Online Terrain-Aware Bipedal Gait Generation via Manifold Projection and Optimization-Guided Motion Libraries

January 1, 2026·
Junhang Lai
Junhang Lai
· 1 min read
Abstract
This work builds an online terrain-aware bipedal gait generation pipeline based on manifold projection and optimization-guided motion libraries, targeting stable humanoid locomotion over stairs, slopes, low obstacles, and mixed unstructured terrain.
Type
Publication
IEEE/ASME Transactions on Mechatronics
Status
Peer-reviewed
publications

First-author manuscript under review. Update publication metadata after acceptance.

Bipedal Walking Terrain-Aware Planning Motion Libraries Optimization
Junhang Lai
Authors
Junhang Lai (he/him)
Ph.D. Candidate in Robotics
Junhang Lai is a Ph.D. candidate at the Institute of Intelligent Robotics, Beijing Institute of Technology. His work focuses on humanoid and legged robot locomotion, motion planning and control, reinforcement learning, simulation-to-real deployment, and embodied intelligence systems.