Obstacle Avoidance Control for Unmanned Tracked Vehicle Formation With Communication Constraints

In this paper, aiming at the ground battlefield environment with communication constraints and many unknown obstacles, the autonomous obstacle avoidance control problem of unmanned tracked vehicle formation is studied. Innovative solutions are proposed to address the challenges of poor dynamic adaptability of obstacle avoidance, high communication dependence, and easy interruption of communication links that exist in traditional methods in the dense scene of unknown obstacles. Firstly, the adaptive formation method is designed and the leadership dynamic transfer mechanism is constructed to enhance the system’s responsiveness to unexpected obstacles by reconfiguring the formation topology in real time. Secondly, a state space modelling method based on perceptual field is proposed, which effectively reduces the communication demand within the formation and improves the convergence speed of the algorithm through the encoding of environmental features and fusion of local information. Finally, a communication constraint model is constructed to evaluate the communication link state in combination with the channel quality, and dynamic rewards are generated in real time to motivate the formation to maintain smooth communication. Based on the above innovations, an adaptive formation communication constraint perceptual field-based multi-agent proximal policy optimization is proposed, and the effectiveness of this method is evaluated through simulation experiments. Experimental results show that the proposed method has strong generalization ability and robustness, and can complete the autonomous obstacle avoidance control task of unmanned tracked vehicle formation with communication constraints. The research results provide theoretical support and technical realization paths for unmanned system formations to collaborate in complex environments. Note to Practitioners—The purpose of this paper is to solve the problem of autonomous obstacle avoidance of unmanned tracked vehicle formation with communication constraints, but it can also be applied to other types of unmanned system formation. The existing unmanned system formation obstacle avoidance control methods are usually rule-based and assume that the communication state is good, which limits its wider application. In this paper, a new method of adaptive formation obstacle avoidance control is proposed. The decision of formation is made only according to the finite state in the perceptual field, which greatly reduces the dependence on communication. Furthermore, the motion mode of unmanned tracked vehicle is described mathematically, and the control scheme is designed based on intelligent algorithm. The new control method fully considers the communication constraints and the robustness of the formation. Preliminary simulation experiments show that this method is feasible, but it has not been tested in physical systems. In future studies, we will conduct physical experiments and verify the effectiveness of the proposed method in more types of control tasks.