Dexterous Hand Practice 22 builds a closed-loop process from visually recognizing a cup to grasping it stably, connecting perception output into the control stage to complete the grasp. The content is organized around the task chain of "seeing the cup—grasping it stably" and is practical, tutorial-style instruction.
Apple's research team published the paper "Perceptive Humanoid Motion from Simulation" on arXiv, unveiling HALO, an autonomous control system for humanoid robots, and simultaneously open-sourcing the training and deployment code.
Robbyant has open-sourced LingBot-Touch, a tactile foundation model pretrained on about 5,000 hours of real tactile data that reads magnetic, vision-based tactile, and piezoresistive sensor inputs into a unified token space. Across 26 real robot tasks and two dexterous hands, four policies using these tokens all achieved the highest average success rates; on specific sensor models unseen during pretraining, a frozen encoder reduced contact-center error from 5.89 mm to 0.16 mm, outperforming an adapter built specifically for it.
Why it matters: The tactile foundation model unifies multiple types of tactile sensors into the same token space and provides cross-sensor and cross-task transfer results.
ASICS America has deployed Corvus Robotics' Corvus One autonomous inventory drones at its Byhalia, Mississippi distribution center, flying 7 times each night. Counting frequency in two high-risk storage areas has increased 6x and those areas are now counted monthly, while misplaced inventory is recovered about 60% faster.
IIC Shenzhen 2026 will be held on December 9-10 at the Sheraton Shenzhen Futian Hotel. The series of summit forums will break down the industrialization path of physical AI across dimensions including edge computing power, visual perception, storage architecture, power devices, quadruped robot practices, and open-source AIoT solutions.
A developer built an open-source desktop companion robot using a MacBook, an Android phone, and a Raspberry Pi Pico W: the Pico W drives the two servos of a 2-DOF pan-tilt head, the MacBook uses MediaPipe and AprilTags to recognize people and convert angles, and the phone screen displays eye animations that can track faces.
On November 10, Ant Group's Lingbo Technology open-sourced LingBot-Touch, a tactile foundation model pre-trained on nearly 5,000 hours of real tactile data. It covers three sensing principles—magnetic tactile, visuotactile, and piezoresistive—and outputs unified tactile tokens for use by different dexterous hands and action models.
Why it matters: The original text provides a unified representation across three types of tactile sensing and data from 26 real-machine tests, which can be used to assess the gains a tactile foundation model brings to existing action models.
Point clouds are discrete geometric samples of three-dimensional space. Unlike regular images, they must be converted into a computable BEV representation through structured processing such as voxels or pillar bucketing. The article reviews point cloud algorithms along the chain from data fundamentals to models, engineering deployment, and quantization, and organizes common computation modules.
In a Shanxi project, the Rhino unmanned forklift had its radar raised by 1 m; this is not parameter tuning but a structural design for shelf compatibility. Using the unmanned forklift as an example, the article notes that robot engineering deployment failures are often due to misalignment between on-site requirements and equipment configuration, rather than inadequate algorithms.
Berkshire Grey has released multiple enhanced capabilities driven by its unified AI software platform BG Pilot, covering Scoop unloading perception, Stride intelligent orchestration, and Core precision packing.
EDGE Sound Research has raised a new $2 million round led by Stormbreaker Ventures, bringing its cumulative funding to $9 million, to expand its sound-based perception technology into robotics and physical AI.
The article discusses why injecting only Gaussian white noise in ROS2 mobile robot simulation is insufficient for validating fusion algorithms, focusing on systematic errors such as bias, scale, and mounting misalignment that cannot be averaged out by filtering, and provides a complete approach for injecting systematic errors into odom, lidar, and IMU.
The authors release the FHNW Muttenz dataset, covering a contiguous 10 km street network and two mobile mapping campaigns about 1.5 years apart, with ground-truth poses accurate to 6-9 mm in position and orientation error below 0.1°. Using this dataset, they systematically evaluate sparse-feature visual localization; with daytime imagery and pre-calibrated cameras, median translation error is 1-5 cm and rotation error is 0.03-0.1°. The authors note that, provided the scene representation is precisely georeferenced, visual localization can approach survey-grade GNSS positioning accuracy and additionally provides full orientation information.
The authors propose the first mmWave radar-based onboard perception-and-control dynamic obstacle avoidance system for small quadrotor UAVs. The system combines a lightweight interacting multiple models (IMM) tracker with a control barrier function (CBF) controller that directly outputs evasive accelerations, and derives sufficient avoidance conditions relating sensing range, relative speed, and control delay.
NoContactNoWorries proposes a Transformer-based multimodal framework that fuses RGB-D vision with robot proprioception to infer binary contact states for hand-object interaction as a pseudo-tactile signal. The authors train a single contact prediction model on multiple objects and verify that this inferred signal can support downstream reinforcement learning agents in completing in-hand object reorientation, and that it generalizes to new objects. Both simulation and real-robot experiments validate the feasibility of inferring contact from vision and proprioception. The paper has been accepted to IROS 2026.
The authors propose SGM-SLAM, a distributed SLAM framework for multi-robot collaboration in GPS-denied, communication-constrained environments, where robots can each initialize their coordinate frames independently and share only limited spatial overlap. The method uses fused RGB-LiDAR point clouds to build a scene graph containing a semantically segmented point cloud layer and a discrete bounded object layer, and performs collaborative matching by exchanging and matching object data with neighboring robots, using multi-step data exchange and optimization to improve communication efficiency. The authors validated effectiveness and communication efficiency on real-world datasets containing limited trajectory overlap, viewpoint variation, repeated structures, and degraded visual conditions.
MosaicIMU is a carrier-conditioned Mixture-of-Experts (MoE) pretraining and adaptation framework for generalizable neural inertial odometry. It composes carrier-specific expert features through a prototype-based router, decodes local velocity and uncertainty constraints, and integrates them with a history-aware EKF. For unseen domains, it freezes the pretrained base model and learns a lightweight expert residual branch; for edge deployment, it reuses the router to select online samples for incremental updates.
The authors propose GELS, a gradiometer-based electromagnetic localization system that uses a compact magnetometer array as an embedded quasi-gradiometer to estimate the local magnetic field and gradient tensor, then maps displacements via the Euler homogeneous relation and uses Procrustes registration with at least three noncollinear sources to recover the array's orientation and position. The closed-form algorithm requires only known source positions and array geometry, with no pre-calibrated workspace field maps, initial pose guesses, or calibrated source magnetic moments; in benchtop experiments, the solver's median runtime was 172.00 microseconds. Perturbation analysis identifies inter-sensor inconsistency and dipole model mismatch as the main accuracy limits.
The authors propose XIT, a sampling-based planner that formulates active gas distribution mapping (GDM) as a next-best-trajectory informative path planning problem, growing a single shared tree toward multiple frontier goals while taking informative gas samples along the way.
The authors propose the Seeing What Matters benchmark to evaluate task-conditioned actionable spatial belief in simulated tabletop manipulation, covering 860 frozen evaluation cases across three scenarios—coffee workstation, drawer/cabinet, and microwave—which expand to more than 115,000 evaluation episodes.
The paper proposes a lightweight path-aware frontier re-ranking method for LiDAR UAV autonomous exploration: a lightweight utility function first retains a small top-N set of frontier candidates, then time-bounded RRT* planning estimates their feasible path costs for re-ranking, and the path for the selected frontier is reused for navigation to reduce redundant global planning.
The study deploys a cucumber length-measurement pipeline based on YOLO26n detection, prompt-based segmentation, and medial arc spline estimation to NVIDIA Jetson Orin Nano and Raspberry Pi 4, and compares both against an RTX 3050 Ti workstation.
The research team proposed a two-component optoelectronic soft suction cup designed for kiwifruit that uses optical signals to estimate interaction forces with real fruit within known contact intervals. On fruit not used in training, it achieved grouped cross-validation R²=0.968±0.005 (MAE=0.138 N), with a median per-fruit held-out R² of 0.973 (MAE=0.134 N).
To address the inaccuracy of conventional frequency-domain readout caused by long interconnects and environmental noise in robotic tactile sensing, this study proposes a time-domain optimization readout method that determines resistance values by fitting a time-domain observation model and using the full waveform rather than only spectral amplitudes. In simulations of a 16×16 array, the method reduced the mean readout error by 20% compared with the baseline; validation over 1,000 trials showed reductions of 24.6% in mean error and 77.6% in standard deviation, and the improvement was also confirmed in cross-point array simulations emulating tactile sensor resistance distributions.
The paper formalizes the attribution problem of inconsistencies between scans and models in LiDAR-inertial odometry and BIM fusion as fault identification over three fault classes, and derives a geometric criterion for determining whether attribution is feasible. On cases that are detectable but not attributable in a two-dimensional study, the method based on this criterion had a conditional error rate of 0.83%, compared with 31.67% for a baseline using uncertainty-based abstention; across 5,600 attributable cases, it produced 5,574 unique attributions with no erroneous attributions.
FATHOM proposes a world-model-guided diffusion action-trajectory selection framework for relative tracking of autonomous underwater vehicles (AUVs) under sparse active light markers. The method first fuses marker observations, inertial measurements, action history, and temporal dynamics using a multimodal task-supervised recurrent state-space model to form a predictive tracking belief; a diffusion action prior then generates multiple multi-thruster trajectories, and the best candidate is selected under a receding-horizon objective and continuously updated based on onboard observations.
The 17th installment of the Dexterous Hand Practice series focuses on the differences between piezoresistive and capacitive tactile sensor approaches, and discusses how to use them to determine contact and slip. The content addresses the tactile perception stage in dexterous hand R&D and is part of this tutorial series.
Apple's research team published the paper "Perceptive Humanoid Motion from Simulation" on arXiv, unveiling HALO, an autonomous humanoid robot control system, and validating it on the Unitree G1.
Why it matters: Apple's HALO splits task orchestration, motion generation, and tracking into three layers, and reports comparative results of 91% in unfamiliar environments and 70% on composite tasks, offering a look at its layered design and data synthesis pipeline.
REEMAN completed on-site deployment of the Fly Boat factory delivery robot (AMR) at a large garment factory in Dongguan in about half a day, and configured multiple pickup and drop-off points. The robot has a rated payload of 300kg and actually carries about 200–250kg per trip (including a custom rack of about 80kg). It lifts the rack with a 10cm autonomous lift, navigates via laser SLAM, travels about 150m per loaded trip, and performs 70–90 tasks per day on average. Workers trigger deliveries via a physical one-button call button. Reference price is about USD 5,780, depending on configuration.
AMR forklifts rely on onboard LiDAR and laser SLAM for mapping and localization, enabling autonomous path planning and obstacle avoidance without ground infrastructure such as magnetic tape, guide wires, or reflectors; route changes are usually made via software. The REEMAN Ironhide 3.0 is a 1.5-ton autonomous forklift with ±5 cm pick-and-place accuracy on 1.2 × 1.2 m pallets, 6 hours of runtime at full load and 8 hours empty, and a maximum mapping area of 40,000 m².
At the ninth China International Import Expo (CIIE), Linde Material Handling exhibited the compact pallet-handling robot L-MATIC, which demonstrated an automated workflow in unmanned operation. L-MATIC weighs 850 kg and has a load capacity of up to 1.2 tons; it features an ultra-narrow body design with built-in laser navigation and a multi-sensor fusion scheme. On site, a "threading the needle" segment, in which a thin steel wire at the front of the fork passes through a small-diameter ring, demonstrated positioning and control precision. Linde also presented KION's multimedia interactive software system, which can automatically generate suitable logistics solutions based on customer operating conditions and logistics needs.
The article reviews the core issues in robotic tactile classification, pointing out that robots need to judge material, attributes, and state from contact signals rather than simply categorizing them. It emphasizes that task definitions must distinguish between material, attributes, and state to avoid label confusion, and notes that tactile readings are affected by action conditions and must be modeled in temporal and conditional contexts.
NavGPT-3 proposes a vision-language navigation system for continuous environments, in which Harness (context management), Runtime (permission control), and NavGPT VLA (local waypoint prediction) work together to resolve the tension between long-horizon planning and high-frequency execution.
Amodal3R at ICCV 2025 performs occlusion reasoning and geometry generation directly in 3D latent space, using visible-region masks and occluded-region masks to preserve real observations and infer occluded geometry and appearance.
Stereo distance measurement calculates depth from disparity using Z=f×B/d, showing near-accurate, far-sparse characteristics: a longer baseline is better suited for long distances, and a longer focal length gives higher depth resolution. Stereo calibration consists of two steps: monocular intrinsic calibration and stereo extrinsic calibration. Extrinsic angle error is the main source of amplified error at long distances, and calibration accuracy largely determines the system's accuracy ceiling. Calibration targets should be selected by scenario: standard glass checkerboards for routine industrial inspection, dot calibration boards for high-precision micro-assembly, large-size calibration boards for large-field outdoor use, and ChArUco or AprilTag for dynamic workstations.
SUPMAST brought its "Satongxue" AGV unmanned transport robot and in-plant smart logistics solution to the 22nd China International Grain and Oil Expo in Xiamen, held November 6 to 8, with live demonstrations of fork picking, stacking, and alignment correction at Booth C123 in Hall C1. The AGV uses laser SLAM + reflective pillars + IMU fusion navigation and requires no magnetic strip guide rails; the accompanying Yunshu S7 RDS scheduling system supports a maximum concurrency of ≥20 units and task allocation latency of ≤500ms. In the Yihai Grain and Oil project, it handles a production takt of 3 minutes per pallet, reducing the error rate by more than 95% and the cargo damage rate by 94%.
Ubot's "Xingzhe Taishan" humanoid robot completed the official course as the only fully autonomous entrant in the first CMG World Robot Mount Tai Climbing Competition, taking second place. The entire climb was free of remote-control intervention, with the "brain" recognizing terrain and planning routes in real time and the "cerebellum" dynamically adjusting gait to handle uneven steps and slippery surfaces. Ubot says its commercial quadruped robot for inspection scenarios has carried out regular autonomous inspection in limited scenarios across multiple pilot projects in power and mining.
Changan Automobile announced that its first painting visual AI adhesive robot has officially gone live at the Liangjiang plant, with risk detection rate and reporting accuracy both exceeding 99.5%. The robot uses a large-model detection system built on a deep convolutional neural network (CNN) to lock in optimal adhesive application points and standardized operating trajectories, with single-inference response at the millisecond level. After clearing the PVC area bottleneck, line JPH increased by about 8%.
At CeMAT ASIA 2026, Jingsong Intelligent held a global debut of its magnetic levitation stacker crane and introduced its flying-wing loading/unloading vehicle solution for the first time. The magnetic levitation stacker crane replaces wheel-rail transmission with magnetic levitation and linear drive, achieving a maximum travel speed of 240 m/min and a lifting speed of 60 m/min; with laser ranging and magnetic-control full closed-loop positioning, it reaches ±2 mm storage and retrieval accuracy, and its dual-station design can carry 2 sets of 50 kg bins per cycle. The flying-wing loading/unloading vehicle solution relies on FMR robot cluster deployment and near-field visual scanning, and targets bidirectional synchronized loading and unloading for flying-wing carriages of multiple sizes.
PaXini has entered a strategic partnership with 58AIR, the AI industry service platform under 58.com, and 58AIR has officially become a PaXini inside ecosystem partner. The two parties will work with N data-collection companies to explore a "1+1+N" open collaboration model. PaXini will provide partners with high-precision tactile sensors and device integration support, aligning tactile, posture, and visual information by task timeline; 58AIR will open up real-world operation scenarios such as home repair and intra-city logistics. The two parties will also connect collaboration steps including collection interfaces, quality requirements, task design, and data security compliance.