Skip to content
#Safety and reliability
Oct 27Tue
  1. arXiv cs.RO53

    SHAPE: An Energy-Based Fault Detection and Isolation Framework for Flat Port-Hamiltonian Mechanical Systems

    The paper proposes an energy-based fault diagnosis framework, SHAPE, which combines port-Hamiltonian variables and differential flatness to generate nominal power references from planned motion, constructing mechanical port-power residuals without a dedicated dynamic observer. In simulations of a six-degree-of-freedom Kinova manipulator under model predictive control, SHAPE-CWT achieved an average accuracy of 84.69% on 3000 held-out tests, compared with 81.67% for torque residuals and 97.17% for a generalized momentum observer. The authors note that severity-excitation profiles and offline velocity perturbations limit the method's diagnostic capability.

  2. arXiv cs.RO58

    MDPO: A Multi-Constraint Policy Optimization Reinforcement Learning Method for Dexterous Manipulation

    The authors propose Multi-constraint Dexterous Policy Optimization (MDPO), which formulates dexterous manipulation as a multi-constraint policy optimization problem involving both cumulative cost constraints and state-wise cost constraints, and integrates multi-constraint projection into the policy optimization process. In experiments on ReDMan for safe joint control, finger safety, and grasping tasks, MDPO achieved better constraint satisfaction than baselines while maintaining competitive task performance. The code is publicly available.

  3. arXiv cs.RO56

    SCET-CBF: A Speed-Calibrated Soft Guidance Method for One-Step Dynamic Navigation

    Xiaoyu Han of University College London and Dongkun Han of the Chinese University of Hong Kong proposed SCET-CBF, a speed-calibrated soft guidance law for one-step dynamic navigation. It uses a short lateral escape maneuver to provide an allowable approach speed, and lets a quadratic program weigh the two effects of braking when choosing acceleration and steering.

  4. arXiv cs.RO66

    Active Reachability Learning for Humanoid Recoverability Monitoring

    The paper proposes an active reachability learning framework for learning the stochastic finite-horizon recoverability envelope of a given fallback policy, so that fallback can be triggered before an unrecoverable fall occurs. The method reframes data exploration as a reach-avoid problem, in which an uncertainty-guided explorer discovers unseen states while avoiding confidently unrecoverable regions, and combines temporal-difference and Monte Carlo objectives to improve in-distribution value prediction and uncertainty estimation. The authors evaluate on stochastic benchmarks and a 29-DoF Unitree G1 humanoid robot, reporting clear improvements in monitoring coverage and accuracy, and zero-shot transfer to real hardware.

  5. arXiv cs.RO70

    Neuromorphic Decision Dynamics: Deadlock-free Safe Navigation in Congested Environments

    The authors propose Neuromorphic Decision Dynamics (NDD), a decentralized, event-driven controller for safe navigation in congested environments, with guarantees on safety, liveness, robustness, and scalability. NDD requires no communication, preassigned priorities, or models of neighbors' objectives; by coupling internal decision and attention states to physical states and the relative motion observed locally, it enables agents to self-organize into local conflict groups and form a passage order.

  6. arXiv cs.RO34

    Causal-based approaches to explain and learn from self-extension: A review

    This review surveys recent advances in aware, self-evolving robots, with the goal of enabling robots to give humans explanations for the reasons behind their actions, plan failures, or perceptual anomalies. The authors note that deep learning alone cannot close this gap, because the reasons behind the behavior of black-box models are not accessible; an underlying architecture based on deeper reasoning is needed. When explanations arise from cognitive processes and are shared with and validated by humans, the robot should be able to incorporate them into its knowledge base to improve subsequent behavior.

  7. arXiv cs.RO60

    Interpretable Collision Avoidance and Feasibility Guarantees via an Uncertainty-Aware Pipeline

    The study proposes a modular perception-prediction-planning pipeline that propagates uncertainty as structured, interpretable quantities across modules and guarantees a feasible solution in every cycle within convex safe sets. In Monte Carlo trials at a four-way unsignalized intersection, the pipeline achieved 0% collisions and 0% solver infeasibility, with 12.7 ms per cycle; the deterministic planner and chance-constrained baseline had collision rates of 36% and 10%, respectively. Ablation experiments confirmed the necessity of each uncertainty source, and a hardware deployment on a Clearpath Jackal UGV verified real-time feasibility.

  8. OFweek Robotics58

    Shenmou Intelligent's "Dragon Claw" drone wins State Grid power aerial operations project

    Shenmou Intelligent's "Dragon Claw" aerial-operations drone has won State Grid Power's "High-Stability UAV for Aerial Operations" project, using a drone with a robotic arm and dexterous hand to clear foreign-object hazards on overhead transmission lines. The system identifies foreign objects such as bird nests through inspection, then uses the robotic arm and dexterous hand to handle and verify them, integrating detection, handling, and verification into one workflow. Shenmou says its core technologies include a self-developed flight control system, robotic arm and dexterous-hand grasping control, dynamic visual guidance, and piezoelectric six-axis force sensors, and cites research estimating annual procurement demand of about 150 million to 200 million yuan per province.

  9. Ars Technica55

    Over 300 Boston Dynamics employees petition against selling robot dogs to ICE; company insists on honoring the deal

    More than 300 Boston Dynamics employees signed a petition demanding that executives stop selling robot dogs to U.S. Immigration and Customs Enforcement (ICE), while company management insists on the deal. On September 25, the U.S. Department of Homeland Security awarded Radeco a $1.3 million contract to supply ICE with 4 modified Spot units named Cerberus, equipped with oxygen sensors, 30x zoom cameras, 360-degree cameras, infrared thermal imaging, and an optional robotic arm, for detecting radiation, toxic gas leaks, and explosive threats.

Oct 26Mon
  1. Robohub27

    September 2026 robotics society, business, and culture roundup: Global industrial robot stock tops 5 million units, Toyota plans to invest 1 trillion yen, ETH Zurich achieves humanoid hand walking

    The International Federation of Robotics' "World Robotics 2026" report shows that more than 5 million industrial robots are now in operation in factories worldwide; in 2025, more than 600,000 units were newly installed, up 11% year on year, with China accounting for 354,000 units, or 59% of the global total.

  2. 雷科技71

    Agibot launches safety-standard heavy-duty embodied robot Genie G2 Max with 40 kg dual-arm rated payload

    Agibot launched the safety-standard heavy-duty embodied robot Genie G2 Max on October 26, targeting depalletizing and palletizing, material transfer, and production-line feeding. It has a 20 kg single-arm payload, a 40 kg dual-arm rated payload, a 50 kg peak payload, and repeat positioning accuracy within ±0.5 mm. The robot uses 7-DOF force-controlled dual arms, has a vertical working range of 0.1 to 2.5 meters, features dual-battery hot swapping and autonomous recharging, and is compatible with the Genie Studio 2.0 embodied work platform. Agibot disclosed that in e-commerce warehousing scenarios its overall task success rate exceeds 99%, with 80 boxes depalletized and 75 boxes palletized per hour.

  3. OFweek Robotics62

    Underwater robot orders surge, but why is only half the business case settled?

    The underwater robot sector saw a concentration of financing and orders in 2026: Shihang Intelligent's orders for the year exceeded 1 billion yuan; Shenhai Zhiren signed 8 work-class ROVs in South America and Southeast Asia with total contracts of 300 million yuan; and Ouka Zhibo won two consecutive water-area cleaning projects totaling about 27.17 million yuan. The article uses Shenhai Zhiren, Weilan Shenhai, and Ouka Zhibo to represent three hurdles—selling equipment, completing tasks, and replicating platforms—and notes that scarce underwater data, unproven stable profits, and a shortage of onshore engineering talent remain the parts of the industry that have not been settled.

  4. Robotics & Automation News58

    beRobox launches 3D Dynamic Safety module for PALTZ palletizing robots

    beRobox has launched the 3D Dynamic Safety module for its PALTZ palletizing robots, mounting the detection device on the robotic arm so the protective zone adjusts dynamically with the robot's position. The module uses Sonair's ADAR One 3D ultrasonic sensor, which the company says is the first 3D ultrasonic sensor to receive SIL 2 / Pl d safety certification.

  5. OFweek Robotics38

    Unitree's new robot "Kung Fu Legend" loses a leg on landing after side flip at conference performance

    Unitree's new robot "Kung Fu Legend" was billed in a promo video as able to lift a car bare-handed, but on October 22 at its own conference it performed a side flip and failed to stick the landing, with one mechanical leg detaching on the spot and flying off. The next day Unitree Technology hit the daily limit up on the STAR Market, with market value surpassing 250 billion. The article also cites Musk saying that although Tesla's Optimus is already in production, about 10,000 details still need tuning, and external deliveries will be pushed to the end of 2027.

  6. arXiv cs.RO62

    TouchSafeBench: A Human-Robot Collaboration Benchmark for Evaluating Collision Grounding in Vision-Language Models

    The authors propose TouchSafeBench, a physically grounded benchmark built in Habitat 3.0 for evaluating the collision grounding capability of vision-language models (VLMs). It contains 2,940 simulated indoor co-presence episodes covering two scenario types, social navigation and social rearrangement, and provides synchronized multi-view RGB-D observations, top-down trajectory maps, calibrated camera metadata, and simulator contact labels.

  7. arXiv cs.RO58

    FH-HBTS: A Haptic Bilateral Teleoperation System for Free-Hand Dental Procedures

    The authors propose FH-HBTS, a haptic bilateral teleoperation system for free-hand dental procedures, featuring a mechanical end-effector compatible with standard clinical tools and equipped with an endoscopic camera. The system improves the operator's sensory awareness and motor accuracy through motion and force correspondence between the operator and the robot, and limits interaction forces by scaling the motion references provided to the admittance controller based solely on measured contact forces, thereby achieving force limitation without prior knowledge of the environment. Technical evaluations and in vitro preclinical studies on dental models showed that the system improves the naturalness, safety, and precision of teleoperation.

  8. arXiv cs.RO58

    Study Reveals Flaws in Humanoid Robot Safety Filters; SIREN Can Automatically Generate Collision Attacks

    The paper points out that the mathematical guarantees of humanoid robot safety filters hold only when parameters are configured correctly, while the parameters depend on the specific task and environment and may fail after deployment. The authors propose goal insertion attacks and develop an automated tool, SIREN, which identified attacks in all vulnerable scenarios in the SPARK dataset, three times the number found by baseline methods, and triggered collisions on a real humanoid robot.

  9. arXiv cs.RO32

    How Human Corrections Travel Across Robots: Safety Issues in Federated VLA Adaptation

    Regarding the prospects of combining federated learning with VLA policies, the research points out that sharing human-correction adaptations across tasks, environments, people, and robot embodiments carries safety risks: a shared update that raises average task success may suppress rare safety interventions or produce unsafe responses on another robot.

  10. arXiv cs.RO60

    VERID: A semantic digital twin infrastructure for agentic self-play in automated laboratories

    The authors propose VERID, a persistent, high-fidelity semantic digital twin for automated laboratories that provides AI agents with four core services: queryable introspection, time-compressed simulation, comprehensive trajectory persistence, and a safe execution framework. Three case studies using mixtures of water-soluble dye and deionized water validated its optical and fluidic foundations, demonstrating pre-detection of critical protocol errors, in-run stateful error recovery, and post-hoc identification of subtle physical faults.

  11. arXiv cs.RO60

    AWARE: A world-model framework that predicts robot dynamics and identifies anomalies using only external cameras

    AWARE is a world-model framework that relies solely on external camera monitoring, performing anomaly detection through system dynamics estimation and state prediction without onboard sensors such as motor encoders. On 240 real-world robotic crane trajectories (including payload collisions and motor degradation), its anomaly classification improves by 26% over the strongest baseline; the inferred dynamics parameters can also be decoded to localize the source of faults, achieving 88% accuracy in localizing degrading motors.

  12. arXiv cs.RO58

    Do VLAs Know They Will Fail? Interpretable Success and Failure Patterns in Physical AI Systems

    Research shows that sparse autoencoder (SAE) representations of OpenVLA activations contain outcome-predictive information and can predict task failure before the robot reaches an unrecoverable state. The authors also identify individual SAE features associated with specific stages of success or failure, and uncover recurring behavioral patterns by localizing temporal intervals where these features show significant activation changes. The team further builds a VLM-based interpretation pipeline to generate behavioral descriptions of these patterns and validates their correspondence to feature activations on held-out intervals.

  13. arXiv cs.RO56

    Move, don't think: Sample-efficient collision avoidance learning with lossy proximity sensors

    This paper investigates whole-body proximity sensors as an inductive bias for learning reflexive collision-avoidance behavior, finding that low-resolution lossy sensors were a stronger bias than high-resolution full coverage. The study compared eight candidate representations in a reinforcement learning environment, including latent compression, future contact anticipation, lossy corruption, and privileged state estimation, and trained over 200 humanoid robots end-to-end to play a simplified version of dodgeball under intentionally constrained sample and compute budgets. The authors also found that anticipating when and where contact occurs actually reduced learning sample efficiency.

  14. arXiv cs.RO63

    Lunar swarm robots exhibit emergent cooperative thermoregulation via reinforcement learning

    The study used reinforcement learning to train a single policy shared by a set of twenty robots, with only the mean temperature of the five coldest robots at the end of the night as a terminal reward, and two behaviors emerged: aggregation and rotation between the core and the edge. The minimum stored energy required per robot for survival was measured at 155.6 +/- 1.3 pcm-h (1 pcm-h is 7.2 kJ of stored heat), 37% lower than stationary huddling and within 13% of a physics-based lower bound; under an energy budget in which stationary huddling would lose six robots, all robots survived.

  15. arXiv cs.RO62

    SEVI: A Safety Evidence Framework for Real-World Reinforcement Learning

    The authors propose the SEVI framework, which uses an evidential safety critic to simultaneously estimate future constraint-violation risk and epistemic uncertainty in a single forward pass, judging before an action is executed whether it is unsafe or insufficiently supported by evidence, and overriding that action with an offline-trained recovery policy.

  16. HuggingFace Daily Papers59

    OSMA-Bench: An Obstacle-Conditioned Safety Benchmark for Household Robot Navigation

    The study proposes OSMA-Bench, an obstacle-conditioned safe motion adaptation benchmark built on spatially matched scenario sets that differ only in obstacle type, and introduces the Semantic Sensitivity Index (SSI) to measure whether a robot's clearance and its distance-dependent speed and smoothness follow the caution ordering of obstacles.

Oct 25Sun
  1. 电子工程专辑 · 微信公众号42

    The mass-production secret of humanoid robots lies in the automotive supply chain

    Humanoid robots are extensively reusing resources from the automotive and industrial automation supply chains. McKinsey data shows their BOM cost is about $30,000 to $150,000 per unit, with a long-term target of pushing it below $20,000. However, actuators, force and tactile sensing, and computing and control are listed as high-risk bottlenecks, and dexterous hands are difficult to develop by reusing adjacent industries because of the lack of a standardized, large-scale supply ecosystem. Tesla Optimus's two hands, integrated with the forearms, contain more than 50 actuators, while the rest of the body has only 25 to 30; Tesla has used the Chinese supply chain to cut costs to $20,000 to $30,000, and Unitree's R1 is officially priced from 29,900 yuan.

  2. 机器人全球资讯 · 微信公众号40

    Unitree H2-X Humanoid Robot Malfunctions During Side Flip Demo, Mechanical Lower Leg Detaches

    During a live kung fu demonstration at the "2026 World Physical AI Conference and Unitree Partner Ecosystem Conference," the Unitree H2-X humanoid robot lost balance on landing from a side flip and fell sideways, with one mechanical lower leg detaching from its joint. The robot was just released on October 21, officially named "H2-X Kung Fu Legend Guardian," and is positioned for security and emergency rescue scenarios with a maximum battery life of about 8 hours. Promotional videos show it kicking in car doors, lifting a car bare-handed, and performing a flip after jumping from a platform about 4 meters high. As of now, Unitree has not commented on the incident.

Oct 24Sat
  2. 钛媒体20

    From the humanoid robot "Mobao" to automotive-grade chips: automotive safety pushed to the margins under carmakers' cross-industry narratives

    In 2026, multiple carmakers densely showcased the humanoid robot "Mobao," automotive-grade 5nm chips, eVTOL and Robotaxi at launch events, with cars no longer the primary subject of the narrative. In the same period, four government departments convened more than 30 carmakers to require strict control of smart-driving promotional messaging and to prohibit blurring the boundary between assisted driving and autonomous driving; in August, the State Administration for Market Regulation recalled more than 4.36 million new energy vehicles, with most issues pointing to emergency mechanical door handles that are hard to identify and operate after a collision. A J.D. Power survey shows reliability/durability ranks first among car-purchase factors with a 78% mention rate, and only 23% of consumers are willing to pay for autonomous driving.

  3. Intel Capital · X58

    Figure 03 walks 348 miles over nine consecutive days on a treadmill

    Intel Capital, citing a report by Interesting Engineering, said Figure 03 walked 348 miles on a treadmill over nine consecutive days, logging 1.2 million steps in total. Figure CEO adcock_brett said the test began as a side project aimed at putting stress on the robot and exposing hardware failures, and that it tests reliability better than short demonstrations.

Oct 23Fri
  1. 机器之心 · 微信公众号58

    Shanghai Jiao Tong University and other institutions propose Physical RSI framework, using L0–L5 levels to measure robot self-improvement

    Researchers from Shanghai Jiao Tong University, UC Berkeley, Stanford University and other institutions proposed the Physical RSI framework in the paper "Who Improves the Robot? Toward Physical Recursive Self-Improvement," incorporating task capability updates and improvement process updates into a unified analysis and presenting an L0–L5 grading scale.

  2. Robotics & Automation News51

    Should inspection robots replace people or assist them? An engineer's two-year field review

    Daniela Shchegelskaya, an electronics and robotics engineer at Elimpus, shares two years of experience using Boston Dynamics' Spot for hazard-zone inspections in live power substations. She argues that robots should take on tasks such as carrying instruments into hazardous locations and providing more detailed viewpoints, keeping people away from danger while preserving key judgment.

Oct 22Thu
  1. The Verge – Robot56

    Eiliko parent company Energize Lab says it will hire an independent security firm to review its systems, but a reporter's tests found vulnerabilities remain

    After The Verge reported serious security vulnerabilities in its robot toy Eiliko, Energize Lab issued a statement saying it has confirmed the core issues identified in the report. The company said it will hire an independent security firm to conduct a full review of its systems, set up a security contact email at security@energizelab.com, and has already closed external access to its backend, disabled the interface that allowed remote modification of the robot's replies, taken the long-term memory and custom avatar features offline, and will delete stored conversation records, user profiles, and avatar images.

  2. 3D视觉工坊 · 微信公众号76

    SenseTime and joint team evaluate GPT-6-Astra embodied capabilities, open-source SenseNova-RoboRSI framework and 27B model

    A joint team including SenseTime and The Hong Kong University of Science and Technology (Guangzhou) released the report "Has GPT-6-Astra Achieved Embodied Intelligence?", systematically evaluating GPT-6-Astra's embodied capabilities across 12 benchmarks, and simultaneously open-sourced the SenseNova-RoboRSI evaluation framework and a 27B embodied model.

    Why it matters: The report breaks model capabilities into testable conditions and provides harness ablation data, so readers can judge the real bottlenecks in embodied capability evaluation and cost.

  3. arXiv cs.RO51

    FallGown: A Soft Pneumatic Wearable Device for Fall Prevention

    FallGown is a soft wearable support structure made of pressurized air pouches inside a fabric shell; it can support the full weight of a user's body to prevent falls. When stored it measures 6.9 kg and 0.051 cubic meters, with an average donning time of under 2.5 minutes and doffing time of under 0.5 minutes. Single-user fall prevention testing showed a tradeoff between device size and resistance to tipping; a configuration with a base, 12 air pouches, and a skirt-bottom diameter of 1.4 meters prevented all falls except for one mechanical failure. Onboard pressure sensors can sense the user's weight distribution, offering potential for monitored exercise and adaptive support.

  4. arXiv cs.RO60

    Safe-Night VLA: Thermal-Perceptive Vision-Language-Action Models for Safety-Critical Manipulation

    Safe-Night VLA is a multimodal manipulation framework that integrates long-wave infrared thermal perception into a pre-trained vision-language backbone, enabling robots to perform semantic reasoning based on thermodynamic properties and to enforce deterministic workspace constraints during policy execution through a control barrier functions safety filter.

  7. arXiv cs.RO45

    Sufficient and Necessary Causal Explanations for Autonomous System Failure Analysis

    An arXiv paper proposes distinguishing two goals in autonomous system failure analysis, reproduction and avoidance, and formalizes the concept of "necessary causal explanations," which together with existing "sufficient explanations" forms a unified causal analysis framework. Based on actual causality and causal responsibility, the paper presents an efficient approximation algorithm for generating necessary explanations, evaluates its efficiency across multiple environments, and uses counterfactual generation to verify that it outperforms existing methods on failure avoidance.