“…In recent years, the application of deep reinforcement learning [ 3 , 4 , 5 , 6 , 7 , 8 ] in the robot field [ 9 , 10 ] has deepened and has been widely used in grasping [ 11 , 12 ], assembly [ 13 ], path planning [ 14 , 15 ], and other fields [ 16 , 17 ]. A few scholars have used deep reinforcement learning to study the constant force-tracking process, showing the great potential for applying deep reinforcement learning to solving the issue of constant force-tracking.…”