基于在线支持向量机的无人机航路规划技术

Path Planning of UAVs Based on Online Support Vector Machine

研究了一种基于在线支持向量机的无人机航路规划方法,以保证无人机在完成任务时候能以最小的被发现概率以及最短航程安全到达目标点。首先建立多约束的无人机航路规划数学模型,并进行分析。应用A*算法产生初始航迹获取样本数据,在此基础上应用在线支持向量机具有在线训练、模型精确度高、需要样本少、泛化能力强等特点,实现无人机航路优化。最后将所研究的方法应用于无人机的航路规划仿真,仿真结果表明所研究的基于在线支持向量机的无人机航路规划方法是有效的。

A path planning method for Unmanned Aerial Vehicles （UAV） was proposed based on the onlineSupport Vector Machine （SVM） to ensure UAVs reach the destination safely with the minimum probability of being found and through the shortest path. Firstly, the mathematic model was established and analyzed for the path planning of UAV considering the various constraints. The initial path was given by using A ＊ algorithm to obtain the sample data for UAV. On the basis of the initial path, the online SVM, which has the features of online training, high precision model, small training sample and strong approximation ability, was employed to optimize the path of UAV. Finally, the path planning based on the online SVM was used to the simulation of path planning for UAV. The simulation results proved the effectiveness of the proposed method.