Robust Control of OpenMANIPULATOR-X Using Reinforcement Learning

This study introduces an innovative method for enhancing the control of the OpenMANIPULATOR-X robot, integrating Fixed-Time Sliding Mode (FTSM) control with the Deep Deterministic Policy Gradient (DDPG) algorithm. The main objective of this research is to minimize positional inaccuracies and mitigate chattering that can occur during the robot’s operation. By utilizing the DDPG algorithm, a reinforcement learning (RL) agent is trained to optimize the control gains for the FTSM technique. The simulation results presented in the study highlight the effectiveness of this approach in reducing both position errors and chattering. Incorporating advanced state and reward selection in the RL framework is essential for optimizing the robot’s performance. Overall, this methodology shows promise for improving the precision and reliability of the OpenMANIPULATOR-X, particularly in uncertain environments.