2021/04/21 by Min Dai, Xiaobin Xiong, Dai, Min +4 · 2 citations
Computer Science · Engineering · Mathematics · #Angular momentum #Artificial intelligence #Bipedalism #Classical mechanics #Computer science #Control (management) #Control theory (sociology) #Controller (irrigation) #FOS: Computer and information sciences #FOS: Electrical engineering #Geology #Geometry #Ground reaction force #Inverted pendulum #Kinematics #Mathematics #Momentum (technical analysis) #Nonlinear system #Pendulum #Physics #Quadratic equation #Robot #Robotic Locomotion and Control #Robotics (cs.RO) #Soil Mechanics and Vehicle Dynamics #Sports Dynamics and Biomechanics #Systems and Control (eess.SY) #Underactuation #cs.RO #cs.SY #eess.SY #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2104.10367
published in arXiv (Cornell University) (Cornell University)
openalex publication_date 2021/04/21 · arxiv created 2021/09/23 · arxiv updated 2021/09/27 · openalex created_date 2022/10/03 · openalex updated_date 2026/07/28
This paper presents an online walking synthesis methodology to enable dynamic and stable walking on constrained footholds for underactuated bipedal robots. Our approach modulates the change of angular momentum about the foot-ground contact pivot at discrete impact using pre-impact vertical center of mass (COM) velocity. To this end, we utilize the underactuated Linear Inverted Pendulum (LIP) model for approximating the underactuated walking dynamics to provide the desired post-impact angular momentum for each step. Desired outputs are constructed via online optimization combined with closed-form polynomials and tracked via a quadratic program (QP) based controller. This method is demonstrated on two robots, AMBER and 3D Cassie, for which stable walking behaviors with constrained footholds are realized on flat ground, stairs, and randomly located stepping stones.