USV地形测量路径自主规划研究与应用

Research and application of path autonomous planning based on USV terrain measurement

针对未知近岸海域水上水下地形一体化测量需求,无人水面艇(Unmanned Surface Vehicle,USV)搭载导航雷达、深度计、POS MV、多波束测深仪及激光扫描仪等设备,通过设计在线地图构建及测量路径规划方法,使得USV在未知环境下自主完成地形一体化测量任务。通过离线电子海图获得待测区域初始基准地图,结合雷达、深度计等传感器实时感知的信息,采用一种基于贝叶斯估计方式的占据栅格建图方法,实现对基准地图的修正、更新,进而根据测量需求自适应调节测量路径间距。同时构建测量地图,使用基于神经元激励方法自主实现完全遍历的测量路径规划,兼顾使用A星算法避免路径规划锁死,以获得合理可行的测量路径实时规划结果。USV按照自主规划的测量路径自主航行,多波束测深仪、激光扫描仪实时测量、获取地形云数据后进行无缝拼接,完成未知近岸海域的水上水下地形一体化测量工作。

To meet the needs of integrated survey of underwater and underwater topography in unknown coastal waters, Unmanned Surface Vehicle (USV) are equipped with navigation radar, depth gauge, POS MV, multi-beam bathymeter and laser scanner, etc. By designing online map construction and survey path planning methods, USV can independently complete the integrated survey of topography in unknown environment. The initial reference map of the area to be measured is obtained by off-line electronic chart, and the reference map is revised and updated by combining the real-time sensing information of sensors such as radar and depth meter, and adopt an occupancy grid mapping method based on Bayesian estimation to realize the correction and update of the reference map. At the same time, the measurement map is constructed, and the measurement path planning based on neuron excitation method is realized independently, and the A-star algorithm is used to avoid the locking of path planning, so as to obtain reasonable and feasible real-time planning results of measurement path. The USV navigates autonomously according to the self-planned measurement path, and the multi-beam sounder and laser scanner measure in real time, obtain terrain cloud data, and then perform seamless splicing, thus completing the integrated measurement of underwater and underwater terrain in unknown coastal waters.