Chapter Four - User experience-enhanced teleoperation control

Abstract

In this chapter, several new types of EMG signal-based variable stiffness control based on tremor suppression and VF human–robot interactive control are proposed for a teleoperation robot system. Focusing on the characteristics of the tremor signal, a linear learning weight function is designed. In the process of tremor filtering, the SVM filter can effectively use small samples to filter the tremor, and the variable stiffness control can achieve personalized control according to muscle activity. The combination of tremor filtering and personalized control methods can improve the control performance of remote operation. In addition, combined with the hybrid control of variable stiffness control and the VF program, the activity of the operator's hand muscles can be effectively adjusted, providing natural and user-friendly HRI and improving the operating level. Additionally, a controller with a variable stiffness scheme is designed. By utilizing the proposed controller, integrating DOB technology, the RBFNN algorithm, and the variable stiffness strategy, the trajectory tracking performance of the robot manipulator is better than that of previously mentioned controllers. Model dynamics uncertainty problems are out of account, since the proposed combination of DOB and RBFNN reduces the reliability of dynamic nonlinear models. Finally, the experimental results show that in the case of tremor filtering, the teleoperation robot system can improve the tracking control performance, and the control stiffness of the teleoperation system can adapt to changes in the operator's muscle activity.