Advanced control of network interconnected systems for high performance collaborative tasks

This project targets on achieving high performance collaborative tasks which are highly demanding, in application areas such as aerospace, military applications, manufacturing, transportation and in health care services. Advanced control of network interconnected systems for collaborative tasks, such as multi-agent systems and multilateral teleoperation systems, addresses one of the recognized major integrated challenges and opportunities facing control researchers. Previous research in our group has been successful on developing various robust nonlinear control approaches for remotely controlled systems in which one communication loop is involved in the system. Instead of remotely operating a single system, this proposal is to deal with a system of subsystems interconnected over networks. Note that existing work focuses on scalar agent without dynamics or point-mass model only, centralized control architecture and the entire system operated in a reliable network environment. There are many issues have not been well addressed such as dynamics, coordination mechanism to environment changes or disturbance, imperfect sensors, communication constraints, and performance evaluation index. The short term goals of this proposal are to: build a framework for the control and robust estimation strategies of interconnected systems over networks for high performance collaborative tasks; develop a new theoretical control architecture for multiple teleoperation systems with experimental studies; apply the theoretical work to multiple Pioneer mobile robots and car-like robotic vehicles in my lab addressing various constraints in sensors, communications and kinematics/dynamics. The proposed research program covers the theoretical development on the decentralized control design for interconnected systems over networks and its application to multiple collaborative autonomous robotic systems, and the advanced control of multilateral teleoperation systems with collaborative tasks.