| 摘要: |
| 针对无人机(unmanned aerial vehicle, UAV)/无人艇(unmanned surface vehicle, USV)对海面区域的最优动态覆盖观测问题, 提出了一种以最大化观测收益为指标的航路优化方法。采用区域分解、子区域分配、航路规划相结合的分层求解思路: 首先, 根据信息密度等先验知识, 采取基于高斯混合模型(Gaussian mixture model, GMM)的区域特征提取理论, 从任务区域中提取出若干个子区域; 然后, 将子区域进行排序与分配, 从而将复杂的协同优化问题转化为多个简单的单UAV或USV航路规划问题;最后, 各UAV或USV在分配的子区域内采用并行滚动时域控制(receding horizon control, RHC)算法进行航路规划。仿真结果表明, 本文提出的GMM-RHC 方法具有更高的观测效率, 可有效解决无人机/无人艇最优动态覆盖观测问题, 具有重要的应用价值。 |
| 关键词: 无人机 无人艇 最优动态覆盖观测 航路优化 |
| DOI：10.11759/hykx20171011015 |
| 分类号: |
| 基金项目:山东省自然科学基金(ZR2018BF016); 中国博士后科学基金资助项目(2017M622278); 中央高校基本科研业务费(201713046) |
|
| Optimal dynamic coverage for UAV/USV surveillance |
|
YAO Peng,QI Sheng-bo,LI Ming
|
| Abstract: |
| In this paper, we propose a path optimization method with the objective of maximum surveillance to solve the problem of achieving optimal dynamic coverage in the surveillance of the sea surface by unmanned aerial vehicles (UAVs) and unmanned surface vehicles (USVs). We propose a layer-based framework that includes regional decomposition, subregion allocation, and path planning. First, based on the previously obtained information density, we utilize the Gaussian mixture model (GMM) to extract regional features with respect to several subregions. Next, we sequence these subregions and allocate them to UAVs or USVs to simplify the complex cooperative optimization problem into several single-UAV/USV path planning problems. Then, we plan each agent’s path using the concurrent receding horizon control (RHC) method. Simulation results indicate that the proposed GMM–RHC method achieves higher surveillance efficiency. |
| Key words: unmanned aerial vehicle (UAV) unmanned surface vehicle (USV) optimal dynamic coverage for surveillance path optimization |
