Computationally efficient predictive adaptive control for robotic operation in dynamic environments and task domains

Abstract

This paper presents a new adaptive predictive control algorithm and its refinement for robotic utility. The controller addresses the need for practical, computationally efficient, robust real-time adaptive control for multivariable robotic systems working in challenging industrial environments. It exploits a special matrix representation to obtain substantial reductions in the computational expense relative to standard methods. Controller performance is established for a simple robotic manipulator directing motion through sharply changing loading conditions and on an industrial robot loading heavy shells within the weapons magazine of a naval vessel. The new controller demonstrates the ability to adapt to varying actuator performance and rapidly changing sea states for which a classic proportional—integral—derivative controller cannot adjust. Control commands and parameter adjustments are executed in time frames suitable for real-time use, even on platforms and in environments with limited computational resources. Future work involves the implementation and testing of the controller on a prototype robot during facsimile naval operations. This work may serve as a foundation to address control issues for robots working in uncertain dynamic environments with varying task domains through the implementation of computationally efficient predictive adaptive control.