Stair-Climbing Strategy of a Leg-Wheel Transformable Robot Using Visual Feedback and Varying Leg-Wheel Contact Points

作者：Yen-Li Lai, Wei-Shun Yu, Pei-Chun Lin · 发表于：IEEE Access · 年份：2026 · DOI：10.1109/access.2026.3651768 · 被引用次数：1 · 研究领域：Robotic Locomotion and Control、Soft Robotics and Applications、Prosthetics and Rehabilitation Robotics

Staircases represent one of the most challenging terrains commonly encountered in urban environments. This paper presents a comprehensive stair-climbing strategy for a leg–wheel transformable robot that exploits the varying contact points of its leg–wheel mechanism. The proposed strategy integrates workspace analysis, stability adjustment through center of mass shifting and rolling along stair edges, foothold planning under mechanical constraints, swing trajectory design to reduce slippage, impact, and moment of inertia, as well as gait transition from flat ground to stair climbing. Furthermore, the applicable range of stair dimensions is analyzed. A vision-based method is further developed to estimate stair dimensions in real time from depth images, enabling the robot to autonomously generate climbing behaviors. The proposed strategy is validated through a series of indoor and outdoor experiments on staircases with varying dimensions. Results showthat the robot can accurately estimate stair depth and height, and successfully climb staircases with varying step sizes. Moreover, the robot closely follows the planned trajectories throughout the climbing process. These results demonstrate the effectiveness and robustness of the proposed approach, highlighting its potential for reliable stair climbing in complex real-world environments.