Fast Replanning Slosh-Free Trajectory for Mobile Manipulation Using Model Predictive Control
Fang-Wei Chen, Feng‐Li Lian · 2024
In this paper, a model predictive control (MPC) framework is proposed to deal with slosh-free liquid transportation task with a mobile manipulator (MM). Two variations of slosh-free trajectroy using MPC with a spherical pendulum model are presented, which are the feedforward and feedback method respectively. The simulation experiments were conducted using a MM combined with a tracked mobile base and a 6-DoF robotic manipulator. The simulated slosh suppression results showed the feedback version of the proposed method goes beyond the existing Trajectory Optimization (TO) based approach by 44% reduction in Root Mean Square Error (RMSE), providing a fast re-planning solution for generating a suitable slosh-free motion reference trajectory for a whole-body controlled MM.