ETH Zürich Develops Humanoid Robot Capable of Climbing and Recovering Using ACP
- Published
- Source
- HumanoidsDaily
Researchers at ETH Zürich's Robotic Systems Lab have successfully demonstrated a Unitree G1 robot capable of climbing ladders, overcoming obstacles, and recovering from falls using a single locomotion policy. This innovative approach, known as Attention-based Contact Planning (ACP), emphasizes optimal body placement against the environment, enabling the robot to determine movements effectively.
The significance of this research lies in its potential applications in various fields, including search and rescue operations and advanced manufacturing. By utilizing reinforcement learning and human motion data, the system learns to predict contact positions for nine body parts, enhancing the robot's adaptability and efficiency in navigating complex terrains.
ETH Zürich's project, titled Humanoid Mobility via Attention-based Contacts, is currently under review after being submitted to arXiv on October 22. The research team aims to further refine the robot's capabilities, with ongoing developments in the perception system and contact planning to improve real-time responsiveness and operational efficiency.
Original report
ETH Zürich Teaches a Humanoid to Climb, Vault and Recover by Planning Body Contacts
HumanoidsDaily · [email protected] (Humanoids Daily Staff)This briefing is an independently written summary based on publicly available reporting and is provided for industry information and news discovery. The original report and source publication are credited and linked where applicable. RobotToday does not claim ownership of third-party source material.
Rights concerns? Contact [email protected] with the relevant URL and details. We will review the matter and take appropriate action where warranted.
More related news
-
1
ROBOTTODAY Weekly October 19 – 23, 2026 RobotToday · 2 days ago
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
- 11
- 12
Related Suppliers
NVIDIA Robotics
United States
NVIDIA robotics division delivering Isaac simulation and training tools, Jetson edge compute modules, and Omniverse digital twin frameworks for physical AI.
Deci AI (Acquired by NVIDIA)
Israel
Israeli deep learning optimization startup acquired by NVIDIA in May 2024 for ~$300M; known for AutoNAC neural architecture search technology.
Acheron Labs
Austria
Vienna-based robotics software startup building an LLM-driven control layer that lets Unitree's G1 humanoid and Go2 quadruped follow spoken commands.
Humanola
United States
US start-up founded in 2025 offering VR teleoperation software for humanoids and robot arms and turning each operator session into robot-learning training data.
Palit Microsystems
Taiwan
Taipei-based Palit has manufactured NVIDIA-based graphics cards since 1988 and now also builds Jetson Orin edge-AI compute modules for robotics.
SINTRONES Technology Corp. 鑫創電子股份有限公司
Taiwan
Taiwan-listed (TPEx: 6680) maker of rugged Edge AI computers for transportation, AGV/AMR, industrial automation, and defense; ISO 9001/27001/22163 certified.
Wuhan Langyi Robotics Co., Ltd. 武汉朗毅机器人有限公司
China
Wuhan startup supplying embodied spatial-intelligence navigation modules that let humanoid and mobile robots move autonomously.
Honda
Japan
Tokyo-based Honda Motor, founded 1948, runs a robotics program that built ASIMO and now develops the Avatar Robot and UNI-ONE mobility device.
Zhejiang Jitai Intelligent Technology Co., Ltd. 浙江吉泰智能科技有限公司
China
TRI (Zhejiang Jitai, est. 2023) builds wall-climbing robots for boiler, coal-conveyor, and wind-turbine tower and blade inspection at power plants.
Luoyang BECOT Scientific Development Co., Ltd. 洛阳百克特科技发展股份有限公司
China
Luoyang mine-hoist friction-drive specialist (NEEQ: 837067) whose magnetic flux leakage climbing robot inspects stay cables of cable-stayed bridges.
CNRS-AIST Joint Robotics Laboratory (JRL), IRL3218
Japan
French-Japanese CNRS-AIST lab in Tsukuba researching humanoid autonomy, motion planning and control on HRP-5P, HRP-4CR and Kawasaki RHP Friends.
Wild Robotics
Uruguay
Palletizing software on ROS 2 for system integrators: configure, simulate and deploy Fanuc and Universal Robots cells without custom robot code.
Robotics needs a service framework.
RSF defines a common language for robot service capability, lifecycle operations, certification pathways, and service-provider networks.
Join the RobotToday community on LinkedIn
Daily robotics news, in-depth analysis, conference highlights, and discussions with professionals worldwide.
