Related Experiment Video
Updated: May 10, 2025
08:19
Asymmetric Walkway: A Novel Behavioral Assay for Studying Asymmetric Locomotion
Published on: January 15, 2016
8.7K
Efficient Hybrid Environment Expression for Look-and-Step Behavior of Bipedal Walking
Chao Li1, Qingqing Li1, Junhang Lai1
1School of Mechatronic Engineering, Beijing Institute of Technology, Beijing 100081, P. R. China.
Cyborg and Bionic Systems (Washington, D.C.)
|April 24, 2025
Summary
This study introduces an efficient perception method for biped robots, enabling safe look-and-step behavior. The approach uses feasible planar regions and a heightmap for navigation, ensuring obstacle and foot collision avoidance.
Area of Science:
- Robotics
- Computer Vision
- Artificial Intelligence
Background:
- Biped robots require efficient environmental perception for look-and-step locomotion.
- Limited computing resources pose a challenge for real-time environmental analysis.
Purpose of the Study:
- To develop an efficient method for extracting environmental features for biped robot navigation.
- To enable safe and effective look-and-step behavior in biped robots.
Main Methods:
- A hybrid environment representation using feasible planar regions and a heightmap.
- Efficient planar region extraction leveraging organized point structures for nearest neighbor searches.
- Exclusion of collision-prone areas to ensure robot body safety.
Main Results:
- The proposed method achieves perception in 0.16 seconds per frame using only a CPU.
- Demonstrated efficiency and safety in experiments with BHR-7P and BHR-8P robots.
- Successfully enables look-and-step behavior by preventing body and foot collisions.
Conclusions:
- The developed method is suitable for real-time perception in biped robots.
- The hybrid representation enhances navigation safety and efficiency.
- Validated effectiveness for look-and-step locomotion in complex scenarios.
