Optimization-based locomotion planning, estimation, and control design for the atlas humanoid robot
Scott Kuindersma,Robin Deits,Maurice Fallon,Andrés Valenzuela,Hongkai Dai,Frank Permenter,Twan Koolen,Pat Marion,Russ Tedrake
DOI: https://doi.org/10.1007/s10514-015-9479-3
IF: 3.255
2015-07-31
Autonomous Robots
Abstract:This paper describes a collection of optimization algorithms for achieving dynamic planning, control, and state estimation for a bipedal robot designed to operate reliably in complex environments. To make challenging locomotion tasks tractable, we describe several novel applications of convex, mixed-integer, and sparse nonlinear optimization to problems ranging from footstep placement to whole-body planning and control. We also present a state estimator formulation that, when combined with our walking controller, permits highly precise execution of extended walking plans over non-flat terrain. We describe our complete system integration and experiments carried out on Atlas, a full-size hydraulic humanoid robot built by Boston Dynamics, Inc.
robotics,computer science, artificial intelligence