Skip to content
#Perception and localization
Oct 28Wed
  1. 机器人全球资讯 · 微信公众号58

    CASIVIBOT Industrial Embodied Inspection Robot Debuts Two New Models at CIIF

    CASIVIBOT, the industrial embodied inspection robot from CASIVIBOT, showcased its Z1-Lite motor stator model and Z1-SE hard material model at the 26th China International Industry Fair. The Z1-Lite targets mass-production inspection of micro motors, completing loading and unloading, defect detection, and automatic sorting in one station, with 5-micron inspection precision and 5 seconds per part; the Z1-SE relies on a self-developed few-shot visual inspection engine to identify chipping, scratches, dents, and dirt, with 7-micron inspection precision and 20 to 35 seconds per part.

  2. 机器人全球资讯 · 微信公众号55

    First CMG World Robot Mount Tai Climbing Competition Qualifying Round Kicks Off in Rain

    On October 26, the qualifying round of the first CMG World Robot Mount Tai Climbing Competition started on the trail from Red Gate to Zhongtian Gate in Mount Tai, with multiple domestic teams fielding legged and wheel-legged composite robots. The course is about 3.8 kilometers long with more than 3,100 steps. Rain during the race made the stone steps slippery, testing the robots' balance control, perception and decision-making, and battery life in a real mountain environment. One robot from Dongtai Technology completed the speed race in about 58 minutes with about 80% battery remaining. The event also includes a loaded-carrying race for legged robots, simulating mountain emergency supply delivery scenarios.

  3. 高工咨询 · 微信公众号28

    Huaray Technology Showcases New Machine Vision and AMR Products at Shenzhen Electronics and Semiconductor Exhibition and Machine Vision Exhibition

    Huaray Technology exhibited new machine vision and AMR products and solutions at the Shenzhen Electronics and Semiconductor Exhibition and Machine Vision Exhibition held October 27-29. On the vision side, the exhibits included a 65MP CXP large-area-array camera, a 5MP short-wave infrared camera, a 16K 256-level TDI line-scan camera, and the R5000Pro series AI smart code readers, covering processes such as wafer AOI inspection, scribe-line positioning, and micro-crack detection.

  4. 维科网机器人 · 微信公众号58

    Topstar signs orders for about 10,000 industrial robots in first three quarters, robotics business revenue up 88.83%

    Guangdong robotics industry chain lead company Topstar disclosed its performance for the first three quarters of 2026: industrial robot orders totaled about 10,000 units and shipments about 8,600 units, while revenue from its industrial robot and automated application systems business reached 1.028 billion yuan, up 88.83% year on year. Overall revenue was 1.969 billion yuan, up 16.68% year on year, and net profit attributable to shareholders was 146 million yuan, up 196.38% year on year. The company has submitted an application for a Hong Kong listing and disclosed embodied intelligence products including Xiaotuo and Xingzai, as well as the GMU-400E humanoid robot component machining solution.

  5. 节卡机器人 · 微信公众号60

    JAKA Kargo wheeled humanoid robot deployed for bin handling at auto parts factory

    The JAKA Kargo wheeled humanoid robot, running the JAKA EVO embodied intelligence operating system, has been deployed for bin handling at the headlight plant and electronics plant of a leading Chinese auto parts manufacturer. Per-bin handling time dropped from 105 seconds per bin to 28 seconds per bin, nearly a 4x efficiency gain, and the robot also stacks bins five layers high and recovers empty bins by layer.

  6. 众擎机器人 · 微信公众号47

    First URKL Qualifier's Top 16 Revealed, Each Receives an EngineAI T800 Robot

    The 32-to-16 qualifying round of the first URKL Global Humanoid Robot Free Combat League has ended, with 16 teams advancing to the group stage and each receiving an EngineAI T800 robot. The qualifiers held six days of offline training at the Dasha River Cultural and Sports Center in Nanshan, Shenzhen. All teams used the T800 as their unified hardware platform, deploying code refined in simulation onto real robots for the first time, and were evaluated on basic combat moves such as getting up after a fall, navigating complex terrain, and punching and kicking. The event features a prize pool of tens of millions, and industry partner Dison Digital will provide the top 16 with motion capture experience benefits.

  7. 3D视觉工坊 · 微信公众号58

    Suzhou sandisiwei releases NexusPickit-S1 bin-picking software with source code and tutorials

    Suzhou sandisiwei has launched NexusPickit-S1 bin-picking software, aimed at automated picking of industrial parts that have no fixed pose and are randomly stacked. It provides source code, video tutorials, and Q&A support. The software's hand-eye calibration can be completed within 5 minutes with millimeter-level accuracy. In a configuration with a KW-MINI structured-light camera and a Dobot CR5 collaborative robot arm, picking accuracy is within about 5 mm; on an i7-13700H with 24 GB of memory, the cycle time is about 1.2 s per part, and it can run in a Windows environment without a GPU.

  10. arXiv cs.RO62

    Goal2Pixel: A Vision-Language Navigation Method That Grounds Goals to Pixels

    Goal2Pixel proposes a pure pixel paradigm that reformulates vision-and-language navigation in continuous environments (VLN-CE) as navigable pixel grounding: the model predicts navigable pixels on the image plane, which are then back-projected into 3D waypoints for forward movement, while turning and stopping are encoded by auxiliary left, right, and bottom regions.

  11. arXiv cs.RO60

    QUAN: Quaternion Approximate Networks Improve Image Classification and Oriented Object Detection

    The paper proposes Quaternion Approximate Networks (QUAN), which use Hamilton product decomposition to approximate quaternion convolution with real-valued operations, paired with custom CUDA kernels to achieve rotation-equivariant image classification and object detection. The authors also propose Independent Quaternion Batch Normalization (IQBN) to stabilize training and extend quaternion operations to spatial attention mechanisms.

  13. arXiv cs.RO58

    Compact Force Sensor and Tension-Aware Outer-Loop Control for Dual-UAV Cable-Suspended Payload Transport

    This study proposes a dual-UAV payload transport framework that uses a custom compact force sensor at the cable anchor point to measure interaction forces, with static and dynamic tests evaluating its linearity, hysteresis, repeatability, and crossload. Control uses a cascade structure: inner loops handle vehicle stabilization, while outer loops compensate for the measured forces, validated in simulations and indoor experiments under position uncertainty. Payload-drop and constrained-space tests compared with literature-based distributed methods show improved stability, coordination, and disturbance rejection.

  14. arXiv cs.RO60

    Temporal Tactile Encoding and Compliant Control for Intent-Aware Robot-to-Human Bimanual Handover

    This study combines a VLA model with a compliant controller, fine-tuning the model using RGB observations, temporally encoded tactile feedback, and proprioception from human demonstrations to achieve robot-to-human bimanual handover. In a human-subject study, the system was compared against two baselines: one without tactile feedback and one with tactile feedback but no compliant control. Results show that compliant control and tactile history are complementary, with performance clearly outperforming the baselines. The project page is publicly available.

  15. arXiv cs.RO60

    From Language to Navigation Goals: A Vision-Language Approach for Semantic Navigation of Mobile Robots Using RGB-D Perception

    The study proposes a language-driven navigation framework that enables mobile robots to understand natural-language requests such as "go to the mailbox," identify the referenced object, estimate its position using RGB-D data, and generate a navigation goal that is then executed by the ROS 2 Nav2 navigation stack.

  16. arXiv cs.RO70

    FAWAM: Force-Aware World Action Model for Closed-Loop Contact-Rich Manipulation

    The authors propose FAWAM, a force-aware world action model that incorporates force information at three levels: perception, prediction, and closed-loop execution. It first encodes historical 6-axis force/torque signals to modulate action generation, then jointly predicts future actions and end-effector wrenches to explicitly model contact evolution, and uses the predicted wrench trajectory as an execution-time reference while correcting actions online with real-time force feedback. In real-world experiments across multiple contact-rich tasks, FAWAM achieved an average success rate 36.25% higher than vision-only baselines and 21.25% higher than existing force-aware baselines.

  18. arXiv cs.RO62

    RAPID: A Robotic Agent Platform for EV Battery Pack Disassembly

    The authors propose RAPID, a robotic agent disassembly platform that integrates a gantry-mounted industrial robotic arm, RGB-D perception, and an automated nut-running tool to remove fasteners on a full-scale EV battery pack. Open-vocabulary object detection achieves 0.9757 mAP50; among three one-shot fastener removal strategies, taught-in poses achieved a 97% success rate in 24 minutes, one-shot vision execution achieved 57% in 29 minutes, and visual servoing achieved 83% in 36 minutes (n=204).

  21. arXiv cs.RO60

    RPT*: Global Planning with Probabilistic Terminals for Target Search in Complex Environments

    The paper studies the Hamiltonian Path Problem with Probabilistic Terminals (HPP-PT), whose goal is for a mobile robot to visit all candidate locations to find a target while minimizing expected path cost. The authors propose RPT*, a search method with optimality guarantees that uses dynamic programming in a new state space to bypass history dependency, along with a heuristic to speed up computation. The RPT*-based HATS system, combined with Bayesian filtering or autonomous exploration, found targets faster than baseline methods on average in both simulation and real-robot experiments.

  22. arXiv cs.RO62

    CLEAR-IR: Reconstructing Infrared Images into RGB-Style Images to Support Robot Perception in Dark Environments

    CLEAR-IR proposes a dual-stream architecture that reconstructs clean, RGB-style images from infrared images containing structured-light emitter patterns, allowing vision pipelines originally trained on RGB to keep working when no visible light is available. Inspired by DeepMAO, the architecture was evaluated on object detection, ArUco marker detection, and visual SLAM, covering scenarios ranging from well-lit to completely dark; in complete darkness, passive RGB enhancement methods fail because the sensor receives no signal. The paper has been accepted to IEEE/RSJ IROS 2026.

  23. arXiv cs.RO59

    Study: Active sampling with vision-based tactile sensors for hardness classification

    The paper investigates active sampling strategies based on vision-based tactile sensors (VBTS) for sample-efficient hardness classification, using information-theoretic measures of model uncertainty. The authors evaluated three probabilistic classifiers, two active sampling strategies, and a random sampling baseline on a hardness classification task using their own VBTS dataset collected by a robot and a previously published dataset, and also conducted a human study on hardness classification with the same objects. Results show that the active sampling methods using variance and entropy as uncertainty measures performed best, reaching 90% accuracy, compared with human accuracy of 48%.

  24. arXiv cs.RO62

    Research team builds egocentric tennis dataset and evaluates vision models' spatiotemporal prediction capabilities

    The research team collected an annotated egocentric tennis dataset and trained and evaluated image encoders, video encoders, and vision-language models on seven spatiotemporal prediction tasks. The best models achieved current-frame localization errors of 0.14 m for the egocentric player and 0.54 m for the ball, with trajectory prediction errors of 0.34 m and 0.91 m respectively within 1.1 seconds, and a mean error of 0.14 seconds for contact-time prediction. Auxiliary depth information notably improved 3D ball localization, while longer image histories did not improve ball trajectory prediction; the team also proposed a conditional diffusion model that generates full-body tennis kinematics from egocentric video at real-time rates.

  25. arXiv cs.RO55

    MCP-Net: A Real-Time Coronary Perception Network and Topological Analysis for Robotic Guidewire Navigation

    The authors propose a simulation-driven framework centered on a multi-task coronary perception network, MCP-Net, for robotic guidewire navigation, integrating coronary vessel segmentation, vascular topology extraction, and roadmap generation. The method generates simulated fluoroscopy images and simulation-derived ground truth from three-dimensional coronary CT angiography, including simulated guidewires, stenotic lesions of varying severity, and diverse vascular geometries. MCP-Net jointly performs vessel segmentation, guidewire-tip localization, and stenosis localization from a single fluoroscopic frame.

  26. arXiv cs.RO38

    Polarity-Time Interaction and Multi-Level Reliability for Event-Based Spacecraft Pose Estimation

    This study proposes a reliability-aware framework that connects polarity-time localization, geometric recovery, event-CAD refinement, and trajectory optimization in sequence, using a three-branch representation that combines positive events, negative events, and recent-event timing. On the official SPADES challenge test set, the complete system achieves a combined pose score of 0.0474, about 5.4% lower than PASAN+IRIS. Ablations show that base localization with uniform PnP scores 0.0772, while the full cascade reduces it to 0.0514; under identical single-frame observations, quality weighting lowers the score from 0.0549 to 0.0514.

  27. arXiv cs.RO72

    RESPLE2: Inertial-Informed Non-Uniform B-Spline Multi-Sensor Odometry with Global Optimization

    RESPLE2 proposes a recursive multi-sensor state estimation framework based on non-uniform B-splines for 6-DoF motion representation, using the knot-clamping boundary property to predict control points directly from inertial kinematics. Prediction and reset are performed within an iterated extended Kalman filter, and the update step fuses LiDAR, camera, IMU, and leg kinematics at their native timestamps; a separate lightweight global optimization module refines control points using LiDAR-visual loop-closure constraints.

  28. arXiv cs.RO71

    Research team uses differentiable simulation to learn quadrotor dynamic narrow-gap traversal from scratch

    The research proposes a differentiable moving narrow-gap modeling method that gates centering, alignment, and smooth collision terms by signed distance to a time-varying gap plane, turning sparse traversal events into dense pathwise objectives, and integrates this into a visual-feature-to-command policy trained in a single stage from random initialization, without a teacher, planner, or distillation. In simulation, path gradients clearly outperform score-function gradients, reaching a 99.3% success rate after a few minutes of training on a single GPU. The policy transfers directly to a real quadrotor using only onboard vision and inertial measurements, successfully traversing the dynamic narrow gap in 8 of 10 real-world trials.

  29. arXiv cs.RO80

    Rendering-Free Event Frame Simulation Accelerates Reinforcement Learning for Autonomous Robot Navigation

    The authors propose an analytic event simulator that synthesizes event streams directly from scene geometry and ego-motion without image rendering, generating 64x64 event observations in just 17 microseconds on a single GPU, over 400 times faster than state-of-the-art rendering pipelines. Reinforcement learning training for an event-camera obstacle-avoidance policy based on this simulator converges in under 30 minutes, three orders of magnitude faster than training on rendered events, and transfers zero-shot to ray-traced event simulation in Isaac Sim.

    Why it matters: This work uses analytic event simulation to bypass image rendering, cutting event-camera reinforcement learning training time to under 30 minutes, and completes real-world deployment on a 56-gram drone.

  30. arXiv cs.RO60

    StabBot: Robotic Pipe Stabbing with Learned Stereo Vision and Residual Reinforcement Learning

    The StabBot robotic stabbing system uses a pendulum Kalman filter to track pipe joints in stereo images, approaches at matched speed, and confirms the final few millimeters by touch. In 200 randomized jobs, StabBot completed every stab, versus 74% for a conventional fixed guide; median contact speed fell from 18.4 cm/s to 2.3 cm/s, and median peak head-pipe force fell from 11.3 kN to 1.4 kN.

  31. arXiv cs.RO58

    Non-line-of-sight passive object sensing using distributed sensing and hyperdimensional computing

    The paper proposes a framework that relies solely on distributed channel measurements from communication infrastructure to sense passive entities outside a robot's field of view and perform risk-aware classification. The method uses hyperdimensional computing (HDC) to embed measurements into high-dimensional representations without explicit channel reconstruction. Numerical simulations based on commercial ray tracing and validation in a ROS smart factory scenario show accuracy of up to 89% under severe occlusion while maintaining low computational complexity.

  32. arXiv cs.RO60

    Reinforcement Learning-Based Autonomous Fixed-Wing Drone Interception

    A team at Delft University of Technology in the Netherlands proposed a fixed-wing interceptor controlled by a neural network that maps filtered monocular vision observations directly to actuator commands; the policy was trained in a custom Python environment using reinforcement learning and proximal policy optimization (PPO). In simulation, the policy outperformed two classical guidance controllers under most conditions, and it has successfully intercepted maneuvering fixed-wing targets in real flight. This work provides validation for the autonomous control of fixed-wing interception drones.

  33. arXiv cs.RO60

    AuRaSim: A Ray-Tracing Automotive FMCW Radar Simulator for Dynamic Scenarios

    AuRaSim is a ray-tracing-based automotive FMCW radar simulation framework that generates coherent multi-channel ADC data for dynamic scenarios. It builds dynamic scenes with Blender, performs GPU-accelerated electromagnetic ray tracing with Sionna RT, and adds path-level complex-coefficient propagation, motion-aware Doppler modeling, and waveform-configuration-driven ADC cube synthesis. In a digital twin scenario, the simulated and measured range-Doppler maps achieve an RMSE of 0.147 and a PSNR of 16.93 dB.

  34. arXiv cs.RO60

    Target-free joint camera-LiDAR intrinsic and extrinsic calibration method

    The authors propose a fully automatic, target-free framework that jointly estimates camera intrinsics, radial-tangential distortion, and camera-LiDAR extrinsics from raw images and LiDAR scans. The method extends deep pixel-point correspondence-based calibration from rectified images to raw cameras through intrinsic- and distortion-aware correspondence learning, and iteratively couples correspondence prediction with nonlinear optimization during calibration. Zero-shot evaluations on an unseen KITTI configuration and on three real robotic platforms show that joint calibration improves extrinsic accuracy while recovering accurate intrinsics.

  35. arXiv cs.RO57

    KEPT: Conformal Prediction Keepout Planning for Enclosed-Space Navigation Under Degraded Sensing

    KEPT proposes a navigation method for enclosed spaces such as tunnels and mines, using conformal prediction to calibrate keepouts around mapped walls under degraded sensing, and adjusting the margin according to a label-free degradation severity estimate to maintain coverage. The planning and control layer enforces a braking-distance condition on the keepout at plan time and on every command, using the measured tracking residual as its only margin and never altering the keepout.

  36. arXiv cs.RO53

    LSMM: Learning Spatial Memory for End-to-End Legged Robot Exploration

    The authors propose Local Spatial Matrix Memory (LSMM) based on Hadamard memory, which lets reinforcement learning agents store task-relevant information in position-based memory cells. On an autonomous maze exploration task, LSMM's maze coverage is comparable to strong recurrent networks and finite-history baselines, and it is the only architecture that remembers the maze layout, reducing reconstruction error to one third. The work targets the problem that humanoid and legged robots in complex environments need to remember information long after it leaves their field of view before it is used.

  37. arXiv cs.RO70

    RAVEN: A Unified Active Visual SLAM System for Quadrotor Localization Uncertainty

    RAVEN is an active SLAM framework for quadrotors that unifies planning of exploration and active loop closure under localization uncertainty, and uses a priority parameter from 0 to 1 to adjust the emphasis between exploration efficiency and localization accuracy. The framework first identifies frontiers and active loop closure candidates and generates global tour plans, then evaluates them with a utility vector, and finally a perception-aware navigation planner generates smooth trajectories based on an online Fisher information field to reduce visual state estimation error.

  38. arXiv cs.RO54

    State Estimation for a Hybrid Rigid-Continuum Space Manipulator System

    This study proposes a state reconstruction pipeline for a hybrid rigid-continuum space manipulator whose distal link is a tendon-driven continuum robot. The method formulates the estimation problem as maximum a posteriori inference conditioned on joint encoders and vision measurements, transcribes it as a factor graph, fuses heterogeneous measurements with the rigid-soft kinematic model, and uses the Geometric Variable Strain (GVS) approach to obtain a minimum-order parameterization of the configuration. Closed-loop trajectory simulations verified the state reconstruction accuracy and computational performance.

  39. arXiv cs.RO64

    Omni-IGN: Omnidirectional Vision Onboard Image-Goal Navigation System for Drones

    The authors propose Omni-IGN, a fully onboard drone image-goal navigation (ImageNav) system that integrates quad-fisheye perception, an attention-based reinforcement learning policy, and a collision-aware safety layer. The policy adaptively fuses goal-relevant features across multiple views to provide directional guidance for continuous navigation, and the quad-fisheye RGB images support both ImageNav and obstacle avoidance without a dedicated depth sensor.