Distributed Cooperative Tracking Algorithm Based on Finite Time Neurodynamics

Abstract

In the rapidly growing field of multi-robot system applications, this study focuses on exploring the challenges associated with multi-robot cooperative tracking of time-varying targets. The core objective of the research is to achieve efficient cooperative target tracking by minimizing the distance between the robots and the target, facilitated by the exchange of local information among the robots. To address this challenge, a novel distributed finite-time algorithm is proposed, integrating predictive correction methods and sliding mode control techniques. On a theoretical level, a carefully constructed Lyapunov function is developed, and a comprehensive convergence analysis is conducted to ensure that the proposed algorithm can successfully solve the time-varying optimization problem within a finite time. To validate the practical effectiveness of the algorithm, multiple rounds of robot-target tracking experiments are conducted. The experimental results vividly demonstrate the efficiency of multi-robot cooperative tracking, showing that the performance remains unaffected by the number of obstacles or the time-varying target's movement trajectory, thereby confirming the algorithm’s outstanding effectiveness and reliability in real-world applications.