Safety control system technologies for UAVs: review and prospect
Lei Guo, Xiang Yu, Xiao Zhang, Youmin Zhang
Abstract
Unmanned aerial vehicles (UAVs), considered as platforms for AI technology, play an important role in national security and economic development. With the growing degree of task complexity and the development of anti-UAV technologies, the UAV-incident rate is increasing significantly. Therefore, it is highly desirable to improve UAV safety through anti-disturbance and resilience control. This paper reviews existing studies upon disturbance estimation and fault diagnosis, anti-disturbance control, fault-tolerant control, and task reconfiguration. In addition, technical challenges and potential solutions are presented, including an overall architecture of a UAV safety-control system. In terms of research focus, particular attention must be paid to UAV-characteristics recognition, causality and trace-to-the-source analysis, quantitative analysis of ability, intelligent detection in a local loop, safety control in an overall loop, and task-control-optimization design. In the future, advanced control algorithms must be transferred to the software, chips, and systems of safety control, allowing the proposed control scheme to satisfy major practical needs and applications.