Safe and reliable dynamic coverage control for multi-agent systems

Coverage control;multi-agent systems;swarming;collective behavior;cooperative control;autonomous systems;collision avoidance;proximity maintenance

Atinc, Gokhan

关键词: Coverage control;    multi-agent systems;    swarming;    collective behavior;    cooperative control;    autonomous systems;    collision avoidance;    proximity maintenance;   

Others  :  https://www.ideals.illinois.edu/bitstream/handle/2142/49682/Gokhan_Atinc.pdf?sequence=1&isAllowed=y

美国|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

In this dissertation, we study the dynamic coverage problem for multi-agent systems, where the main objective of a group of mobile agents is to explore a given compact region. Qualitatively, the coverage goal can be described as gathering sensory information for each point in the compact domain up to a desired level. In order to achieve the coverage goal, we propose two different novel control schemes. In supervised coverage control, we introduce a stationary supervisor that assists a group of coverage agents with the centralized coverage control input and the global trajectory tracking control input.The coverage control input ensures the coverage task is performed until the agents end up in local minima, and when they do, the global trajectory tracking control input ensures that the agents are deployed to uncovered regions.Our control scheme is designed such that the two control inputs are decoupled, meaning that only one of them is active at a given time. In addition to the coverage objective, we design control inputs for coverage agents for avoiding collisions and maintaining proximity to the supervisor.In swarm-based coverage control, we consider groups of coverage agents that behave as swarms for completing the coverage task.Unlike the supervised coverage control, there is no stationary agent; all agents move as a group in order to explore a given domain.Moreover, contrary to the supervised coverage control scheme, there is no trajectory tracking; instead, agents are deployed to uncovered regions via swarming control where a leader agent selects a target position that is in an uncovered region while all other agents are commanded to swarm around the leader agent's target position. In this scheme, coverage and swarming control inputs are also decoupled, meaning that only one of them is active at a given time.In addition to the coverage objective, we design control inputs for coverage agents for avoiding collisions with each other and static obstacles and maintaining proximity to each other.For both control schemes, we introduce a smooth transition signal that enablesthe coverage agents to continuously transition between thecoverage control and global trajectory tracking control in the case of supervised coverage control scheme and between thecoverage control and swarming control in the case of swarm-based coverage control scheme. Through the decoupling of these control inputs, we attain simpler control problems that we analyze separately for different modes and provide stability results for the overall control schemes through Lyapunov-like analysis. Finally, to illustrate the effectiveness of our schemes, we provide numerical simulations for various scenarios.