A Disturbance-Aware Trajectory Planning Scheme Based on Model Predictive Control

Despite the development of numerous trajectory planners based on computationally fast algorithms targeting accurate motion of robots, the nowadays robotic applications requiring compliance for interaction with environment demand more comprehensive schemes to cope with unforeseen situations. This letter discusses the problem of online Cartesian trajectory planning, targeting a final state in a desired time interval, in such a way that the generated trajectories comply with the tracking abnormalities due to considerable motion disturbances. We propose a planning scheme based on Model Predictive Control. It utilises a novel strategy to monitor the tracking performance via state feedback and consequently update the trajectory. Also, it ensures the continuity of the generated reference while accounting for realistic implementation constraints, particularly due to computational capacity limits. To validate the efficacy of the proposed scheme, we examine a practical robotic manipulation scenario in which a given task is executed via a Cartesian impedance controller, while an external interaction interrupts the motion. The performance of the proposed strategy as compared to that of a state-of-the-art study is demonstrated in simulation. Finally, a set of experiments verified the effectiveness of the proposed scheme in practice