Abstract: In order to test and evaluate the performance of autonomous navigation algorithms of orchard vehicles, this paper constructed a simulation test platform for autonomous navigation algorithms of orchard vehicles by using mathematical analysis software MATLAB. The simulation test platform built orchard planting model, light detection and ranging (LiDAR) measurement model, vehicle kinematics model, vehicle dynamics model, and motor control model to realize virtual orchard measurement and vehicle motion process simulation. The navigation results were displayed and the performance of navigation algorithm was evaluated through visualization of measurement data, navigation process and navigation deviations. The performance test results of single models and the overall platform have shown that the simulation test platform is reliable. For the navigation path generation and tracking algorithms based on a single 2D LiDAR, the simulation test platform can effectively test and evaluate the algorithm performance under various orchard planting parameters and vehicle initial states, and meet the testing requirements of autonomous navigation algorithms of orchard vehicles.

Key words: simulation test platform, autonomous navigation, orchard vehicle, LiDAR

摘要： 为测试和评价果园车辆自主导航算法的性能，采用数学分析软件MATLAB搭建面向果园车辆自主导航算法的仿真测试平台。仿真测试平台通过构建果园种植模型、激光雷达（Light detection and ranging，LiDAR）测量模型、车辆运动学模型、车辆动力学模型和电机控制模型实现虚拟果园测量和车辆运动过程模拟，通过可视化测量数据、导航过程和导航偏差显示导航结果并评价导航算法性能。对单一模型和平台整体的性能测试结果表明，仿真测试平台可靠，针对基于2D LiDAR的导航路径生成和跟踪算法，仿真测试平台能够有效测试和评价不同果园种植参数及车辆初始状态下的算法性能，满足果园车辆自主导航算法的测试需求。

关键词: 仿真测试平台, 自主导航, 果园车辆, 激光雷达