학술논문
Robust and Collision-Free Formation Control of Multiagent Systems With Limited Information
Document Type
Periodical
Source
IEEE Transactions on Neural Networks and Learning Systems IEEE Trans. Neural Netw. Learning Syst. Neural Networks and Learning Systems, IEEE Transactions on. 34(8):4286-4295 Aug, 2023
Subject
Language
ISSN
2162-237X
2162-2388
2162-2388
Abstract
This article investigates the collision-free cooperative formation control problem for second-order multiagent systems with unknown velocity, dynamics uncertainties, and limited reference information. An observer-based sliding mode control law is proposed to ensure both the convergence of the system’s tracking error and the boundedness of the relative distance between each pair of agents. First, two new finite-time neural-based observer designs are introduced to estimate both the agent velocity and the system uncertainty. The sliding mode differentiator is then employed for every agent to approximate the unknown derivatives of the formation reference to further construct the limited-information-based sliding mode controller. To ensure that the system is collision-free, artificial potential fields are introduced along with a time-varying topology. An example of a multiple omnidirectional robot system is used to conduct numerical simulations, and necessary comparisons are made to justify the effectiveness of the proposed limited-information-based control scheme.