Skip to content

Technical Directions

Robot safety Latest coverage

Robotics coverage relating to robot safety.

124 selected reportsLast 30 days 113 recordsAll coverage 361 records

updated

Signal archive — Page 5

2026-11-02Mon81–100 records
  1. arXiv Robotics — research abstracts75

    Robot Navigation and Canine Welfare: A Pilot Study Surfacing Challenges for Cross-Species Legibility and Proxy Validity in Animal-Machine Interaction

    This paper presents a pilot study evaluating a curve-out maneuver designed to improve animal welfare during robot navigation. Contrary to expectations, the maneuver caused higher stress responses in companion dogs compared to straight-line trajectories. The authors attribute this to discontinuous execution that reduced behavioral legibility, highlighting structural challenges for cross-species interaction design.

  3. arXiv Robotics — research abstracts75

    VLA Policies Do Not Know When Not to Act: Unintended Task Execution and Start Suppression

    This research investigates the behavior of vision-language-action (VLA) policies when given missing, meaningless, or incorrect instructions. Using six model checkpoints in the LIBERO-Goal simulation environment, the authors found that robots consistently initiated movement even with empty inputs and frequently completed unspecified tasks. While a gating mechanism successfully suppressed motion on non-task inputs for five of the six models, fine-tuning approaches failed to reliably keep robots still. The study concludes that current VLA architectures lack the ability to distinguish between valid commands and noise, posing potential risks in real-world deployment where input reliability cannot be guaranteed.

  4. arXiv Robotics — research abstracts75

    Joint-Health-Aware Reinforcement Learning for Robust Robot Control Under Actuator Degradation

    This paper introduces a joint-health-aware reinforcement learning framework designed to maintain robust robot control despite actuator degradation. The method encodes actuator condition in a continuous joint-health vector and models degradation as multiplicative attenuation of actions, using a HealthNet estimator to infer per-joint health from observations. Evaluated on Walker, Go2, Cassie, and Panda-based tasks under diverse degradation masks, the approach outperforms nominal baselines, with task-level locomotion returns improving by up to 250.1% and manipulation success rates by up to 15.2 percentage points relative to the nominal baseline.

  5. arXiv Robotics — research abstracts75

    TRACE: Target-State Risk-Aligned Conditional Execution for Embodied Tracking

    This research presents TRACE, a method for embodied visual tracking that balances persistence with safety. Using a compact 0.6B vision-language-action (VLA) tracker, the authors update only the planner head with proximal policy optimization (PPO). While PPO improves success rates, it increases collision risks in distracted scenarios. TRACE addresses this by introducing an execution-time risk gate that processes 84 cues and selectively rescales translational waypoints when calibrated risk exceeds a threshold. In paired evaluations, TRACE reduces the distracted-tracking collision rate by 3.96 percentage points while further improving success and following rates, demonstrating that target-state-aware execution can recover safety costs with minimal intervention.

  6. arXiv Robotics — research abstracts75

    Veloxity: Towards Memory-Safe UAV Embedded Firmware using the Rust Programming Language

    This paper introduces Veloxity, an open-source embedded autopilot for unmanned aerial vehicles (UAVs) written in the memory-safe Rust programming language. The authors address the fragmented adoption of memory-safe languages in aerospace by providing a minimal, modular foundation that integrates with established autonomy ecosystems. They compare Veloxity to equivalent C++ autopilots using simulation, hardware testbenches, and outdoor waypoint-following flight tests, demonstrating functional parity and improved performance. These results suggest that modern memory-safe languages are a viable option for reliable and performant firmware in autonomous aerospace systems.

  7. arXiv Robotics — research abstracts75

    Confidence-Modulated Safety Filtering for End-to-End Policies

    This research paper introduces a method where the conservatism of a safety filter is dynamically adjusted based on the confidence level of an end-to-end robot policy. The approach aims to maximize performance within the filter's certified operational domain while adding necessary safety margins when the policy operates beyond its competence or flags potential failures, validated through preliminary simulations on the F1TENTH platform.

  8. Robotics — Chinese web discovery75

    Amicro Technology (2065.HK) Files for IPO as Robot Chip Platform Sees Growth Opportunity

    Amicro Technology, a provider of dedicated SoCs for robot SLAM, navigation, and mapping, is filing for an IPO on the Hong Kong Stock Exchange under Chapter 18C rules. The company's proprietary Amicro Core platform integrates self-developed algorithms with chips, sensors, and motion control modules. Revenue grew from approximately 206 million yuan in 2023 to 407 million yuan in 2025, with multi-modal solutions becoming the primary revenue source by 2025.

2026-11-01Sun
  1. ROS Discourse — Community updates75

    A Deterministic Safety Gate for Differential-Drive Robots in ROS 2

    The qerra_gate_node is a deterministic supervisor node for differential-drive mobile robots, designed to mediate between high-level planners and base controllers. It operates in simulation and CI environments, enforcing safety measures like kinematic stopping envelopes, fail-closed trips, and operator resets. The node is implemented in Python with no external dependencies, and its behavior is verified through simulation and GitHub Actions CI. However, hardware validation and real-world deployment readiness remain unverified.

  2. IT Home — Robotics75

    AI Performs Brain Surgery on Mice for Epilepsy Treatment

    Exin Therapeutics, a biotech startup, claims to have successfully used an AI agent to perform brain surgery on a mouse to treat epilepsy. The AI selected gene therapy vectors, dosages, and brain regions using an electronic spreadsheet, three research papers, and a camera-equipped robotic arm. The AI injected 250 nanoliters of medication while monitoring for bleeding. Human oversight and safety protocols, including an emergency stop button, were in place. The AI was limited to the injection phase, with craniotomy and suturing performed by human surgeons. The team initially attempted to use Anthropic's Opus 5.5 model, which refused the task due to concerns about incomplete execution.

2026-10-31Sat
  1. Chinese robotics — Research news discovery75

    Tumi Robot Completes First Remote Surgery in U.S. History

    Tumi robot has completed the first remote surgery in U.S. history, marking a milestone in the expansion of remote surgical capabilities within the American healthcare system. The procedure, conducted under FDA IDE guidelines, was performed by Dr. Vipul Patel and supported by Dr. Marcio Moschovas, with no visible bleeding and smooth communication throughout the 103-minute operation.

2026-10-30Fri
  1. arXiv Robotics — research abstracts82

    Robust Taylor-Lagrange Control for Safety-Critical Systems with Guaranteed Feasibility

    This research proposes a robust Taylor-Lag, control (rTLC) method to address feasibility preservation in safety-critical systems. By expanding the safety function with a higher-order Taylor expansion and Lagrange remainder, the method ensures safety constraints are satisfied over the entire discretization time interval. The approach is demonstrated on an adaptive cruise control problem and compared with existing safety-critical control methods.

  2. Robotics — Chinese web discovery75

    Embodied Intelligence Collective Experience Learning (CEL) Framework Released: Robots Teach Robots to Perform Tasks

    On October 30, the Collective Experience Learning (CEL) framework for embodied intelligence was released in Shanghai. Developed by Agibot in partnership with Tsinghua University and Wuwen Xinqiong, CEL enables robots to learn from each other's real-world operation trajectories rather than relying solely on human demonstrations. In tests involving five tasks such as flower arranging and phone packaging, robot strategies achieved success rates exceeding those of remote human operators, with the highest improvement reaching 16.7 percentage points. The framework also demonstrated that increasing the number of participating robots within the same physical training time directly improves task execution success rates, breaking previous scalability limits.

2026-10-28Wed
  1. Robotics — Web news discovery75

    Trump Says Imported Robotics Pose National Security Risks but Does Not Impose Tariffs

    President Donald Trump determined that imported robotics and machinery present national security risks but decided against immediate tariffs. Instead, he ordered officials to negotiate with foreign governments on import levels, allowing existing humanoid and quadruped robots to be imported while barring new models of Chinese robots and power inverters following an FCC rule. The Commerce Department and U.S. Trade Representative are directed to report back in 180 days on the status of talks, with the possibility of future tariffs if negotiations fail. The probe covers various industrial machinery including machine tools, autoclaves, laser cutting tools, and programmable computer-controlled mechanical systems.

  2. MIT — Robotics

    A soft robot could help doctors diagnose and treat disorders of the small intestine

    MIT engineers have developed a soft robotic enteroscope that can navigate the small intestine, perform imaging, deliver drugs, and remove tissue samples. Inspired by snake locomotion and kirigami, the robot uses inflatable actuators and textured coatings to enhance traction and maneuverability, offering a potential alternative to traditional endoscopic procedures.

  3. arXiv Robotics — research abstracts75

    Gradient-Free Neural Hamilton-Jacobi Reachability for Scalable Safety-Critical Control

    The paper proposes a discrete-time neural reachability framework for control-disturbance-affine systems that learns backward reachable tubes (BRTs) and backward reach-avoid tubes (BRATs) through Bellman-Isaacs value propagation. It combines equation-driven self-supervision with structured policy learning to avoid explicit PDE-gradients, using gradient-free value probes to construct approximate teacher actions. The method demonstrates scalability up to 80 dimensions and generalizes to real-world applications like F1-tenth racing with over 16,000-dimensional inputs, achieving real-time robust collision avoidance.

2026-10-27Tue
  1. IEEE Spectrum — Robotics78

    Your Robot’s Safety Functions Already Work. What If the Input Lies?

    VicOne LAB R7 explores how adversarial inputs can alter robot behavior, challenging safety assumptions. The article discusses risks from manipulated sensor data, cyber attacks, and the importance of testing safety systems against deliberate threats. It also outlines five key questions for safety and security teams to ensure robot behavior remains safe under adversarial conditions.

  2. Chinese robotics — Hardware and sensing

    2026 Hardcore Achievements | Industry's First. Humanoid Robots Start Regular Duty, Achieve 98% Material Loading Success Rate

    Lepu Robotics and Zhao Feng机电 jointly released the 'Embodied Intelligence + Automotive Parts Loading Solution,' with 15 humanoid robots deployed on Zhao Feng机电 production line for regular operation, completing over 10 action sequences, stably grasping small parts with inner diameters between 80 to 120 millimeters, achieving a stable material loading success rate of 98%, running continuously without faults for more than two months, accumulating approximately 1000 hours of operation, and assisting in the production of over 60,000 parts.

2026-10-26Mon
  1. Chinese robotics — Deployment discovery

    JD Logistics × Zhiyuan Robotics: Embodied Robot G2 Max Applied in Zhi Wang Warehouse

    Zhiyuan Robotics has deployed the G2 Max, an industrial-grade embodied robot, in JD Logistics' Zhi Wang warehouse for inbound handling tasks. The robot can carry up to 50kg with high precision, achieving over 99% operational success rate. This marks a real-world application of embodied intelligence, with JD Logistics emphasizing the importance of industrial scenarios for testing and refining such technologies.

Timezone · UTC

Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).