Distributed Optimal Self-Deployment Control of Multiple Dynamic Systems subject to Various Uncertainties and Communication Constraints

Over the past two decades, cooperative control of multiple dynamic systems, also called multi-agent systems, has drawn increasing interest in the control and robotics communities. This is motivated by many complicated tasks in a wide range of real-world applications, including environmental monitoring, search and rescue, and precision agriculture, which cannot be easily accomplished by a single system. Those complicated tasks can often be completed by networked multiple dynamic systems through their cooperation/coordination. One important and interesting topic in this area is distributed optimal self-deployment control of multi-agent systems, also known as distributed coverage control, the objective of which is to achieve optimal coverage performance by properly deploying agents in a given mission field through distributed control strategies that use only certain information from the agent itself and its neighbors.   Numerous researchers, including the PI, have devoted themselves to the problem of distributed optimal self-deployment control for multi-agent systems over the past two decades. However, most existing research assumes that networked agents can always obtain accurate measurements and knowledge of terrain roughness. In addition, each agent in practical multi-agent systems is often equipped with simple hardware components with limited energy and communication resources, resulting in a limited operational lifetime, limited communication bandwidth and range, and transmission delays. Unfortunately, few studies in the open literature on distributed optimal self-deployment control for multi-agent systems deal with measurement errors and unknown terrain roughness, especially when they are also subject to limited energy and communication resources. Thus, it is critical to take into consideration those uncertainties as well as energy and communication limitations in the development of distributed optimal self-deployment control strategies. This project aims to investigate several fundamental problems in distributed optimal self-deployment control under various real-world uncertainties and constraints, including measurement errors, unknown terrain roughness, limited energy resources, limited communication bandwidth and range, and transmission delays. To overcome the challenges arising from those uncertainties and practical constraints, this project will develop novel distributed optimal self-deployment control strategies to achieve the optimality and stability of the resulting closed-loop multi-agent systems. It is expected that the results of this project will provide solutions to several fundamental and critical challenges in distributed optimal self-deployment control of multi-agent systems, thus facilitating their real-world applications.