针对四旋翼飞行器系统中图像识别与航迹规划问题,提出了一种基于视觉的智能识别规划方法,并搭建了基于Linux嵌入式控制的飞行实验平台.在控制过程中,搭载云台相机并结合OpenCV库函数对相机捕捉到的画面进行实时图像处理,实现无人机的目标识别与定位;通过改进遗传算法对无人机飞行路径进行合理规划,使其完成从起点到终点并途经所有中间点而路径最合理的任务;结合Guidance视觉传感导航系统中的超声波数据,利用获取的近地面高度信息实现飞行器精确定位.最后,通过实验验证了该方法的有效性.%An autonomous identification and planning system is proposed;and then a flight platform based on Linux embedded control system is developed for a quadrotor unmanned aerial vehicle(UAV) based on vision.In the process of control,through the Zenmuse X3 camera and the OpenCV library for real-time image processing,to achieve the UAV target identification and localization.By improved genetic algorithm to plan a reasonable path for the UAV,the mission from the starting point to the end point is completed.The ultrasonic data from the Guidance is acquired;and the UAV accurate position can be realized by the acquired nearground height information.Finally,the proposed approach is verified by experiments.
展开▼
