Design of admittance controller with sliding mode based on disturbance observer for elbow joint actuated by pneumatic muscles

Purpose Active compliance control is the key technology for Tri-Co robots (coexisting-cooperative-cognitive robots) to interact with the environment and people. This study aims to make the robot arm shake hands compliantly with people; the paper proposed two closed-loop-compliant control schemes for the dynamic identification of cascade elbow joint. Design/methodology/approach The active compliance control strategy consists of inner and outer loops. The inner loop is the position control using sliding mode control with disturbance observer (SMCDO), in which a new saturation function is designed to replace the traditional signal function of sliding mode control (SMC) law so as to mitigate chatter. The outer loop is the admittance control to regulate the dynamic behaviours of the elbow joint, i.e. its impedance. The simulation is carried out to verify the performance of the proposed control scheme. Findings The results show that the chatter of traditional SMC can be effectively eliminated by using SMCDO with this saturation function. In addition, for the handshake task, the value of threshold force and elbow joint compliance is defined. Then, the threshold force tests, impact tests and elbow-joint compliance tests are carried out. The results show that, in the impedance model, the elbow joint compliance only depends on the stiffness parameters, not on the position control loop. Practical implications The effectiveness of the admittance control based on SMCDO can improve the adaptability of industrial manipulator in different working environments to some degree. Originality/value The admittance control with SMCDO completed trajectory tracking has higher accuracy than that based on SMC.