Distributed Adaptive Cooperative Tracking Control for UAV Swarm Formation With Dead-Zone Input

Abstract

This paper overcomes the failure of a UAV formation to detect a target in a limited time due to the visual sensor dead zone, thereby delaying the timing of combat. In contrast to the existing actuator failure problem, this paper also considers the influence of the sensor dead zone and obstacles on the process of tracking target. Based on the adaptive back-stepping technology, a cooperative tracking controller is designed that can track the moving targets in a limited time. First, this paper considers the visual sensor model, and then a back-stepping method is used to design a controller for each UAV, and these controllers are used to design an integrated controller for the whole system that overcomes the sensor dead zone and the control signal quantization effect. Second, by applying the approximation property of the neural network, the unknown control gain of the non-strict feedback system can be compensated, and the nonlinear quantization system of UAV actuator dead zone can realize stable dynamic characteristics in a limited time. Finally, based on the Bhat and Bernstein theory, a simulation and an experiment are conducted for comparative analysis to further verify the performance of designed controller.