Abstract
In earlier chapters, a practically achievable posture controller was developed for human-like movement of redundant DOF, based on effort minimisation. In this chapter, safety considerations regarding physical human–robot interaction (HRI) lead to the integration of the posture scheme into dynamic model-free, adaptive compliance control for task motion. This chapter explains the combination of the adaptive control scheme enhanced by the posture controller and an anti-windup (AW) compensator. The AW compensator avoids instability issues when control signal amplitudes reach their amplitude constraints and saturate. Lyapunov stability proofs of the task controller, the adaptive controller and AW compensator are provided to show the convergence of the scheme, implying safety in HRI. A discussion of the posture controller’s stability is provided in the light of passivity theory. The effectiveness of this proposed scheme is confirmed by simulation and real-world experimental results for one-dimensional and multidimensional cases of compliance control.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
