机载视觉探测与目标定位系统设计与实现

Design and development of airborne visual detection and target localization system

摘要:

为了解决GPS(Global Position System)定位精度不够高,无法满足无人机完成货物搬运任务的问题,设计并实现了一种以TMS320DM6437为平台的机载视觉探测与目标定位系统,由机载子系统和地面监视子系统组成。机载视觉信息处理子系统利用圆的几何特性改进了经典的Hough变换圆检测算法,在保证精度的同时大大降低了计算量,提高了算法运行效率。同时设计并实现了区分多个相同目标的方案与算法,其冗余机制增强了方案的可靠性。最后搭建了静目标定位测试平台和动目标定位跟踪机载测试系统,验证了系统的可靠性、鲁棒性、实时性和探测精度。

Abstract:

In order to solve the problem that the GPS(Global Position System)’s low positioning accuracy which can’t meet the needs of UAV(Unmanned Aerial Vehicle)’s cargo handling task. Designed and developed an airborne visual detection and target localization system based on TMS320DM6437, which is composed of an airborne subsystem and a ground monitoring subsystem. The airborne visual information processing subsystem modified the typical Hough transform method taking advantage of the geometric properties of circles, so that the computing burden is decreased significantly while the detection accuracy is preserved, which improved the algorithm efficiency. A framework to distinguish multiple targets in an image was also designed and developed, and a proposed redundancy mechanism enhances its reliability. To verify and demonstrate the reliability, robustness, fastness and accuracy of the proposed algorithm and system, a set of experiment platforms were constructed for both static target localization and moving target tracking tasks.