Purpose The purpose of this paper is to elaborate on the key process of an unmanned aerial vehicle (UAV) collaborative landing experiment, which focuses on the design of pose estimation and control pipelines and decision state machines for UAV in experiments, providing a detailed explanation of the problems encountered and their solutions. Design/methodology/approach Design a UAV pose estimation pipeline by combining sensor data such as inertial measurement unit, real-time kinematic and visual odometry. By combining Lora communication and Aruco code recognition results, communication and relative positioning between UAV and unmanned surface vessel (USV) during collaborative landing were achieved, and a decision state machine was designed based on this. Findings UAV can autonomously land on the deck of moving USV through the above pose estimation pipeline and state machine. Moreover, solutions proposed in the paper for engineering problems encountered in practical experiments have practical guidance. Originality/value This paper fully elaborates on the main parts and schemes of the collaborative landing experiment between UAV and USV, including pose estimation pipeline and state machine design. At the same time, this paper presents in detail the manifestations of engineering problems encountered in the experiment, analyzes the possible causes of the problems and provides our solutions, laying a good foundation for researchers who attempt to reproduce the experiment in the future.
Ge et al. (Thu,) studied this question.