Balancing Control and Pose Optimization for Wheel-legged Robots Navigating Uneven Terrains.
Year: 2021
Type: preprint
Abstract: In this paper, we propose a novel approach on controlling wheel-legged quadrupedal robots using pose optimization and force control via quadratic programming (QP). Our method allows the robot to leverage wheel torques to navigate the terrain while keeping the wheel traction and balancing the robot body. In detail, we present a rigid body dynamics with wheels that can be used for real-time balancing control of wheel-legged robots. In addition, we introduce an effective pose optimization method for wheel-legged robot's locomotion over uneven terrains with ramps and stairs. The pose optimization utilized a nonlinear programming (NLP) solver to solve for the optimal poses in terms of joint positions based on kinematic and contact constraints during a stair-climbing task with rolling wheels. In simulation, our approach has successfully validated for the problem of a wheel-legged robot climbing up a 0.34m stair with a slope angle of 80 degrees and shown its versatility in multiple-stair climbing with varied stair runs and rises with wheel traction. Experimental validation on the real robot demonstrated the capability of climbing up on a 0.25m stair with a slope angle of 30 degrees.
Source: arXiv (Cornell University)
Authors: Junheng Li, Junchao Ma, Quan Nguyen
Institution: University of Southern California
Language: English
Subfield: Control and Systems Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W3198916782