ContactNet: Online Multi-Contact Planning for Acyclic Legged Robot Locomotion

Abstract-- In legged locomotion, online trajectory optimization techniques generally depend on heuristic-based contact planners in order to have low computation times and achieve high replanning frequencies. In this work, we propose ContactNet, a fast acyclic contact planner based on a multioutput regression neural network. ContactNet ranks discretized stepping regions, allowing to quickly choose the best feasible solution, even in complex environments. The low computation time, in the order of 1 ms, makes possible the execution of the contact planner concurrently with a trajectory optimizer in a Model Predictive Control (MPC) fashion. Online motion planning for legged robots remains a challenging Further, automatically navigating terrain with constraints problem. The common approach is to use optimization such as stepping stones is generally not possible with such algorithms in a Model Predictive Control (MPC) approaches. When complex motions are desired, the user is loop to automatically generate trajectories based on sensor then usually forced to design a contact plan suitable for the feedback [1], [2], [3], [4].