Deep Reinforcement Learning-Based Control Framework for Multilateral Telesurgery

The upper boundary of time delay is often required in traditional telesurgery control design, which would result infeasibility across regions. To overcome this issue, paper introduces a new framework based on deep deterministic policy gradient (DDPG) reinforcement learning (RL) algorithm. developed effectively overcomes the phase difference and data loss caused by delays, facilitates restoration surgeon’s intention interactive force. Kalman filter (KF) employed to blend multiple surgeons’ commands predict final local commands, respectively. ensures synchronization tracking performance transparency. Prior knowledge therefore not required. Simulation experiment results have demonstrated merits proposed framework.