面向大型机场草坪的割草机器人路径规划及轨迹跟踪控制研究

为了提高割草机器人的工作效率及环境适应能力，基于移动机器人平台设计了一种既受遥控操作又能自主运行的适用于大型机场草坪作业的割草机器人。首先，运用高精度差分GPS（global positioning system，全球定位系统）采集机场草坪边界和障碍物的位置信息，根据采集的信息将机场草坪分为最少数目的凸多边形工作区域；考虑到割草机器人无法原地无半径转弯，在传统迂回式路径规划算法的基础上提出一种往返直线型路径规划算法，并在凸多边形路径规划区内推导出遍历路径的显示方程表达式。其次，运用高精度差分GPS测得割草机器人实际轨迹并与规划轨迹对比，设计了一种区间判断型轨迹纠偏算法；以执行电机的PID（proportion integration differentiation，比例积分微分）控制和区间判断型轨迹纠偏算法构造割草机器人双闭环轨迹跟踪控制器，对按传统迂回式路径和往返直线型路径行进的割草机器人进行轨迹跟踪仿真分析。最后，以自制的割草机器人为例，按往返直线型路径运行方式进行样机实验。仿真结果发现：当割草机器人跟踪当前路径到达终点后会自动调头跟踪下一条路径，验证了轨迹跟踪算法的稳定性；传统迂回式路径运行方式下割草机器人的漏割率较高，达到46.42%，而往返直线型路径运行方式下其漏割率为7.15%，明显优于传统迂回式路径仿真结果。样机实验测得的漏割率为8.89%，与仿真实验结果一致，表明所设计的轨迹跟踪算法对大型机场草坪作业割草机器人是适用的。研究结果可为大型机场草坪割草机器人的开发提供理论指导。

Research on path planning and trajectory tracking control of mowing robot for large airport lawn

In order to improve the work efficiency and environmental adaptability of mowing robot, a mowing robot for large airport lawn was designed, which could operate remotely and operate autonomously based on the mobile robot platform. Firstly, high precision differential GPS（global positioning system）was used to collect the location information of airport lawn boundary and obstacle. The airport lawn was divided into a minimum number of convex polygonal working areas according to the collected information. Considering that the mowing robot couldn’t turn around without radius in situ, a round-trip straight path planning algorithm based on traditional circuitous path planning algorithm was proposed, and the analytic equations of path in the convex polygon working area were derived. Secondly, the actual trajectory of the mowing robot was measured by high-precision differential GPS and compared with the planned trajectory, and a interval judgment trajectory correction algorithm was designed. The double closed-loop trajectory tracking controller was constructed by PID control of executive motor and interval judgment trajectory correction algorithm. Trajectory tracking simulation experiment of a mowing robot running along the traditional roundabout path and the straight-line roundabout path was carried out. Finally, taking the self-made lawn mower as an example, the prototype experiment was carried out according to the round-trip straight path running mode. It was found that the mowing robot would automatically turn its head to track the next path when the mowing robot tracked the current path to the bound, which verified the stability of the trajectory tracking algorithm. The missing cutting rate of mowing robot was 46.42% under traditional circuitous path running mode, while the missing cutting rate was 7.15% under round-trip straight path running mode, which was obviously better than the simulation result of traditional circuitous path. The missing cutting rate measured in the prototype experiment was 8.89%, which was consistent with the simulation results. It showed that the proposed trajectory tracking algorithm was effective for the mowing robot. The research results can provide theoretical guidance for the development of lawn mowing robot for large airport lawn.