Position control of a planar cable-driven parallel robot using reinforcement learning

Abstract

This study proposes a method based on reinforcement learning (RL) for point-to-point and dynamic reference position tracking control of a planar cable-driven parallel robots, which is a multi-input multi-output system (MIMO). The method eliminates the use of a tension distribution algorithm in controlling the system’s dynamics and inherently optimizes the cable tensions based on the reward function during the learning process. The deep deterministic policy gradient algorithm is utilized for training the RL agents in point-to-point and dynamic reference tracking tasks. The performances of the two agents are tested on their specifically trained tasks. Moreover, we also implement the agent trained for point-to-point tasks on the dynamic reference tracking and vice versa. The performances of the RL agents are compared with a classical PD controller. The results show that RL can perform quite well without the requirement of designing different controllers for each task if the system’s dynamics is learned well.

Research on workspace visual-based continuous switching sliding mode control for cable-driven parallel robots

“…Moreover, closed-loop control methods can converge the errors caused by uncertainty. Various control methods are used in CDPRs, including PID [17, 18], robust control [19], reinforcement learning [20], neural networks [21], and adaptive control [22]. Sliding mode control (SMC) is an effective method for controlling nonlinear and uncertain systems and has considerable robustness to external disturbances and uncertain dynamics modelling [23].…”