Cooperative Target Enclosing Control of Nonholonomic Dynamic Systems with Practical Constraints and Uncertainties

Cooperative control of multiple dynamic systems, also known as multi-agent systems, has been a central topic in control and robotics due to its significant potential in realworld applications, including search and rescue, and environmental monitoring. Cooperative target enclosing control, in particular, has drawn increasing attention for tasks such as spatial mapping, data acquisition, and support and escort operations. The objective of cooperative target enclosing control is to drive a group of mobile agents to track and surround a static or moving target while maintaining a prescribed formation pattern.Various cooperative target enclosing control strategies have been developed for different types of agent dynamics, such as integrator-type and Euler-Lagrange systems in recent years. However, most existing works do not fully consider practical constraints and uncertainties, which are inevitable in real-world applications. For example, practical systems such as fixed-wing drones, differential drive ground robots, and underwater gliders are subject to nonholonomic constraints due to their inherent motion limitations.Practical multi-agent systems are also often subject to distance constraints due to limited interaction ranges and potential collisions between neighboring agents. In addition, mobile robots might be subject to velocity constraints, including saturated heading rates and bounded maximum and minimum velocities, due to physical limitations such as pitch rate limits and thrust limits. All these constraints presentsignificant challenges in cooperative target enclosing control of multi-agent systems.Furthermore, in scenarios such as patrol and escorting, the assumption that target motions are fully known is often invalid, leading to uncertainties about target dynamics.Recent literature reveals that cooperative target enclosing control problems of multiagent systems under nonholonomic constraints, distance constraints, and velocity constraints, along with uncertainties about target dynamics, have not been adequatelyaddressed.This project will investigate the problems of cooperative target enclosing control for nonholonomic multi-agent systems, focusing on those practical constraints and uncertainties. Specifically, this project will investigate three problems: (i) static target enclosing control for nonholonomic dynamic systems with distance constraints; (ii) moving target enclosing control for nonholonomic dynamic systems with both distance and velocity constraints; and (iii) moving target enclosing control for nonholonomic dynamic systems when there are uncertainties about target dynamics. This project, by addressing these problems and challenges, will develop novel robust and adaptive distributed target enclosing control strategies. The expected outcomes of this project will enhance the safety and performance of multi-agent systems in cooperative target enclosing control tasks, greatly facilitating their practical applications in real-world scenarios.