结合混合粒子群算法的植保无人机航线设计方法

Route Planning Method for Plant Protection Unmanned Aerial Vehicles Combined with Hybrid Particle Swarm Optimization

随着农业航空技术的发展,自主化作业方式成为了目前农业航空领域的研究热点.植保无人机是一种集成了通信技术、自动控制技术、传感技术以及地理信息定位技术等多种相关技术的智能农业设备,用植保无人机进行喷雾作业具有效率高、速度快等优点,并且无人机能够在没有跑道的小型区域内垂直起降,能够很方便的在各种地形上进行喷洒作业.针对含障碍作业区域,提出了一种无人机航线设计算法,算法分为路径点采样以及路径点排序两部分.首先使用栅格法来获得路径点,用路径点表示作业区域的特征信息,并结合混合粒子群算法来对路径点进行排序,用路径点的排列来表示无人机飞行路径,设计出一条能够有效规避障碍物并对工作区域完成全覆盖的航线.实验结果表明本文提出的无人机航线设计方法能够应用在多种含障碍工作区域内,目标区域包含单个或者多个障碍物时,本文所提出的航线设计方法均能设计出一条合理的无人机航线,并有效减少无人机航线的转弯次数,降低无人机飞行时的能耗.

With the development of agricultural aviation technology,the autonomous operation mode is a hi-tech research focus in the field of agricultural aviation.Plant protection unmanned aerial vehicle is an intelligent agricultural equipment that integrates various related technologies such as communication technology,automatic control technology,sensing technology and geographic information positioning technology.Spraying pesticides on crops is an effective method,with several advantages,including high speed and high efficiency.The unmanned aerial vehicle can take off and land vertically on a small area without a runway,making it easy to spray on various terrains.With the development of satellite positioning technology,the autonomous flight operation mode based on global positioning system navigation and capable of automatically generating drone operation routes according to the target area has become the main development trend of plant protection drones.Research on the route planning for unmanned aerial vehicles is particularlynecessary.A route planning algorithm based on hybrid particle swarm optimization was proposed for the obstacle-containing operation area.The route planning algorithm was divided into path point sampling and path point sorting.In our unmanned aerial vehicle route planning method,the path points on work area were sampled based on grid method and all effective path points were obtained.The unmanned aerial vehicle route planning problem was transformed into a sorting problem for the path points.The sorting problem for the path points was a combinatorial optimization problem that could be solved using an intelligent optimization algorithm.The path points were sorted based on Hybrid Particle Swarm Optimization.The number of turns of the path was used as a criterion for evaluating the meritsofthe route.Comparative simulation experiments were performed in six different work areas.The experimental results showed that the number of turns of the route designed by the route planning method proposed in this paper were 36,34,27,40,46 and 37 times respectively.The number of turns of the route designed by the comparison method were 60,60,60,50,54 and 46 times respectively.A route which could effectively keep away from all obstacles and complete full coverage of work area was designed.The experimental results show that the proposed unmanned aerial vehicle route planning method can be applied on a variety of work areas with single or multiple obstacles.Routes of unmanned aerial vehicle are planned before operation,and then let the drone operate according to the specified route.The energy and pesticide consumption generated by unmanned aerial vehicle can be estimated in advance.Operating in this mode can save the time required for preparation and increase the efficiency of unmanned aerial vehicle operations.As the popularity of unmanned aerial vehicle increases,the operation and management of unmanned aerial vehicle becomes more and more important.The route planning method proposed in this paper is applicable to automatic unmanned aerial vehicle and can be widely used in precision agriculture.