Huaner Technology offers four 18-DOF hexapod robots—miniHexa, SpiderPi, JetHexa, and ROSpider—for different learning needs, all supporting 360° omnidirectional movement and inverse kinematics algorithm learning.
The Ruiman Wind Fire Wheel (WBOT11B) chassis has an open-source SDK and API, and can be used for mobile robot algorithm validation in university laboratories and MVP development of commercial service robots for enterprises, eliminating the work of building low-level drivers and motion control from scratch. This content takes ROS secondary development and navigation deployment as its thread and provides a deployment guide for the above scenarios.
REEMAN autonomous forklifts perform a task-level handshake with turntable stretch wrappers via an external wireless module: the forklift places the pallet and, after retreating from the rotating area, sends a start signal; the wrapper completes rotary film wrapping and automatic film cutting, then returns a completion signal, after which the forklift re-identifies and aligns to pick up the load.
REEMAN automated forklifts integrate with WMS, MES, ERP, and WCS through open APIs and SDKs. Business systems issue transport requests, the fleet scheduler assigns tasks, and the robots execute them using laser SLAM navigation and report completion status back.
The author has open-sourced the claude-hardware-in-the-loop method and a Claude Code skill, enabling Claude Code to complete circuit board, firmware, sensor and motor selection, and CAD design, and to repeatedly iterate and validate in an Isaac Sim or Unreal virtual twin.
The author wrote open-source firmware for the Feetech STS32XX series servos, claiming it nearly eliminates positioning error in most cases. SO-ARM101 positioning error dropped from 35 mm to 10 mm, and SO-ARM102 from 25 mm to 0–2 mm. The firmware is open-sourced on GitHub.
The article notes that humanoid robots are moving from demonstrations to mass production, but the manipulation success rate drops sharply from 89.4% in simulation environments to 12% in real home scenarios, and reliability must be supplemented by a testing system. Integrated joint modules account for about 40% of the BOM cost of the complete machine, and GB/T 43200-2023 has provided a testing framework in four categories: motor performance, control performance, positioning accuracy, and mechanical performance. NI uses a cRIO joint test bench and a hardware-in-the-loop simulation solution based on Isaac Sim, PXI, and RDMA to try to insert an HIL stage between pure simulation and real machines, narrowing the gap in migration from simulation to reality.
Sunzhan Technology released the fully customized AI computing power platform Axon Platform in Shenzhen on October 20. Built on a RISC-V plus DSA foundation, the hardware integrates scalar cores, vector units, a dual-source tensor NPU array, and a NoC, with computing power ranging from 0.1T to 128T.
Tsinghua University and UC Berkeley jointly proposed and open-sourced uni-robot-agent, a cross-environment unified embodied agent harness for connecting tool interfaces across different robots and simulation environments. Based on this framework, Qwen3.8-Flash-Next scored 31.39 on the RoboDojo benchmark, exceeding the publicly reported 28.97 for GPT-6 Astra, at about 1/20 of Astra's cost.
KUKA Mobile Robotics unveiled three core offerings at CeMAT ASIA 2026: the KMP L series underride lifting mobile robots, a one-stop pallet handling solution, and the KUKA.AMR software 3.0 platform, as the implementation of the "KAI" intelligent platform in warehousing and logistics scenarios.
Xinyan Robotics launched the Bubbo 1 emotional companion robot for the home on October 29, 2026. It is 65 cm tall, carries 26 sensors across 13 categories, is priced at 9,499 yuan on JD.com, and is available to the first 1,000 users at a discounted price of 7,999 yuan.
At CeMAT 2026, Geek+ demonstrated a robotic arm equipped with the Gravity embodied model and its general-purpose humanoid robot Gino 1, completing grasping of unfamiliar objects, picking, and bin handling. Gravity uses a hybrid Transformer (MoT) dual-brain collaborative architecture and a hierarchical memory mechanism, and through natural-language correction it raised the overall success rate of complex manipulation tasks from 17.5% to 87.5%, with a single picking action stable at about 2 seconds.
At the 2026 Tencent Global Digital Ecosystem Summit, Tencent Cloud showcased AI Infra deployments in embodied intelligence. Manifold Space used a cluster of 512 domestic chips to train stably, improving training performance nearly 3x in about two weeks. Machine Science built a unified data lake in Nanjing with COS, enabling cross-region direct reads of the same data between Beijing and Nanjing and saving tens of millions of yuan that would have gone to building high-performance storage in both locations. Zeroth used COS to handle PB-scale data across multiple campuses in Shanghai, North America, and Europe, with MetaInsight helping engineers quickly extract high-value segments.
The author used the RKNN profiler to complete a full end-side optimization, reducing latency from 41 ms to 13 ms. The material provides only an abstract and does not give the specific optimization steps or the hardware platform used.
The article explains ROS2 TF2 coordinate frame management, extending from transform tree principles to multi-sensor calibration in practice. The material provides only the title and summary; it does not give specific method steps or calibration results.
Intel will hold an Embodied Intelligence and Robotics Solutions Exchange in Shanghai on November 10, showcasing a multi-channel GMSL real-time perception solution and a 358H embodied intelligence complete system based on the Intel Core Ultra X7 358H (Panther Lake) platform.
After merging with ATOMIX, Force Machine made its debut at CeMAT ASIA 2026 with a robot multi-agent tiered operation system that divides case-piece picking into three tiers by difficulty: economical robotic arms handle about 60% of standard goods, embodied robots handle about 20% of irregular, hard-to-grasp goods, and the remaining 20% of long-tail tasks are covered by human workers.
REEMAN developed an enclosed food delivery robot based on the Hussar Pro chassis (HBOT20P), combining a fully enclosed delivery compartment with indoor autonomous mobility. The chassis has a rated payload of 100 kg, uses laser SLAM navigation, avoids obstacles with laser, 3D camera, and ultrasonic sensors, and offers open SDK/API interfaces supporting HTTP API, Android SDK, RS232, and local MQTT integration.
With annual shipments of thousands of mobile robots and chassis units, industrial-grade engineering standards, cross-application customization experience, and a self-developed technology chain spanning navigation algorithms to cloud management, REEMAN provides OEM and custom robot development services. Its public projects include a warehouse picking robot using the Big Dog chassis, a 7-DOF robotic arm, and a 1,200 mm lifting mechanism; an AI library inventory robot based on the Hot Wheels chassis; and an enclosed food delivery robot based on the Hussar Pro HBOT20P chassis.
Hiwonder's three native ROS 2 mobile manipulation platforms, LanderPi, JetRover, and ROSOrin Pro, all include SLAM navigation, 3D vision, and multimodal AI examples; they differ mainly in compute power, perception hardware, chassis, and robotic arm torque.
Using python-control as the vehicle, the article breaks down state-space control design, covering controllability and observability criteria, pole placement, the algebraic Riccati solution for LQR, the dual observer, and frequency-domain margins. It also deploys discrete LQR and an energy-based swing-up strategy into a MuJoCo cart-inverted-pendulum closed loop.
The article analyzes GTSAM factor graph state estimation, introduces how pose graph and IMU preintegration factors are constructed, and conducts a comparison experiment between batch LM and iSAM2 incremental solving. It also analyzes the process by which loop closure detection triggers global correction, as well as numerical validation of incremental inference on the Bayes tree.
Gu Yue Ju (Ancient Moon Residence) publishes the seventh installment of the Dexterous Hand in Practice series, themed on connecting the CAN bus to a single finger and covering three parts: protocol, wiring, and transceiving. The material provides only the title and summary, without specific protocol details, wiring schemes, or transceiving implementation.
Part 06 of the Dexterous Hand in Practice series discusses how STM32, ESP32, and Raspberry Pi divide responsibilities in dexterous hand control. The content centers on the roles each of these three types of control boards plays, and it is a tutorial-style article in this hands-on series.
This practical guide to Zephyr stepper motor control covers three features: microstepping, positioning, and event callbacks. It explains how to use microstep driving, position control, and the event callback mechanism for stepper motors.
Rockchip and SigmaStar reported a combined net profit of 1.174 billion yuan in Q3 2026, with Rockchip at 473 million yuan for the quarter, up 90.49% year on year, and SigmaStar at 701 million yuan, up 752.76% year on year. Rockchip's RK3588, RK3576 and other chips handle robot visual recognition, 3D mapping and path planning, and the company has launched a multi-sensor data acquisition solution; SigmaStar shipped more than 10 million robot vision chips in 2025 for a 23% global share, and its SSU9366 targets outdoor and companion robots.
REEMAN published three verified AMR forklift pallet-handling deployments: in Huzhou, Ironhide 3.0 moves 1.5-ton pallets along a route of about 47 meters, with daily throughput of 160–200 trips; in Jingmen, Ironhide passes through two electric doors and interfaces with a stretch-wrapping machine, at about 75 cycles per day; in Shanxi, a Rhinoceros 1.5-ton stacker handles high-level handoffs in the packaging area.
Stardust Intelligence founder Lai Jie said the company does not compete with general large models; instead, through Stardust OS (AOS), it packages perception, mobility, manipulation, and interaction into callable Meta capability packages and Skills, enabling models such as GPT and DeepSeek to divide the work and collaborate with its self-developed Lumo embodied foundation model.
Geek+ announced the "24/7 Autonomous Intelligent Warehouse Plan," establishing embodied intelligence and full-warehouse intelligence as its next strategic direction, driving warehouses toward 7×24-hour unmanned autonomous operation and continuous evolution. The plan relies on end-to-end robotic solutions and a warehouse brain software platform comprising RMS, WES, and IOP, focusing on three areas: end-to-end process integration, strengthening intelligent closed loops, and large-scale deployment. Geek+ says it already has a global large-scale commercial deployment and localized operations system, but the autonomous capabilities in the plan remain development goals.
The REEMAN Fly Boat AMR achieves cross-floor material transport through an elevator integration module: the robot can autonomously call, enter, ride, and exit elevators without human assistance. The model has a rated payload of 300 kg and autonomous jacking of 10 cm, uses LiDAR SLAM navigation, has positioning accuracy of ±5 cm (±3 cm with reflectors, ±2 cm with QR codes), and a travel speed of 0.3–1 m/s.
REEMAN autonomous forklifts use laser SLAM navigation and can connect to existing WMS, MES, and ERP systems without magnetic strips or QR codes; the integration focus is on location mapping and open API connectivity rather than rebuilding navigation infrastructure. The guide covers site surveys, map creation, task triggering, and multi-vehicle fleet scaling. Each model is rated for 6 hours of continuous operation (fully loaded)/8 hours (empty), supports automatic return to charging, and some models allow manual battery swapping.
A developer combined a $250 SO-101 robot arm with Meta's Muse Spark and SAM 3.1 to find, grasp, and hand over a screwdriver on voice command, with no training required. The project was built for a Meta AI hackathon, and the demo video was posted on X.
The WeChat Pay Smart Operations Robot made its in-store debut at Guangdong 7-ELEVEN's 2,000th store (at Tencent's Shenzhen headquarters campus), handling greeting, shopping guidance, and ordering, with customers scanning to pay themselves after choosing. Built on Tencent Hunyuan, the robot uses its multi-turn dialogue and Agent capabilities to assist with daily store sales, payment collection, and operations, and features the Tencent Tianlai inside audio solution to hear users clearly in noisy stores.
At WTCC 2026, Arm China's Jishu Community released the open-source AIOS platform JishuFriends, along with its first prototype robot dog Puppy, and JishuBuddy, an AI hardware development tool for Agentic AI.
Geek+ announced the "24/7 Autonomous Intelligent Warehouse Plan," with embodied intelligence and whole-warehouse intelligence as its two main technical thrusts, driving warehouses toward 7×24-hour unmanned autonomous operation and continuous evolution. The plan relies on end-to-end robotic solutions, a warehouse brain software platform (including RMS, WES, and IOP), and a global large-scale delivery system, focusing on three directions: connecting the entire process, strengthening the intelligent closed loop, and achieving large-scale implementation. Its Gravity unified embodied intelligence framework and Gravity 4D embodied world model are used to improve autonomous execution capabilities in complex physical tasks and abnormal scenarios.
Geek+ has released the "24/7 Autonomous Intelligent Warehouse Plan," advancing 24/7 unmanned autonomous operation and continuous evolution of warehouses along two main technical lines: embodied intelligence and whole-warehouse intelligence. Its end-to-end robotic solution relies on the self-developed Gravity unified embodied intelligence framework and the Gravity 4D embodied world model, while the warehouse brain platform consists of three major modules: RMS, WES, and IOP. The plan focuses on three directions: connecting the entire process, strengthening the intelligent closed loop, and achieving large-scale deployment.
In Lecture 7 of the "Twenty-One Lectures" series, the Xbotics Embodied Intelligence Lab converts the target object pose obtained in Lecture 6 into gripper actions, linking the grasp-and-place loop with five poses—pre-grasp, grasp, lift, place, and retreat—and a state machine.
Gu Yue Ju analyzes the MuJoCo real-time predictive control framework, which uses a residual cost to unify task descriptions, with four planners—predictive sampling, gradient descent, iLQG, and hybrid ILQS—replanning at every step within the same loop. A Cartpole comparison experiment reports the convergence speed and per-step computational cost of the four.
The article documents unsupported operator issues encountered when running VLA models on RK3588 and compiles them into a list and troubleshooting log. The material provides only the title and summary; specific operators, model versions, or performance data are not given.
Autel Robotics launched the AAGI (Autel Air-Ground Intelligence) solution brand in Shenzhen, and together with Daohe Tongtai and Saifang Technology introduced the AIP foundation along with three industry application platforms: EIS 2.0, Avant Hub 2.0, and Skyshield 2.0.