Teradyne Robotics and Sereact have formed a partnership under which Sereact's Cortex physical AI platform will drive Universal Robots robot arms for unstructured tasks such as returns handling and picking; the two parties have already delivered dual-arm systems made up of hundreds of UR robot arms to customer sites in Europe.
ETH Zurich's ANYmal quadruped robot solved a 10-pound giant Rubik's Cube using only its feet, three times, taking about 8 to 9 minutes each time. The demonstration combines computer vision, a traditional Rubik's Cube solving algorithm, and reinforcement learning to teach the robot to balance, apply force, and correct errors. This "pedipulation" (manipulation with feet) could enable future robots to open doors, turn valves, or move obstacles without hands.
HD Hyundai Robotics showcased its self-developed intelligent dual-arm robot at FIX 2026 and FABTECH 2026. It replicates the human upper body, with a five-finger dexterous hand at the end of each arm, a maximum payload of 7 kg per arm and a working radius of 630 mm. It runs on the company's self-developed Robot Foundation Model (RFM) and can autonomously perceive its environment and generate motion plans.
Enact Intelligence, a physical AI company headquartered in Zurich, Switzerland, has closed a €26.68 million (about $30 million) pre-seed funding round, with investors including redalpine, Notion Capital, and firstminute Capital.
DEEP Robotics has released five research results built around its self-developed embodied brain, DEEPRobotics-Brain: the world model DR-WM-1.0, the embodied foundation model DR-VLA-2.0, the embodied agent DR-Agent, the policy optimization framework DR-APO, and the data closed-loop platform DR-Benchmark.
Ding Ning's team at Tsinghua University, together with researchers from Beijing University of Posts and Telecommunications, Peking University, and Nature Will, built HandsOn, a simulation platform and evaluation benchmark for cross-embodiment dexterous manipulation, containing 66 manually verified tasks and 18,000 human simulation teleoperation demonstrations.
Why it matters: Readers can learn why dexterous manipulation evaluation should distinguish execution from transfer, and which specific failure points current multi-task policies have on bimanual collaboration tasks.
DEEP Robotics has released five research results built around its self-developed embodied brain DEEPRobotics-Brain: the world model DR-WM-1.0, the embodied foundation model DR-VLA-2.0, the embodied agent DR-Agent, the policy optimization framework DR-APO, and the data closed-loop platform DR-Benchmark. Partners include seven universities, among them MIT, Tsinghua University, and Zhejiang University.
A research team from Harvard, NVIDIA, Berkeley, MIT and Stanford proposes RAX (Robot Agent eXploration), which connects a general coding agent to robot perception and control interfaces, uses online exploration to write successful decisions into executable programs and accumulate them as reusable skills, while keeping model parameters frozen.
Why it matters: RAX uses a frozen model plus online exploration to generate reusable programs, leading by an average of 37.4 percentage points on five simulation benchmarks with zero action demonstrations, offering a path for robot learning different from scaling data and models.
The research team proposes the RynnVLA-Latent framework, which uses RynnLAM to learn continuous, 3D-grounded latent actions from visual changes without requiring native robot control signals. The method processes 30.1K hours of human, robot, and simulation video, generating 745.52M latent action descriptors, then uses flow matching to predict latent action chunks and adapt them to specific embodiments.
Why it matters: This work converts heterogeneous interaction videos into 3D latent action supervision, offering a reusable data pathway for VLA pretraining that bypasses the scarcity of action annotations.
At IROS 2026 LRC, a team from Nanyang Technological University, Singapore, used the DEEP Robotics Tongchui M1 quadruped robot with an AgileX Robotics PiPER-X robotic arm mounted on its back (with a depth camera on the end effector), winning the Fully Autonomous Group championship and the Best Autonomous Capability Award. The PiPER-X features a spherical wrist joint design at the end effector, has a payload of 1.5 kg, and can move the camera to adjust viewing distance and angle to complete hazardous chemical label recognition. Through device feedback, coordinate calibration, and task scheduling, the team organized movement, observation, and recognition into a complete autonomous inspection process.
DexHoldem, a project involving the University of Hong Kong and Yisheng Technology, brings Texas Hold'em to real hardware, using a UR10e robotic arm and a Shadow dexterous hand to simultaneously evaluate dexterous manipulation, table-state perception, and closed-loop operation. It has been accepted to the NeurIPS 2026 Workshop.
Lingxin Qiaoshou and Xbotics have opened registration for the third dexterous hand online training camp, which starts on November 1 and runs for 8 weeks with 6–8 hours of commitment per week. The hands-on project is to recreate continuous in-hand ball rotation with the O20 in MuJoCo simulation, with O30 ball rotation as an extended example.
Authors from the Hong Kong University of Science and Technology propose World-to-Wrist VLA (W2-VLA), which uses global task context to predict future wrist latent variables and then guide fine action generation. The method achieves 98.5% average success rate on LIBERO, 60.71% and 18.21% on the Easy/Hard settings of RoboTwin 2.0, respectively, and outperforms baselines under standard and OOD conditions on three real-world tasks: tabletop cleanup, occluded placement, and dual-arm plug insertion.
STIAL has built four product series around complex grinding scenarios: AI flexible grinding robots, AI flexible grinding workstations, embodied core components, and the NextBrain™ AI brain, forming a physical AI technology loop of "perception—analysis—decision—execution."
The open-source humanoid robot Asimov 1 has entered day 363 of development, with new grippers being added to and tested on its robotic arm to give it the ability to grasp and manipulate objects. The robot is built from off-the-shelf parts and is fully open source, so developers can reproduce it themselves. The arm previously had sufficient strength; this test marks progress from mobility toward manipulation tasks.
Sereact this week reached a partnership with Teradyne Robotics (including Universal Robots and Mobile Industrial Robots) to have the Cortex physical AI platform drive UR robotic arms in handling unstructured tasks such as returns sorting and picking; the two sides say they have already delivered hundreds of UR dual-arm robotic systems at customer sites in Europe.
Eka Robotics announced this week that it has closed a $150M Series A round led by Andreessen Horowitz (a16z) and F-Prime, with the funds to be used to scale its technology and customer deployments.
DEEP Robotics has released five research results centered on its self-developed embodied brain, DEEPRobotics-Brain, including the world model DR-WM-1.0, the embodied foundation model DR-VLA-2.0, the embodied agent DR-Agent, the policy optimization framework DR-APO, and the data closed-loop platform DR-Benchmark. The company is also collaborating with seven universities, including MIT, Tsinghua, and Zhejiang University.
Why it matters: The five results span world models, VLA, policy optimization, and data benchmarks, offering a way to compare the division of labor and metrics across the full embodied brain pipeline.
At the 2026 iFlytek 1024 Developer Festival, Lingdong General released the iFlyBot-WAM industrial world action model and the LindenBot-Harness execution organization system, and debuted the LDU environmental data collection solution. The LDB embodied intelligence general-purpose robot has obtained CR certification and EU CE-MD market access, and in B2B warehouse picking has validated an operational path covering autonomous movement between shelves, recognition of goods in different poses, and flexible grasping. Its data collection, training, and inference robot LDT had previously completed validation in small-item feeding scenarios.
The authors propose a spatiotemporal retargeting method, STR, which uses a vision-language model to select relevant motion signals and body parts and identifies keyframes accordingly, allocating additional solver iterations to keyframes and higher interaction weights to the selected parts to reduce retargeting computational overhead.
On October 20, Schaeffler's Taicang embodied intelligence robot digital factory opened, with 5 ROKAE Zhuxing H series wheeled humanoid robots participating in the intelligent assembly line for tasks including bin handling, parts sorting, press-fitting and screw assembly, functional testing, marking and packaging, producing joint module components including actuators for next-generation humanoid robots.
Why it matters: The original text provides the specific processes and technical configuration of complete machines entering a mass-production line, helping readers judge the current usable boundaries of wheeled humanoid robots in real manufacturing scenarios.
At the 2026 World Physical AI Conference and Unitree Partner Ecosystem Conference, Unitree Technology founder Wang Xingxing judged that as soon as the end of this year or next year, users may be able to use a single voice command to have a Unitree humanoid robot independently complete household tasks such as tidying a room.
DiscoDemo uses reinforcement learning as a data generator, training a privileged-information latent-conditioned generator that discovers multiple successful solutions through a parallel reverse curriculum and a success-gated diversity objective, then exports them after safety filtering as a static dataset for downstream visuomotor imitation learning. On four manipulation tasks, the method raises downstream success in simulation from 40% for the strongest baseline to 62%, and on a real robot from 39% to 78%. The authors believe that the diversity of successful solutions can improve imitation learning for a pretrained flow-matching VLA, and that it complements corrective data collection.
Maxinsights has released two new products: MaxLoop, an agent-in-the-loop system that scores and corrects data in real time during collection, and MaxGlove, a sensor glove developed with Genesis AI for collecting tactile and force signals.
The KAI Halo case study EgoWild2Dex shows how high-quality bimanual dexterous manipulation capabilities can be derived from highly challenging egocentric data. The project page is https://mmlab.hk/egowild2dex.
Why it matters: The material presents a complete training path from first-person human data to robotic bimanual manipulation, along with success rates; its transferable aspect lies in its approach to data utilization.
Infineon announced the winners of its 2026 Startup Challenge. Italian robotics startup AdapTronics took first place with a grasping technology that can pick up a wide variety of objects without traditional suction cups or mechanical grippers; the technology can be applied to humanoid robots, logistics, packaging, and space robotics.
Swedish robotics startup 120ft announced it has emerged from stealth and completed a €576,000 angel funding round to develop autonomous microfactories driven by humanoid robots, serving localized small-batch manufacturing. The factory design uses three 40-foot containers, respectively handling component preparation, assembly, inspection and packing; the system combines humanoid robots, AI, virtual production line simulation and self-developed orchestration software.
MaxInsights announced a $30 million seed round led by PeakXVPartners and released two new products: MaxLoop, which can generate quality scores for each video clip the same day, and MaxGlove, used to collect tactile and force data that cameras struggle to capture. The company says it has collected real work data at natural speed in real homes, restaurants, and hotels, totaling more than 2 million hours.
Researchers from the University of Hong Kong, Yisheng Technology, and Shenzhen Hetao College have proposed YOCO, a few-shot session calibration method for 3D keypoints, aimed at correcting bias in low-cost motion-capture gloves used in dexterous-hand teleoperation. The paper has been accepted to CoRL 2026.
At the October Shanghai CIIF, Longqi Robotics showcased its self-developed wheeled humanoid robot HR-05W for the first time, along with a product matrix including SCARA robots, high-speed high-precision six-axis industrial robots, the CubeBraid intelligent loading and unloading embodied robot system, and dual-arm robots.
Embodied intelligence evaluation platform RoboColiseum released an evaluation report: Claude Opus 5.5 ranked first among four multimodal models with 76.28 points, followed closely by GPT-6 Astra with 75.99 points, while GPT-6.1 Sol and Gemini 4 Argon scored 70.91 and 66.86 points respectively; the top two on the official overall leaderboard remain specialized embodied policy models Paimon (77.54 points) and HiDream-O1-Embodied (76.92 points).
RoboCT is participating in AllPack Indonesia in Jakarta with its Elephant collaborative palletizing all-in-one machine and AGV transport cart, exchanging views with packaging, food and beverage, pharmaceutical, and personal care companies and system integrators from Indonesia and Southeast Asia on upgrade paths for end-of-line palletizing and in-plant transport.
ROBOTIS has released the HX Friends series robotic hand lineup, built on DYNAMIXEL and covering a variety of structures ranging from five-finger hands to grippers and even single-finger designs. The models include the HX5-D20 (5 fingers, 20 joints), HX4-D12 (4 fingers, 12 joints), HX3-D11 (3 fingers, 11 joints), HX2-D01 (a gripper driven by a single DYNAMIXEL), and HX1-D03 (starting from a single finger and expandable into a full hand).
Robotiq has launched the 2FC physical AI gripper, a production-ready adaptive gripper with tactile sensing and embedded intelligence. Pre-orders open today, with shipping starting in Q1 2027. It is offered in two stroke versions: 85 mm (2FC-85) and 140 mm (2FC-140).
Hunan Viewtrix Robotics Co., Ltd. exhibited a dual-machine collaborative welding workstation, a humanoid welding robot, an intelligent residual-material scanning system, an intelligent grinding workstation, and the KUNWU industrial simulation platform at the 2026 China Marine Economy Expo in Shenzhen.
Origami Robotics has released the S1 skill-capture glove, aimed at high-quality dexterous manipulation data collection. It records 20 finger joint angles via encoders and synchronously captures fingertip and palm pressure and wrist-camera video, all aligned to the same timeline, so it can keep measuring even when the hand is fully occluded and the vision pipeline fails. The original text states that its fingertip accuracy has been verified to reach the millimeter level.
PaceVLA proposes an adaptive temporal execution framework for VLAs that uses multi-granularity action decoders on a shared action expert, along with quantizability-guided mode selection and context-conditioned routing, to decide at each inference step whether to adopt coarse-grained execution. Experiments cover three multi-task simulation benchmarks and six real-world tasks, improving execution efficiency while maintaining task success rates across multiple embodiments, including long-horizon and dexterous manipulation.
The paper proposes Vector-Quantized World Model (VQ-WM), a latent feedback motion planning framework that combines discrete long-horizon search with continuous visual feedback control: a discrete latent state predictor models transitions between vector-quantized latent states, allowing the planner to search for the shortest path over long distances, while a continuous learned controller then reaches the planned states step by step through visual feedback to correct deviations and mitigate compounding errors.
The Lambda-humanoid Team released λ0, a humanoid whole-body loco-manipulation foundation model, along with HumanVerse, a 500-hour open-world human loco-manipulation dataset that pairs first-person video with synchronized body and hand motion.
The paper proposes achieving force-aware manipulation by visually predicting the deformation of a compliant Fin Ray gripper, without requiring force or tactile sensors at deployment. The method trains two models: first, a visual force estimator is trained on calibration data to annotate demonstrations with force estimates; second, an action-force proposal policy is trained to jointly generate candidate actions and their corresponding forces. At test time, candidate actions are sampled and the action whose predicted force is closest to a target from the demonstrations is executed. The method was evaluated on four tasks: berry picking, empty-can grasping, in-hand reorientation, and plug insertion.