Purpose This study aims to realize natural and effort-saving motion behavior and improve effectiveness for different operators in human–robot force cooperation. Design/methodology/approach The parameter of admittance model is identified by deep deterministic policy gradient (DDPG) to realize...