基于仿鹰眼视觉的无人机自主空中加油

Biologically eagle-eye-based autonomous aerial refueling for unmanned aerial vehicles

自主空中加油是提高无人机作战能力的重要技术支撑,而鹰在所有的动物中以视觉最为敏锐而著称。针对无人机自主空中加油中的精确对接需求,设计了基于仿鹰眼视觉原理的无人受油机视觉测量方法及半物理试验平台。采用鹰眼视觉注意机制提取虚拟视景中的兴趣区域,剔除噪声和干扰。在对受油机上的颜色标志点进行识别的基础上采用P3P算法,对加油机和受油机之间的相对位姿进行精确估计,并设计了基于比例-积分-微分的硬式加油伸缩管控制律。基于上述技术,在实验室环境下设计开发了无人机自主空中加油半物理仿真平台。实验结果表明:所设计的仿鹰眼视觉测量方法和自主控制律能满足无人机自主空中加油的性能要求,具有较强的实时性、准确性和鲁棒性。

Autonomous aerial refueling is one of the important supports for the improvement of unmanned aerial vehicle （UAV） combat capacity, and the eagle is famous for its sharp eyesight. Aiming at the accuracy requirement during the docking phase of UAV autonomous aerial refueling （AAR） , the UAV visual measurement methods and a hardware-in-loop simulation platform were designed based on eagle-eye vision mechanism. The visual attention mechanism of eagle eye was adopted to extract the region of interest in virtual scene and perform saliency detection and noise removal. The relative pose of the tanker and unmanned aerial vehicle receiver was measured accurately based on the colored marker recognition of the UAV using P3P algorithm. Based on the above techniques, the autonomous control law of the flying boom was designed with the proportional-integral-differential （PID） controller. The hardware-in-loop simulation platform for unmanned aerial vehicle autonomous aerial refueling was developed in laboratory environment. Series of hardware-in-loop experiment results demonstrate the feasibility and effectiveness of the proposed approach in the biologically eagle-eye vision simulation platform.