Meng Luo, Yanlin Li, Yuchen Tian, Xiang Liu, Xiangru Jian, Yan Yang, Zhiheng Xi, Yang Zhang, Ziyang Luo, Mingfei Cheng, Hongzhan Lin

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

Content type: article

Language: en

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

Topics: [Physical AI](<https://devfeed.tech/topics/physical-ai.md>), [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>), [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>), [Test automation](<https://devfeed.tech/topics/test-automation.md>)

Tags: [actuator](<https://devfeed.tech/tags/actuator.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agents](<https://devfeed.tech/tags/agents.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [code](<https://devfeed.tech/tags/code.md>), [coding](<https://devfeed.tech/tags/coding.md>), [coding-agents](<https://devfeed.tech/tags/coding-agents.md>), [continuous-integration](<https://devfeed.tech/tags/continuous-integration.md>), [controllers](<https://devfeed.tech/tags/controllers.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [cs-se](<https://devfeed.tech/tags/cs-se.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [experience](<https://devfeed.tech/tags/experience.md>), [formal](<https://devfeed.tech/tags/formal.md>), [industrial](<https://devfeed.tech/tags/industrial.md>), [intelligence](<https://devfeed.tech/tags/intelligence.md>), [language-models](<https://devfeed.tech/tags/language-models.md>), [learning](<https://devfeed.tech/tags/learning.md>), [libraries](<https://devfeed.tech/tags/libraries.md>), [test](<https://devfeed.tech/tags/test.md>), [world](<https://devfeed.tech/tags/world.md>)

### AI overview

The paper frames physical-world agentic intelligence around a central challenge: programs for robots and other physical systems lack a built-in test suite, and trustworthy evidence of their behavior is costly and may be flawed. It surveys how agents write robot policies and programs that support policy learning, how physical trials compare with traces, learned judges, simulators, and formal checks, and why safety assessment must go beyond task success. It also examines how tested programs become reusable experience through skill libraries and model weights, alongside benchmarks and open evaluation challenges. The authors argue for a physical counterpart to continuous integration: verification that is cheap to run often, reliable during search, and maintained with the experience it validates.

### Source excerpt

arXiv:2611.22291v1 Announce Type: cross Abstract: Coding agents built to repair software now write, test, and repair programs for robots, laboratories, and industrial controllers. Unlike in software engineering, no test suite comes with the code. Whether a program works must be established by running it on a robot, in a physics simulator, or through the policies it helps to train, and such evidence is costly to obtain and can itself be wrong. As these agents take on more physical tasks, how their programs are judged matters as much as how they are written. This paper examines this emerging area, agentic intelligence for the physical world, by treating the world as the program's test suite. We organize the discussion around three questions: how agents write programs for the physical world, how these programs are tested, and how agents learn from the results. We first trace how language models for robots and software engineering agents came together, ask which software testing practices carry over to physical tests, and describe each program by where it sits relative to the actuator and by what judges it. For writing, we cover programs that act as robot policies and programs that support policy learning by producing rewards, tasks, and training worlds. For testing, we examine what physical trials cost, what traces, learned judges, simulators, and formal checks can establish in their place, and what safety requires beyond task success. For learning, we discuss how tested programs are retained as experience, in skill libraries and in model weights. Finally, we turn to benchmarks and evaluation practice and outline open challenges. Writing programs is becoming cheap, while trustworthy evidence about what they do remains costly. Progress will therefore depend on a physical counterpart of continuous integration: verification cheap enough to run often, reliable enough to trust under search, and maintained alongside the experience it validates.

## From Attack Tree to Model Checker: Binding CAPEC Patterns to Verify Robotic System Security

DevFeed: [From Attack Tree to Model Checker: Binding CAPEC Patterns to Verify Robotic System Security](<https://devfeed.tech/articles/from-attack-tree-to-model-checker-binding-capec-patterns-to-verify-robotic-system-security-141487.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.22408>)

Author: Juel Hussain (Department of Computer Science, University of Manchester, UK), Louise A. Dennis (Department of Computer Science, University of Manchester, UK), Clare Dixon (Department of Computer Scienc

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

Content type: article

Language: en

Sources: [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>)

Topics: [Security](<https://devfeed.tech/topics/security.md>), [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>), [ai security](<https://devfeed.tech/topics/ai-security.md>)

Tags: [approach](<https://devfeed.tech/tags/approach.md>), [attack](<https://devfeed.tech/tags/attack.md>), [autonomous](<https://devfeed.tech/tags/autonomous.md>), [configuration](<https://devfeed.tech/tags/configuration.md>), [controls](<https://devfeed.tech/tags/controls.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [cs-lo](<https://devfeed.tech/tags/cs-lo.md>), [effect](<https://devfeed.tech/tags/effect.md>), [logic](<https://devfeed.tech/tags/logic.md>), [model](<https://devfeed.tech/tags/model.md>), [patterns](<https://devfeed.tech/tags/patterns.md>), [pipeline](<https://devfeed.tech/tags/pipeline.md>), [property](<https://devfeed.tech/tags/property.md>), [reasoning](<https://devfeed.tech/tags/reasoning.md>), [robotic](<https://devfeed.tech/tags/robotic.md>), [robotics](<https://devfeed.tech/tags/robotics.md>), [security](<https://devfeed.tech/tags/security.md>), [system](<https://devfeed.tech/tags/system.md>), [temporal](<https://devfeed.tech/tags/temporal.md>), [threat](<https://devfeed.tech/tags/threat.md>), [verification](<https://devfeed.tech/tags/verification.md>), [verify](<https://devfeed.tech/tags/verify.md>)

### AI overview

The research preview proposes a pipeline that turns robotic-system attack trees into checkable attacker models using CAPEC patterns. Each attack-tree leaf is expressed as a first-order rule linking attacker prerequisites to an attack effect; binding its variables to system components lets the prerequisites be evaluated under selected security controls. The pipeline derives an attacker model and a temporal-logic property from the bound pattern, then model-checks the system with and without a countermeasure and reports the control that defeats the attack. The abstract also says the work describes a signature, schemas, and profiles, but provides no evaluation results or detailed performance evidence.

### Source excerpt

arXiv:2611.22408v1 Announce Type: new Abstract: Security-minded verification of autonomous and robotic systems presupposes an attacker, yet the path from a threat catalogue to a checkable attacker model is still traversed by hand. This research preview presents the theory and construction of a pipeline that mechanises it. Each leaf of an attack tree is mapped to a CAPEC attack pattern and written in first-order logic as a rule: the attacker's prerequisites imply the attack's effect. Casting the pattern as such a rule is what makes it operational - its variables can be bound to a specific system's parts, and its prerequisites can then be tested against that system under a chosen configuration of security controls. Binding is our means of verifying whether a system is vulnerable to a given attack: from a bound pattern we derive an attacker model, turn its effect into a temporal-logic property, and check that property against the system with and without a countermeasure, reporting the control that defeats the attack. We describe the signature, schemas, profiles for approach, and explore how CAPEC can support reasoning about robotics security.

## 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.

## Physics-Informed Causal Reasoning for Nonstationary Robot Dynamics

DevFeed: [Physics-Informed Causal Reasoning for Nonstationary Robot Dynamics](<https://devfeed.tech/articles/physics-informed-causal-reasoning-for-nonstationary-robot-dynamics-138848.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.19251>)

Author: Jungjin Lee, Se Jin Kim, Sung Kwon On, Insoon Yang

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

Content type: article

Language: en

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

Topics: [Isaac Sim](<https://devfeed.tech/topics/isaac-sim.md>), [Physical AI](<https://devfeed.tech/topics/physical-ai.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>)

Tags: [context](<https://devfeed.tech/tags/context.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [motion](<https://devfeed.tech/tags/motion.md>), [neural-networks](<https://devfeed.tech/tags/neural-networks.md>), [nvidia-isaac-sim](<https://devfeed.tech/tags/nvidia-isaac-sim.md>), [payload](<https://devfeed.tech/tags/payload.md>), [physics-informed](<https://devfeed.tech/tags/physics-informed.md>), [reasoning](<https://devfeed.tech/tags/reasoning.md>), [robot](<https://devfeed.tech/tags/robot.md>)

### AI overview

PICR adapts robot dynamics to changing physical conditions by re-inferring a context from recent interactions and using it to condition a fixed physics-informed network for inertia and potential energy. It requires neither retraining nor a hand-specified robot model. In NVIDIA Isaac Sim experiments with a Franka Panda arm, PICR predicted motion under changing and unseen payloads, with 1.6-3.8x lower error on unseen payloads than the reported baselines, and supported counterfactual queries under alternative torques. The authors report that interaction-inferred context tracked payload properties and changes; context-induced gravity and inertia terms agreed with the simulator at the level of individual Lagrangian components.

### Source excerpt

arXiv:2611.19251v1 Announce Type: cross Abstract: Reasoning about what an action will cause requires internalizing how a system behaves. As conditions seldom stay fixed, this must often happen within a few interactions. Yet systems that look identical in a snapshot can respond differently to the same action, depending on physical conditions the snapshot hides. Context-based models capture this separation by summarizing recent transitions into a latent context, but do not specify how that context should change the equations of motion. Physics-informed neural networks build those equations into the model, but describe only a single system and cannot recognize when it has changed. We present Physics-Informed Causal Reasoning (PICR), which bridges the two. Only a context, re-inferred from recent interactions, is updated; it conditions a fixed physics-informed network that determines the robot's inertia and potential energy, so the robot adapts to new physical conditions from its own experience without retraining or a hand-specified robot model. On a Franka Panda arm in NVIDIA Isaac Sim, PICR predicts motion under changing and unseen payloads, reducing error on the latter by 1.6-3.8x over baselines, and extends to counterfactual queries under alternative torques. Our analysis shows that the context, inferred solely from interaction, encodes payload physical properties, tracks their changes, and governs the arm's response. The gravity and inertia terms induced by the context further demonstrate its physical validity, each agreeing with the simulator at the level of individual Lagrangian components.

## 28 Zero-Day Findings in Under Two Days. Is Robotics Security Keeping Pace?

DevFeed: [28 Zero-Day Findings in Under Two Days. Is Robotics Security Keeping Pace?](<https://devfeed.tech/articles/28-zero-day-findings-in-under-two-days-is-robotics-security-keeping-pace-135486.md>)

Original publisher: [Read original article](<https://spectrum.ieee.org/zero-day-robotics-vulnerabilities-vicone-security>)

Author: VicOne

Published: 2026-11-16T10:00:01Z

Content type: article

Language: en

Sources: [IEEE Spectrum](<https://devfeed.tech/sources/ieee-spectrum.md>)

Topics: [Security research](<https://devfeed.tech/topics/security-research.md>), [Physical AI](<https://devfeed.tech/topics/physical-ai.md>), [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>), [AI software supply chain security](<https://devfeed.tech/topics/ai-software-supply-chain-security.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-assisted](<https://devfeed.tech/tags/ai-assisted.md>), [cve](<https://devfeed.tech/tags/cve.md>), [cybersecurity](<https://devfeed.tech/tags/cybersecurity.md>), [deployment](<https://devfeed.tech/tags/deployment.md>), [development](<https://devfeed.tech/tags/development.md>), [discovery](<https://devfeed.tech/tags/discovery.md>), [exploited](<https://devfeed.tech/tags/exploited.md>), [google](<https://devfeed.tech/tags/google.md>), [lifecycle](<https://devfeed.tech/tags/lifecycle.md>), [meta](<https://devfeed.tech/tags/meta.md>), [physical-ai](<https://devfeed.tech/tags/physical-ai.md>), [research](<https://devfeed.tech/tags/research.md>), [robot-safety](<https://devfeed.tech/tags/robot-safety.md>), [robotics](<https://devfeed.tech/tags/robotics.md>), [robots](<https://devfeed.tech/tags/robots.md>), [runtime](<https://devfeed.tech/tags/runtime.md>), [safety](<https://devfeed.tech/tags/safety.md>), [security](<https://devfeed.tech/tags/security.md>), [simulation](<https://devfeed.tech/tags/simulation.md>), [software](<https://devfeed.tech/tags/software.md>), [zero-day](<https://devfeed.tech/tags/zero-day.md>), [zero-day-vulnerabilities](<https://devfeed.tech/tags/zero-day-vulnerabilities.md>)

### AI overview

VicOne reports 28 initially validated zero-day findings identified in under two days with AI-assisted research across Physical AI software; vendor and vulnerability-authority review may change the count, and some findings lack CVE numbers. Three quarters involved development, training, simulation, or validation tools. The excerpt describes potential effects including altered models or test results and disrupted operator visibility or control, while noting actual impact depends on access, system design, and safeguards. It also cites a demonstration in which researchers prevented a robot's remote-stop command from working. Teams are advised to locate affected software across training, simulation, and deployed systems; assess what a weakness could change; verify fixes preserve intended behavior; and consider temporary access restrictions when immediate updating is impractical.

### Source excerpt

This article is brought to you by VicOne. VicOne identified an initial set of 28 zero-day findings in less than two days using AI-assisted research across software and tools supporting the Physical AI ecosystem. These include projects developed or maintained by Google, NVIDIA, and Meta, among others. Based on initial validation, all 28 findings are classified as zero-day vulnerabilities. The count may change following vendor and vulnerability-authority review, and some findings do not yet have CVE tracking numbers. Many of these tools help teams train, simulate, test, and operate robots. Weaknesses in them could influence the models teams build, the test results they trust, and the instructions sent to robots. These findings create two demands for robotics teams: responding as AI helps uncover weaknesses faster, and assessing how each one could affect robot safety and operations. What AI changes about vulnerability discovery Vulnerability discovery involves examining code, tracing how systems communicate, and testing whether a suspected weakness can actually be exploited. AI can accelerate parts of this work, helping researchers analyze more software in less time. Attackers could use similar capabilities. When the weakness is a zero-day -- previously unknown and without an available fix -- faster discovery can leave teams with little time to respond. While that exposure does not necessarily mean an attack has occurred, teams still need to determine whether they use the affected software and what an attacker could access or change. Where the findings sit across the robot lifecycle VicOne's findings span three stages of the robotics lifecycle: development and training, simulation and validation, and deployment and runtime. Weaknesses at each stage could affect what a robot learns, how it is tested, and how it is controlled. Figure 1. In this research set, 75% of findings involved development, training, simulation, or validation tools, highlighting risks that can arise b

## A Generalizable Approach to Safety-Aware Robotic Behavior Planning with an Explicit Constraint Space

DevFeed: [A Generalizable Approach to Safety-Aware Robotic Behavior Planning with an Explicit Constraint Space](<https://devfeed.tech/articles/a-generalizable-approach-to-safety-aware-robotic-behavior-planning-with-an-explicit-constraint-space-135859.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.18026>)

Author: Xianxu Wu, Shang-Ching Liu, Zhaocong Liang, Hengxiang Chen, Yujian Ma, Shenwen Deng, Qiang Li, Jianwei Zhang

Published: 2026-11-16T05:00:00Z

Content type: article

Language: en

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

Topics: [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>), [ai safety](<https://devfeed.tech/topics/ai-safety.md>), [lora](<https://devfeed.tech/topics/lora.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>), [Humanoid Robots](<https://devfeed.tech/topics/humanoid-robots.md>)

Tags: [behavior](<https://devfeed.tech/tags/behavior.md>), [constraint](<https://devfeed.tech/tags/constraint.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [framework](<https://devfeed.tech/tags/framework.md>), [image](<https://devfeed.tech/tags/image.md>), [inspection](<https://devfeed.tech/tags/inspection.md>), [knowledge](<https://devfeed.tech/tags/knowledge.md>), [language-models](<https://devfeed.tech/tags/language-models.md>), [learning](<https://devfeed.tech/tags/learning.md>), [lora](<https://devfeed.tech/tags/lora.md>), [manipulation](<https://devfeed.tech/tags/manipulation.md>), [model](<https://devfeed.tech/tags/model.md>), [multimodal](<https://devfeed.tech/tags/multimodal.md>), [natural-language](<https://devfeed.tech/tags/natural-language.md>), [planner](<https://devfeed.tech/tags/planner.md>), [planning](<https://devfeed.tech/tags/planning.md>), [robotic](<https://devfeed.tech/tags/robotic.md>), [rules](<https://devfeed.tech/tags/rules.md>), [safety](<https://devfeed.tech/tags/safety.md>), [selection](<https://devfeed.tech/tags/selection.md>), [space](<https://devfeed.tech/tags/space.md>)

### AI overview

The framework addresses tabletop manipulation hazards that depend on scene context and may be absent from task instructions. It stores reusable textual safety rules in an explicit Constraint Space and uses a query-conditioned dual-tower model to select rules based on task and environment, rather than semantic similarity alone. A multimodal planner receives the selected rules with the task and image. A Safety Inspector checks primitive plans, reconstructs applicable constraints, and returns corrective feedback to address mistaken selections and unseen hazards. Validated feedback supports test-time adaptation of rule selection and LoRA-based adaptation of the Inspector; newly identified constraints can also be added to the Constraint Space.

### Source excerpt

arXiv:2611.18026v1 Announce Type: cross Abstract: Large language and vision-language models have substantially advanced robotic behavior planning from natural-language instructions and visual observations. However, safe manipulation in tabletop environments remains challenging because the safety constraints required for a task depend on the surrounding scene and may not be explicitly specified in the instruction. We propose a safety-aware behavior planning framework based on an explicit Constraint Space that stores reusable textual safety rules. Rather than retrieving rules solely by semantic similarity, we formulate constraint selection as task- and environment-conditioned applicability learning. A lightweight query-conditioned dual-tower model identifies the constraints applicable to the current context, which are then provided together with the original task and image to a multimodal behavior planner. To handle incorrect constraint selections and previously unseen hazards, we introduce a multimodal Safety Inspector that evaluates generated primitive plans, reconstructs the complete applicable constraint set, and provides corrective planning feedback. Validated inspection feedback is further used for test-time adaptation of constraint applicability and parameter-efficient adaptation of the Safety Inspector through LoRA, while newly identified constraints are incorporated into the Constraint Space. The resulting framework integrates explicit safety knowledge representation,context-aware constraint selection, corrective safety inspection, and test-time knowledge adaptation for safety-aware robotic behavior planning.

## Build a Robot Pick-and-Place Simulation with SimReady Assets

DevFeed: [Build a Robot Pick-and-Place Simulation with SimReady Assets](<https://devfeed.tech/articles/build-a-robot-pick-and-place-simulation-with-simready-assets-133556.md>)

Original publisher: [Read original article](<https://huggingface.co/blog/nvidia/simready-assets-robot-pick-and-place>)

Author: Dillon Bailey; Pomi Lee; Rishabh Chadha

Published: 2026-11-13T18:00:39Z

Content type: tutorial

Language: en

Sources: [Hugging Face](<https://devfeed.tech/sources/hugging-face-blog.md>)

Topics: [Isaac Sim](<https://devfeed.tech/topics/isaac-sim.md>), [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>), [Robot Manipulation](<https://devfeed.tech/topics/robot-manipulation.md>), [Omniverse](<https://devfeed.tech/topics/omniverse.md>), [Isaac](<https://devfeed.tech/topics/isaac.md>), [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [AI Chat](<https://devfeed.tech/topics/ai-chat.md>), [Tutorial](<https://devfeed.tech/topics/tutorial.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [arm](<https://devfeed.tech/tags/arm.md>), [asset](<https://devfeed.tech/tags/asset.md>), [assets](<https://devfeed.tech/tags/assets.md>), [behavior](<https://devfeed.tech/tags/behavior.md>), [build](<https://devfeed.tech/tags/build.md>), [can](<https://devfeed.tech/tags/can.md>), [compose](<https://devfeed.tech/tags/compose.md>), [controls](<https://devfeed.tech/tags/controls.md>), [example](<https://devfeed.tech/tags/example.md>), [isaac-sim](<https://devfeed.tech/tags/isaac-sim.md>), [openusd](<https://devfeed.tech/tags/openusd.md>), [pick-and-place](<https://devfeed.tech/tags/pick-and-place.md>), [properties](<https://devfeed.tech/tags/properties.md>), [robot](<https://devfeed.tech/tags/robot.md>), [robotics](<https://devfeed.tech/tags/robotics.md>), [run](<https://devfeed.tech/tags/run.md>), [simulation](<https://devfeed.tech/tags/simulation.md>), [use-case](<https://devfeed.tech/tags/use-case.md>), [uses](<https://devfeed.tech/tags/uses.md>)

### AI overview

The walkthrough uses SimReady Central assets to build and test a scripted beverage-can pick-and-place task in Isaac Sim with a UR17G arm and Robotiq 2F-140 gripper. SimReady profiles define use-case features and requirements; validation checks authored settings such as joint drives, while the assembled simulation tests whether the robot can grasp, move, and release the can. The workflow downloads complete asset folders, assembles and validates the arm and gripper, then runs agent-generated controls, checking joint settings, motion, contact, and whether the can stays held. Isaac Sim 6.2 and an agent able to use its skills are required for the described agent-assisted setup; Hugging Face CLI is optional for command-line downloads. The example uses a fixed setup, scripted waypoints, and a 2 N-m drive-effort cap selected through simulation tests. It omits vision, limits collision checks, and does not model can deformation or liquid motion. Results do not establish robustness to uncertain positions, other payloads, or real hardware.

### Source excerpt

This walkthrough showcases a beverage can pick-and-place task that uses the UR17G arm, Robotiq 2F-140 gripper, and beverage can to compose a scene in Isaac Sim, check their behavior, and run a pick-and-place example with agent-generated controls. For robotics workflows, SimReady means an OpenUSD asset has been prepared with the properties needed for a defined simulation use case, such as collisions, mass, joints, and joint drives.

## RUMI: A Robotiq-Matched Universal Manipulation Interface for Robot Learning with Augmented Reality and Haptic Feedback

DevFeed: [RUMI: A Robotiq-Matched Universal Manipulation Interface for Robot Learning with Augmented Reality and Haptic Feedback](<https://devfeed.tech/articles/rumi-a-robotiq-matched-universal-manipulation-interface-for-robot-learning-with-augmented-reality-and-haptic-feedback-132173.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.15802>)

Author: Kevin Avdiu, Seongjin Bien, Yannik Blei, Philine Meister, Johannes Hechtl, Philipp Schmitt, Kai Wurm, Georg von Wichert, Wolfram Burgard

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

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

Topics: [Robot Manipulation](<https://devfeed.tech/topics/robot-manipulation.md>), [robot sense of touch](<https://devfeed.tech/topics/robot-sense-of-touch.md>), [contact-rich manipulation](<https://devfeed.tech/topics/contact-rich-manipulation.md>), [Glasses & AR Tech](<https://devfeed.tech/topics/glasses-ar-tech.md>), [Virtual reality](<https://devfeed.tech/topics/virtual-reality.md>), [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>)

Tags: [augmented-reality](<https://devfeed.tech/tags/augmented-reality.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [feedback](<https://devfeed.tech/tags/feedback.md>), [haptic-feedback](<https://devfeed.tech/tags/haptic-feedback.md>), [interface](<https://devfeed.tech/tags/interface.md>), [learning](<https://devfeed.tech/tags/learning.md>), [manipulation](<https://devfeed.tech/tags/manipulation.md>), [robot](<https://devfeed.tech/tags/robot.md>)

### AI overview

RUMI is a handheld interface matched to the Robotiq 2F-85 that preserves its gripper geometry and uses exchangeable fingertip sensors for natural input before contact and force measurement after contact. Tracking integrated into a VR headset provides real-time augmented reality feedback on whether motions are kinematically feasible for the target robot; a haptic mechanism supports force-aware teleoperation with simulated and real robots. In a user study, episodes with joint-limit violations were fewer with live robot visualization and pose initialization (3/50) than without augmented reality (16/50). The authors also trained imitation policies from demonstrations collected with two fingertip modalities and assessed force-aware data collection in simulation. The supplied text is an abstract excerpt, so it does not provide implementation details or broader evaluation results.

### Source excerpt

arXiv:2611.15802v1 Announce Type: cross Abstract: Many manipulation tasks require carefully controlled contact forces for successful execution, yet large-scale, force-aware datasets remain scarce. Handheld interfaces can help by transmitting interaction forces directly through the operator's arm, but most existing designs are tailored to simple linear grippers and do not match the nonlinear mechanics of widely used robotic grippers. We present RUMI, a handheld universal manipulation interface matched to the Robotiq 2F-85 with exchangeable fingertip sensors that preserves the target gripper geometry while enabling natural input before contact and force measurement after contact. By integrating the tracking system into a virtual reality headset, our method provides real-time augmented reality feedback about the kinematic feasibility of motion on the target robot platform. We further introduce a haptic feedback mechanism that supports force-aware teleoperation for simulated and real robots. A user study shows fewer episodes containing joint-limit violations with live robot visualization and pose initialization (3/50) than without augmented reality (16/50). We additionally train imitation policies from demonstrations, collected using two different fingertip modalities enabled by the gripper-matched hardware, and assess the haptic interface's capability for collecting force-aware teleoperation datasets in simulation.

## Robo-SWE: An Executable Issue-Resolution Dataset Mined from Robot Software Repositories

DevFeed: [Robo-SWE: An Executable Issue-Resolution Dataset Mined from Robot Software Repositories](<https://devfeed.tech/articles/robo-swe-an-executable-issue-resolution-dataset-mined-from-robot-software-repositories-131728.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.16116>)

Author: Vahid Haratian, Vahid Ansari, Mohammadhossein Malmir, Alois Knoll

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

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

Topics: [agentic-coding](<https://devfeed.tech/topics/agentic-coding.md>), [issue tracking](<https://devfeed.tech/topics/issue-tracking.md>), [Test-driven development](<https://devfeed.tech/topics/tdd.md>), [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [algorithm](<https://devfeed.tech/tags/algorithm.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [cs-se](<https://devfeed.tech/tags/cs-se.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [execution](<https://devfeed.tech/tags/execution.md>), [fairness](<https://devfeed.tech/tags/fairness.md>), [frameworks](<https://devfeed.tech/tags/frameworks.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [issue](<https://devfeed.tech/tags/issue.md>), [kimi-k3-1-pro](<https://devfeed.tech/tags/kimi-k3-1-pro.md>), [leaderboard](<https://devfeed.tech/tags/leaderboard.md>), [ledger](<https://devfeed.tech/tags/ledger.md>), [libraries](<https://devfeed.tech/tags/libraries.md>), [offline](<https://devfeed.tech/tags/offline.md>), [operating-system](<https://devfeed.tech/tags/operating-system.md>), [pass](<https://devfeed.tech/tags/pass.md>), [physics](<https://devfeed.tech/tags/physics.md>), [repositories](<https://devfeed.tech/tags/repositories.md>), [robot](<https://devfeed.tech/tags/robot.md>), [software](<https://devfeed.tech/tags/software.md>)

### AI overview

Robo-SWE provides 236 executable issue-resolution tasks mined from merged pull requests across 28 robotics repositories, covering ROS 2, simulation and physics engines, robotics algorithms, and robot-learning frameworks. Each task includes the original issue, a pre-fix container recipe, the developers' fix, and tests that repeatedly fail before and pass after the fix in offline runs. A provenance ledger records all 1,567 candidate pull requests and rejection reasons. In three runs per primary setup, reported resolve rates ranged from 45.8% for Claude Sonnet 5.5 to 56.5% for Claude Opus 5.5; GPT-6.1 Sol achieved 54.1% at maximum effort and 52.5% at medium effort, while GPT-6 Astra reached 51.4%. The tasks are execution-validated, but human review of evaluation fairness is future work.

### Source excerpt

arXiv:2611.16116v1 Announce Type: new Abstract: Coding agents are evaluated mostly on general-purpose software, but domain-specific systems require domain-aware evaluation. Robot software is a particularly challenging case: Robot Operating System 2 (ROS 2) packages have their own build tools, simulators and learning frameworks are built on physics engines, and behavior that depends on timing and hardware is hard to test. We present Robo-SWE, a dataset of 236 executable issue-resolution tasks mined from merged pull requests in 28 public robotics repositories in four domains: ROS 2 systems, simulation and physics engines, robotics algorithm libraries, and robot-learning frameworks. Each task contains the original issue text, a container recipe for the repository at the commit before the fix, the developers' fix, and the developers' tests, which fail before the fix and pass after it in repeated offline runs. The dataset also keeps a provenance ledger of all 1,567 candidate pull requests, with the stage each reached and the reason for every rejection, and, as a first use, 3,540 graded attempts by five coding-agent setups with recorded patches and 3,529 available trajectories. Across three runs per primary setup, resolve rates range from 45.8% for Claude Sonnet 5.5 to 56.5% for Claude Opus 5.5; GPT-6.1 Sol reaches 54.1% at maximum effort and 52.5% at medium effort, and GPT-6 Astra 51.4%. GLM-5.3 resolves 46.2%, and Kimi K3.1 Pro 43.9%. The tasks are validated by execution; verification of their fairness by human reviewers is future work. The dataset and public leaderboard are available online.

## AgriSense: Vision-Language Decision Support and Robotic Intervention in a Virtual Greenhouse Copy title

DevFeed: [AgriSense: Vision-Language Decision Support and Robotic Intervention in a Virtual Greenhouse Copy title](<https://devfeed.tech/articles/agrisense-vision-language-decision-support-and-robotic-intervention-in-a-virtual-greenhouse-copy-title-132282.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.16752>)

Author: Lyes Saad Saoud, Oualid Doukhi, Reza Ghorbani, Moussa Ayyash

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

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

Topics: [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>), [Isaac Sim](<https://devfeed.tech/topics/isaac-sim.md>), [vlm](<https://devfeed.tech/topics/vlm.md>)

Tags: [autonomous](<https://devfeed.tech/tags/autonomous.md>), [bert](<https://devfeed.tech/tags/bert.md>), [climate](<https://devfeed.tech/tags/climate.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>), [decision](<https://devfeed.tech/tags/decision.md>), [distribution](<https://devfeed.tech/tags/distribution.md>), [external](<https://devfeed.tech/tags/external.md>), [flag](<https://devfeed.tech/tags/flag.md>), [framework](<https://devfeed.tech/tags/framework.md>), [greenhouse](<https://devfeed.tech/tags/greenhouse.md>), [image](<https://devfeed.tech/tags/image.md>), [improvement](<https://devfeed.tech/tags/improvement.md>), [incremental](<https://devfeed.tech/tags/incremental.md>), [integrate](<https://devfeed.tech/tags/integrate.md>), [integration](<https://devfeed.tech/tags/integration.md>), [language](<https://devfeed.tech/tags/language.md>), [robotic](<https://devfeed.tech/tags/robotic.md>), [support](<https://devfeed.tech/tags/support.md>), [virtual](<https://devfeed.tech/tags/virtual.md>), [vision](<https://devfeed.tech/tags/vision.md>)

### AI overview

AgriSense evaluates a virtual-greenhouse framework that combines an Isaac Sim environment, a quadruped mobile manipulator, RGB and LiDAR-inertial sensing, climate control, and supervised vision-language decision support. Its model uses frozen DINOv2 and BERT encoders with environmental inputs to predict climate references, a target zone, and whether to intervene. On a source-grouped test set, full, image-removed, and language-removed models each correctly selected 43 of 44 zones, so the results show no added zone-accuracy benefit from combining image and language inputs; sensor-only inputs matched that accuracy, while language-only reached 68.64%. The full model's reported climate prediction errors were 0.249°C for temperature, 2.030 percentage points for humidity, and 0.997 points for soil moisture. Tests also found real-image distribution-shift limitations, accepted 32 of 40 virtual-arm targets, and measured 5.22-17.32 mm trajectory errors on five offline routes. Navigation and integrated treatment used simulator-state inputs; physical greenhouse performance and agronomic efficacy remain unvalidated.

### Source excerpt

arXiv:2611.16752v1 Announce Type: cross Abstract: Autonomous greenhouse management increasingly requires robotic systems that integrate perception, navigation, environmental regulation, and plant-level intervention. This paper presents AgriSense, a framework combining an Isaac Sim virtual greenhouse, a quadruped mobile manipulator, RGB and LiDAR-inertial sensing, environmental control, and supervised vision-language decision support. Frozen DINOv2 and BERT encoders, together with structured environmental inputs, support joint prediction of twelve environmental references, a target zone, and an intervention flag. Checkpoint replay reproduces 41/43 zone decisions. Across five seeds on a source-grouped test set, the full, image-removed, and language-removed configurations each achieve 43/44 correct zone decisions, providing no evidence of incremental multimodal zone-accuracy improvement. The full model achieves mean absolute errors of 0.249 degrees C for temperature, 2.030 percentage points for humidity, and 0.997 percentage points for soil moisture. Sensor-only inputs match full-model zone accuracy, while language-only inputs achieve 68.64%. External real-image benchmarks reveal distribution-shift limitations. Five offline Point-LIO routes yield rigidly aligned trajectory errors of 5.22-17.32 mm. Of 40 virtual-arm targets, 32 satisfy geometric acceptance criteria and eight are rejected at boundaries. Six long-horizon climate cases and fault protocols characterize integration limits. Navigation and integrated treatment rely on simulator-state inputs; physical greenhouse performance and agronomic treatment efficacy remain unvalidated. The contribution is an evaluated decision-support and robotic-intervention framework for a virtual greenhouse.

## Qvyon launches with $8M to improve safety for physical AI deployments

DevFeed: [Qvyon launches with $8M to improve safety for physical AI deployments](<https://devfeed.tech/articles/qvyon-launches-with-8m-to-improve-safety-for-physical-ai-deployments-130455.md>)

Original publisher: [Read original article](<https://siliconangle.com/2026/11/12/qvyon-launches-with-8m-to-improve-safety-for-physical-ai-deployments/>)

Author: Kyt Dotson

Published: 2026-11-12T14:00:18Z

Content type: news

Language: en

Sources: [SiliconANGLE](<https://devfeed.tech/sources/siliconangle.md>)

Topics: [Physical AI](<https://devfeed.tech/topics/physical-ai.md>), [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>), [Robotics](<https://devfeed.tech/topics/robotics.md>), [autonomous vehicles](<https://devfeed.tech/topics/autonomous-vehicles.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [3D](<https://devfeed.tech/topics/3d.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-startup-funding](<https://devfeed.tech/tags/ai-startup-funding.md>), [ai-validation](<https://devfeed.tech/tags/ai-validation.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [autonomous-vehicles](<https://devfeed.tech/tags/autonomous-vehicles.md>), [deployments](<https://devfeed.tech/tags/deployments.md>), [frantz-saintellemy](<https://devfeed.tech/tags/frantz-saintellemy.md>), [funding](<https://devfeed.tech/tags/funding.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [launches](<https://devfeed.tech/tags/launches.md>), [news](<https://devfeed.tech/tags/news.md>), [physical-ai](<https://devfeed.tech/tags/physical-ai.md>), [qvyon-technologies](<https://devfeed.tech/tags/qvyon-technologies.md>), [robot-simulations](<https://devfeed.tech/tags/robot-simulations.md>), [robots](<https://devfeed.tech/tags/robots.md>), [safety](<https://devfeed.tech/tags/safety.md>), [software](<https://devfeed.tech/tags/software.md>), [the-latest](<https://devfeed.tech/tags/the-latest.md>), [validation](<https://devfeed.tech/tags/validation.md>), [venture-capital](<https://devfeed.tech/tags/venture-capital.md>)

### AI overview

Qvyon launched with $8 million in funding and Mimic Core, which turns customer recordings of camera, lidar, and radar data into 3D, time-based simulated environments and varied scenarios for testing physical AI before deployment. Engineers can reuse one recording to test how vehicles or robots respond to challenging conditions, producing traceable datasets and validation evidence without repeatedly gathering field data. Mimic Core works with existing virtual testing tools and is offered through Qvyon Cloud or as self-hosted software. Qvyon says cloud customer data is kept separate and is not used to train models for other customers. The article says a 14-day guided evaluation is available; planned products include Mimic Validation OS in the first half of 2027 and Mimic Learn in late 2027.

### Source excerpt

Physical AI validation infrastructure and software development company Qvyon Technologies Inc. launched today with $8 million in total funding to help companies validate autonomous vehicles and robots before deployment. MaC Venture Capital led the funding, with participation from BDC Capital, BKR Capital and Spotlight VP. Physical AI is artificial intelligence that moves from digital brains [...] The post Qvyon launches with $8M to improve safety for physical AI deployments appeared first on SiliconANGLE.

## Omnimodal World Models Teach Robots to See, Hear, Read and Act in One System

DevFeed: [Omnimodal World Models Teach Robots to See, Hear, Read and Act in One System](<https://devfeed.tech/articles/omnimodal-world-models-teach-robots-to-see-hear-read-and-act-in-one-system-129454.md>)

Original publisher: [Read original article](<https://hackernoon.com/omnimodal-world-models-teach-robots-to-see-hear-read-and-act-in-one-system?source=rss>)

Author: NVIDIA CORPORATION

Published: 2026-11-11T12:30:02Z

Content type: article

Language: en

Sources: [HackerNoon](<https://devfeed.tech/sources/hackernoon.md>)

Topics: [Cosmos](<https://devfeed.tech/topics/cosmos.md>), [World models](<https://devfeed.tech/topics/world-models.md>), [Physical AI](<https://devfeed.tech/topics/physical-ai.md>), [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>), [Robot Manipulation](<https://devfeed.tech/topics/robot-manipulation.md>), [post-training](<https://devfeed.tech/topics/post-training.md>), [generative-ai-tools](<https://devfeed.tech/topics/generative-ai-tools.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>), [Image](<https://devfeed.tech/topics/image.md>), [Nemotron](<https://devfeed.tech/topics/nemotron.md>)

Tags: [3](<https://devfeed.tech/tags/3.md>), [act](<https://devfeed.tech/tags/act.md>), [ai](<https://devfeed.tech/tags/ai.md>), [audio](<https://devfeed.tech/tags/audio.md>), [cosmos](<https://devfeed.tech/tags/cosmos.md>), [cosmos-3](<https://devfeed.tech/tags/cosmos-3.md>), [image](<https://devfeed.tech/tags/image.md>), [language](<https://devfeed.tech/tags/language.md>), [model](<https://devfeed.tech/tags/model.md>), [models](<https://devfeed.tech/tags/models.md>), [multimodal-ai](<https://devfeed.tech/tags/multimodal-ai.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [omnimodal-model](<https://devfeed.tech/tags/omnimodal-model.md>), [omnimodal-world-model](<https://devfeed.tech/tags/omnimodal-world-model.md>), [one](<https://devfeed.tech/tags/one.md>), [open](<https://devfeed.tech/tags/open.md>), [open-source-ai](<https://devfeed.tech/tags/open-source-ai.md>), [physical-ai](<https://devfeed.tech/tags/physical-ai.md>), [robotics](<https://devfeed.tech/tags/robotics.md>), [robots](<https://devfeed.tech/tags/robots.md>), [tech-robots](<https://devfeed.tech/tags/tech-robots.md>), [video](<https://devfeed.tech/tags/video.md>), [world](<https://devfeed.tech/tags/world.md>), [world-model](<https://devfeed.tech/tags/world-model.md>), [world-models](<https://devfeed.tech/tags/world-models.md>)

### AI overview

Cosmos 3 combines multimodal understanding, generation, simulation and action prediction in a shared Mixture-of-Transformers framework intended as a backbone for Physical AI. It processes language, images, video, audio and action; depending on the input and output configuration, it can perform vision-language reasoning, generate images or video with optional audio, and predict actions alongside visual consequences. Separate autoregressive and diffusion pathways handle reasoning and generation, while a shared attention mechanism connects them. The report describes Edge, Nano and Super model scales; Nano and Super are released, while Edge is planned for a later release. NVIDIA says its evaluations place the system at or above specialized open-source baselines on most capabilities, with particular rankings attributed to external benchmarks at the time of writing. Code, checkpoints, synthetic datasets and an evaluation benchmark are offered under the OpenMDW-1.1 License. The excerpt presents the system as a foundation for downstream specialization, not evidence that robots trained with it already operate reliably in real-world settings.

### Source excerpt

NVIDIA's Cosmos 3 unifies language, image, video, audio and action in one open world model built as a backbone for Physical AI and robotics.

## Industrial automation startup Path Robotics raises $300M in funding

DevFeed: [Industrial automation startup Path Robotics raises $300M in funding](<https://devfeed.tech/articles/industrial-automation-startup-path-robotics-raises-300m-in-funding-127692.md>)

Original publisher: [Read original article](<https://siliconangle.com/2026/11/10/industrial-automation-startup-path-robotics-raises-300m-in-funding/>)

Author: Maria Deutscher

Published: 2026-11-11T01:51:39Z

Content type: news

Language: en

Sources: [SiliconANGLE](<https://devfeed.tech/sources/siliconangle.md>)

Topics: [Automation](<https://devfeed.tech/topics/automation.md>), [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-model](<https://devfeed.tech/tags/ai-model.md>), [automate](<https://devfeed.tech/tags/automate.md>), [automation](<https://devfeed.tech/tags/automation.md>), [emerging-tech](<https://devfeed.tech/tags/emerging-tech.md>), [funding](<https://devfeed.tech/tags/funding.md>), [funding-round](<https://devfeed.tech/tags/funding-round.md>), [industrial](<https://devfeed.tech/tags/industrial.md>), [industrial-automation](<https://devfeed.tech/tags/industrial-automation.md>), [news](<https://devfeed.tech/tags/news.md>), [obsidian](<https://devfeed.tech/tags/obsidian.md>), [path-robotics](<https://devfeed.tech/tags/path-robotics.md>), [raises](<https://devfeed.tech/tags/raises.md>), [robotics](<https://devfeed.tech/tags/robotics.md>), [rove](<https://devfeed.tech/tags/rove.md>), [series](<https://devfeed.tech/tags/series.md>), [startup](<https://devfeed.tech/tags/startup.md>), [the-latest](<https://devfeed.tech/tags/the-latest.md>), [welding](<https://devfeed.tech/tags/welding.md>), [workflows](<https://devfeed.tech/tags/workflows.md>)

### AI overview

Path Robotics raised $300 million in a Series E led by Georgian, valuing the industrial welding automation startup at $2.45 billion. Its Obsidian model is designed to adapt robotic welding to heat-driven metal distortion and variations between parts. Four lasers on the torch illuminate the work; two cameras capture reflected light, and data gathered before, during, and after each run informs a "weld world model" used to generate simulated projects for training. The company first developed Obsidian for welding cells and says it also powers Rove, a four-legged robot for shipbuilding. Path says its software can support manufacturing for vehicles, generators, utility poles, and other heavy equipment.

### Source excerpt

Path Robotics Inc., a startup working to automate industrial welding workflows, today announced that it has raised $300 million in funding. Georgian led the Series E round. It was joined by a consortium that included the venture capital arms of Yamaha Motor Co. and Navier Corp., one of South Korea's largest tech firms. The investment [...] The post Industrial automation startup Path Robotics raises $300M in funding appeared first on SiliconANGLE.

## Look Ahead: Search-Driven Reasoning in Embodied AI

DevFeed: [Look Ahead: Search-Driven Reasoning in Embodied AI](<https://devfeed.tech/articles/look-ahead-search-driven-reasoning-in-embodied-ai-126009.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.10769>)

Author: Ahmed Zeer, Osama Shbib, Muzaffer Kaan Yuce, Himmet Toprak Kesgin, Mehmet Fatih Amasyali

Published: 2026-11-10T05:00:00Z

Content type: article

Language: en

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

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Physical AI](<https://devfeed.tech/topics/physical-ai.md>), [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>), [Machine Intelligence](<https://devfeed.tech/topics/machine-intelligence.md>)

Tags: [act](<https://devfeed.tech/tags/act.md>), [ai](<https://devfeed.tech/tags/ai.md>), [beam](<https://devfeed.tech/tags/beam.md>), [coverage](<https://devfeed.tech/tags/coverage.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [embodied-ai](<https://devfeed.tech/tags/embodied-ai.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [performance](<https://devfeed.tech/tags/performance.md>), [planning](<https://devfeed.tech/tags/planning.md>), [policy](<https://devfeed.tech/tags/policy.md>), [reasoning](<https://devfeed.tech/tags/reasoning.md>), [reward](<https://devfeed.tech/tags/reward.md>), [runtime](<https://devfeed.tech/tags/runtime.md>), [search](<https://devfeed.tech/tags/search.md>), [simulator](<https://devfeed.tech/tags/simulator.md>), [test](<https://devfeed.tech/tags/test.md>), [transformers](<https://devfeed.tech/tags/transformers.md>)

### AI overview

In PushT simulation with a frozen ACT policy, beam search over perturbed action chunks improved success across 50 episodes from 0% to 18% and average cumulative reward from 39.44 to 103.2. The proposed One Step Ahead gate simulates the nominal chunk and triggers full search when predicted coverage drops beyond a threshold; reported ablations show it preserves or improves success while cutting runtime by about 3x versus full beam search. Larger search budgets did not consistently improve performance. The evidence is limited to this setup; language-conditioned VLAs and physical robots remain unevaluated.

### Source excerpt

arXiv:2611.10769v1 Announce Type: new Abstract: Search allows a robotic policy to evaluate candidate futures before committing to an action. Motivated by test-time planning for vision-language action (VLA) models, we study selective search over action chunks in PushT using a frozen Action Chunking with Transformers (ACT) policy. Beam search generates perturbed alternatives around the policy prediction and scores them through rollouts from cloned simulator states. Across 50 evaluation episodes, beam search raises task success from 0% to 18% and increases average cumulative reward from 39.44 to 103.2. We introduce One Step Ahead (OSA), a gate that simulates the nominal policy chunk and invokes full search only when its predicted coverage drop exceeds a threshold. In the reported ablations, OSA preserves or improves success while reducing runtime by approximately 3x relative to full beam search. Ablations over tree depth, branching factor, beam width, and replanning frequency show that larger search budgets do not consistently improve performance. The results support selective test-time search for action-chunking policies; evaluation on language-conditioned VLAs and physical robots remains future work.

## EmbodiedIntentBench: Benchmarking Intent-Driven Runtime Adaptation for Mobile Robotic Missions

DevFeed: [EmbodiedIntentBench: Benchmarking Intent-Driven Runtime Adaptation for Mobile Robotic Missions](<https://devfeed.tech/articles/embodiedintentbench-benchmarking-intent-driven-runtime-adaptation-for-mobile-robotic-missions-125205.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.10498>)

Author: Rajat Bhattacharjya, Changwoon Lee, Hyeonjong Ju, Sing-Yao Wu, Hamidreza Alikhani, Zhuoyi Yang, Ian Harris, Nalini Venkatasubramanian, Nikil Dutt

Published: 2026-11-10T05:00:00Z

Content type: article

Language: en

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

Topics: [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>), [Jetson AGX Thor](<https://devfeed.tech/topics/jetson-agx-thor.md>)

Tags: [ablation](<https://devfeed.tech/tags/ablation.md>), [adaptive](<https://devfeed.tech/tags/adaptive.md>), [battery](<https://devfeed.tech/tags/battery.md>), [benchmarking](<https://devfeed.tech/tags/benchmarking.md>), [connectivity](<https://devfeed.tech/tags/connectivity.md>), [contracts](<https://devfeed.tech/tags/contracts.md>), [coverage](<https://devfeed.tech/tags/coverage.md>), [cpu](<https://devfeed.tech/tags/cpu.md>), [cs-ar](<https://devfeed.tech/tags/cs-ar.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-dc](<https://devfeed.tech/tags/cs-dc.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [energy](<https://devfeed.tech/tags/energy.md>), [execution](<https://devfeed.tech/tags/execution.md>), [fixed](<https://devfeed.tech/tags/fixed.md>), [framework](<https://devfeed.tech/tags/framework.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [jetson](<https://devfeed.tech/tags/jetson.md>), [local](<https://devfeed.tech/tags/local.md>), [mission](<https://devfeed.tech/tags/mission.md>), [mobile](<https://devfeed.tech/tags/mobile.md>), [motion](<https://devfeed.tech/tags/motion.md>), [robotic](<https://devfeed.tech/tags/robotic.md>), [runtime](<https://devfeed.tech/tags/runtime.md>)

### AI overview

EmbodiedIntentBench evaluates whether runtime adaptation fulfills mobile robot mission objectives under physical and resource constraints, rather than judging perception accuracy or execution efficiency alone. It encodes intent as contracts covering coverage, temporal evidence, reporting, urgency, and resources, then tests fixed and adaptive strategies in simulated UAV missions with causal, resource-accounted execution. In the multi-phase Depot study, a configuration with 87.5% target recall across visible observations completed 1 of 16 missions, while a lower-recall configuration completed 14. Ignoring execution-induced observation loss raised aggregate success and reversed two strategies' ranking. Jetson AGX Orin CPU and GPU profile replays also changed the relative success of local and remote strategies. In selected GPU-profile replays, intent-driven adaptation reduced modeled task energy by up to 28% without changing mission success.

### Source excerpt

arXiv:2611.10498v1 Announce Type: cross Abstract: Mobile robotic sensing systems must adapt to operator requirements, limited resources, changing connectivity, and observation opportunities shaped by physical motion. Evaluating runtime strategies therefore requires assessing whether their decisions fulfill mission objectives, beyond isolated perception accuracy or execution efficiency. We present EmbodiedIntentBench, a benchmark for intent-driven runtime adaptation in mobile robotic missions. It represents mission intent through structured contracts for coverage, temporal evidence, reporting, urgency, and resources, and couples simulated unmanned aerial vehicle (UAV) workloads with causal, resource-accounted execution. Computation decisions affect mission time, resource use, and which observations can contribute evidence. We evaluate fixed and adaptive perception strategies in multi-phase missions and under changing urgency, connectivity, and battery conditions. In the multi-phase Depot study, a configuration with 87.5% full-bank target recall across visible target observations completes only 1 of 16 missions, whereas a lower-recall configuration completes 14. In an ablation, ignoring execution-induced observation loss increases aggregate mission success and reverses the ordering of two runtime strategies. Replays using Jetson AGX Orin CPU and GPU profiles change the relative success of local and remote strategies. In selected Jetson GPU-profile replays, intent-driven adaptation reduces modeled task energy by up to 28% with unchanged mission success. EmbodiedIntentBench provides an extensible framework for evaluating runtime decisions against mission fulfillment under physical and resource constraints.

## Accelerated Reinforcement Learning for Soft Robotics: Hours Instead of Years

DevFeed: [Accelerated Reinforcement Learning for Soft Robotics: Hours Instead of Years](<https://devfeed.tech/articles/accelerated-reinforcement-learning-for-soft-robotics-hours-instead-of-years-125440.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.11718>)

Author: Roshan S. Kaundinya, Hehui Zheng, Robert K. Katzschmann, George Haller

Published: 2026-11-10T05:00:00Z

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: [Reinforcement learning](<https://devfeed.tech/topics/reinforcement-learning.md>), [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>), [Neural Network](<https://devfeed.tech/topics/neural-network.md>), [Transformers](<https://devfeed.tech/topics/transformers.md>)

Tags: [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>), [learning](<https://devfeed.tech/tags/learning.md>), [math-ds](<https://devfeed.tech/tags/math-ds.md>), [model](<https://devfeed.tech/tags/model.md>), [reinforcement-learning](<https://devfeed.tech/tags/reinforcement-learning.md>), [robotics](<https://devfeed.tech/tags/robotics.md>), [soft-robotics](<https://devfeed.tech/tags/soft-robotics.md>), [tracking](<https://devfeed.tech/tags/tracking.md>)

### AI overview

The authors report that a six-dimensional adiabatic spectral submanifold (aSSM) model can make reinforcement-learning training for a pneumatically actuated soft arm much faster. They learn a 6,924-parameter polynomial model from pre-collected data in five minutes, then use it as the training environment: a trajectory-tracking policy trains in three hours and deploys without fine-tuning. The abstract estimates that reproducing this training through hardware interaction would take three years. Against otherwise identical policies trained using a 4.8-million-parameter RNN or 1.7-million-parameter transformer, the aSSM-trained policy has 40% lower tracking error; identifying those models takes 13 and 100 hours, respectively. These are reported results for the described soft arm and task.

### Source excerpt

arXiv:2611.11718v1 Announce Type: cross Abstract: Reinforcement learning (RL) requires large amounts of interaction data or a reliable simulator, both hard to obtain for nonlinear soft robotic hardware. Recent work shows that a time-scale separation inherent to soft robot dynamics justifies their reduction to low-dimensional invariant manifolds, known as adiabatic spectral submanifolds (aSSMs). We demonstrate that an aSSM-reduced model, a simple polynomial ordinary differential equation learned purely from pre-collected data, can serve as an RL training environment. On a pneumatically actuated soft arm, we train a trajectory-tracking policy in 3 hours on a six-dimensional aSSM-reduced model and deploy it without fine-tuning, whereas reproducing this policy training extrapolates to 3 years of hardware interaction. Our policy tracks dynamic targets with 40% lower error than otherwise identical policies trained on a 4.8-million-parameter recurrent neural network (RNN) or a 1.7-million-parameter transformer, which take 13 and 100 hours to identify, against 5 minutes for our mathematically justified 6,924-parameter model.

## Same Physics, Different Gradients: A State-of-the-Art Report on Differentiable Physics Simulation

DevFeed: [Same Physics, Different Gradients: A State-of-the-Art Report on Differentiable Physics Simulation](<https://devfeed.tech/articles/same-physics-different-gradients-a-state-of-the-art-report-on-differentiable-physics-simulation-125524.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.12162>)

Author: Lei Shu, Zizhou Huang, Yunuo Chen, Yifei Li, Tao Du, Minchen Li, Tuur Stuyck, Miles Macklin, Kui Wu, Bo Zhu, Yin Yang, Ming C. Lin, Alberta Longhini, Rika Antonova, Leonidas Guibas, Gordon Wetzstein,

Published: 2026-11-10T05:00:00Z

Content type: article

Language: en

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

Topics: [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>)

Tags: [animation](<https://devfeed.tech/tags/animation.md>), [computer-science](<https://devfeed.tech/tags/computer-science.md>), [continuous](<https://devfeed.tech/tags/continuous.md>), [control](<https://devfeed.tech/tags/control.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-gr](<https://devfeed.tech/tags/cs-gr.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-na](<https://devfeed.tech/tags/cs-na.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [design](<https://devfeed.tech/tags/design.md>), [engine](<https://devfeed.tech/tags/engine.md>), [gradient](<https://devfeed.tech/tags/gradient.md>), [graphics](<https://devfeed.tech/tags/graphics.md>), [math-na](<https://devfeed.tech/tags/math-na.md>), [memory](<https://devfeed.tech/tags/memory.md>), [model](<https://devfeed.tech/tags/model.md>), [optimization](<https://devfeed.tech/tags/optimization.md>), [physics](<https://devfeed.tech/tags/physics.md>), [reporting](<https://devfeed.tech/tags/reporting.md>), [research](<https://devfeed.tech/tags/research.md>), [simulation](<https://devfeed.tech/tags/simulation.md>), [simulator](<https://devfeed.tech/tags/simulator.md>), [taxonomy](<https://devfeed.tech/tags/taxonomy.md>), [techniques](<https://devfeed.tech/tags/techniques.md>), [testbed](<https://devfeed.tech/tags/testbed.md>)

### AI overview

The report distinguishes differentiable physics methods by what their gradients refer to--the executed finite computation, a converged equation, or continuous dynamics--and by how the backward pass is implemented: automatic differentiation, an equation-level adjoint, or a custom adjoint. It argues that exactness, approximation accuracy, and memory cost depend on different parts of this design, so the apparent exactness-memory trade-off is not fundamental. A separate taxonomy addresses non-smooth contact. On one testbed, the advantage of exact derivatives over equation-level adjoints depends on the source of the fitting target as well as solver budget; the report also reviews regimes where zeroth-order methods can outperform simulator gradients. It concludes with a reporting checklist and diagnostic tool. This overview is based on the supplied abstract.

### Source excerpt

arXiv:2611.12162v1 Announce Type: cross Abstract: Differentiable simulation propagates derivatives through a physics simulator, turning animation, system identification, control, and computational design into gradient-based optimization. Engines differ in the physics they model and, less visibly, in which mathematical object they differentiate. We separate two questions the literature conflates. The referent is which object the returned derivative belongs to: the finite computation that executed, the converged equation it approximates, or the continuous dynamics. The realization describes the organization of the backward pass: automatically generated differentiation, an equation-level adjoint solve, or a custom adjoint. The referent defines the target against which exactness is assessed; the implementation determines how accurately that target is approximated. Cost depends on the realization together with how the states needed by the backward pass are retained or recovered. Crossing the axes places every method and engine we classify in a cell, separates claims the literature merges, and shows that the apparent trade-off between exactness and memory follows from how methods have been built rather than from a constraint. A second, orthogonal taxonomy covers contact, where the map is non-smooth. We then ask whether a correct gradient helps. Four mechanisms by which useful gradients are lost are cataloged and measured on one testbed, where the advantage of the exact derivative over the equation-level adjoint depends on where the fitting target comes from as much as on the solver budget, and we review the evidence that zeroth-order methods can outperform simulator gradients in the regimes where those mechanisms act. The model classes, systems techniques and applications are surveyed through this lens, and we close with a reporting checklist, a released diagnostic that computes it, and a research agenda.

## Gemini Robotics 2.5 accelerates real-world general dexterity

DevFeed: [Gemini Robotics 2.5 accelerates real-world general dexterity](<https://devfeed.tech/articles/gemini-robotics-2-5-accelerates-real-world-general-dexterity-128944.md>)

Original publisher: [Read original article](<https://deepmind.google/blog/gemini-robotics-2-5-accelerates-real-world-physical-intelligence/>)

Author: Carolina Parada

Published: 2026-11-09T19:51:47Z

Content type: release

Language: en

Sources: [Google DeepMind News](<https://devfeed.tech/sources/google-deepmind-news.md>)

Topics: [Robot Manipulation](<https://devfeed.tech/topics/robot-manipulation.md>), [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>), [Physical AI](<https://devfeed.tech/topics/physical-ai.md>)

Tags: [accelerated](<https://devfeed.tech/tags/accelerated.md>), [agi](<https://devfeed.tech/tags/agi.md>), [agile](<https://devfeed.tech/tags/agile.md>), [argon](<https://devfeed.tech/tags/argon.md>), [dexterity](<https://devfeed.tech/tags/dexterity.md>), [digital-twin](<https://devfeed.tech/tags/digital-twin.md>), [failures](<https://devfeed.tech/tags/failures.md>), [gemini](<https://devfeed.tech/tags/gemini.md>), [gemini-4](<https://devfeed.tech/tags/gemini-4.md>), [gemini-4-argon](<https://devfeed.tech/tags/gemini-4-argon.md>), [gemini-robotics-2-5](<https://devfeed.tech/tags/gemini-robotics-2-5.md>), [google](<https://devfeed.tech/tags/google.md>), [model](<https://devfeed.tech/tags/model.md>), [models](<https://devfeed.tech/tags/models.md>), [real-world](<https://devfeed.tech/tags/real-world.md>), [reliability](<https://devfeed.tech/tags/reliability.md>), [robot](<https://devfeed.tech/tags/robot.md>), [robot-dexterity](<https://devfeed.tech/tags/robot-dexterity.md>), [robotics](<https://devfeed.tech/tags/robotics.md>), [safety](<https://devfeed.tech/tags/safety.md>), [vision-language-action](<https://devfeed.tech/tags/vision-language-action.md>)

### AI overview

Argon can use a real-world camera scene to generate varied simulated practice runs and build a digital twin. Combined with real training demonstrations, this is described as helping robots generalize more quickly to new scenes and objects. Researchers can also use video footage to hypothesize causes of task failures, then test improvements in simulation and on real robots. The excerpt provides no specific performance measurements or details about safety methods.

### Source excerpt

Our most capable vision-language-action model enhances robot dexterity, reliability, and safety

## From Simulation to Reality: Why Your Robot Works Perfectly in the Lab But Fails in Real Life

DevFeed: [From Simulation to Reality: Why Your Robot Works Perfectly in the Lab But Fails in Real Life](<https://devfeed.tech/articles/from-simulation-to-reality-why-your-robot-works-perfectly-in-the-lab-but-fails-in-real-life-124338.md>)

Original publisher: [Read original article](<https://wf.coreweave.com/blog/from-simulation-to-reality-why-your-robot-works-perfectly-in-the-lab-but-fails-in-real-life>)

Author: CoreWeave

Published: 2026-11-09T16:03:04Z

Content type: tutorial

Language: en

Sources: [Coreweave Blog Posts](<https://devfeed.tech/sources/coreweave-blog-posts.md>)

Topics: [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>), [Reinforcement learning](<https://devfeed.tech/topics/reinforcement-learning.md>), [Physical AI](<https://devfeed.tech/topics/physical-ai.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Tutorial](<https://devfeed.tech/topics/tutorial.md>)

Tags: [blog](<https://devfeed.tech/tags/blog.md>), [domain](<https://devfeed.tech/tags/domain.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [reinforcement-learning](<https://devfeed.tech/tags/reinforcement-learning.md>), [robot](<https://devfeed.tech/tags/robot.md>), [robotics](<https://devfeed.tech/tags/robotics.md>), [simulation](<https://devfeed.tech/tags/simulation.md>)

### AI overview

The article argues that high simulation reward can hide policies that fail on real hardware, such as a simulated walker that earns reward by vibrating its legs until motors overheat. It describes training across randomized gravity, friction, motor power, and sensor noise to improve robustness, then tracking hardware-relevant measures such as action smoothness and torque alongside reward. In the reported BipedalWalker experiments, domain randomization raised reward from 211 to 247 but worsened smoothness from 0.615 to 0.714; a 0.05 smoothness penalty reduced jitter to 0.535 while reward fell to 59. The article presents this as a safety/performance tradeoff and recommends validating behavior with metrics and videos, applying task-specific safety penalties, and taking precautions such as reduced motor power, an emergency stop, and temperature monitoring before hardware deployment. It reports that randomization takes 2-3 times longer and slightly lowers ideal-condition performance, while claiming only a 20% performance drop under real-world variation versus 80% or more without it.

### Source excerpt

Explore the simulation-to-reality problem in robotics and learn how domain randomization, system identification, reinforcement learning, and evaluation can help policies transfer to the real-world.

[Next page](<https://devfeed.tech/topics/robotics-simulation.md?cursor=WyIyMDI2LTExLTA5VDE2OjAzOjA0KzAwOjAwIiwgIjhmZTBhOTBlLWEyYjgtNDkzNC05Y2U2LTMzMjEyNjI1MjgzNSJd>)