基于卷积神经网络的无人驾驶农用车辆路径优化研究  

Research on Path Optimization of Driverless Agricultural Vehicle Based on Convolutional Neural Network

作　　者：赵小娟 Zhao Xiaojuan(Luoyang Open University,Luoyang 471000,China)

机构地区：[1]洛阳开放大学,河南洛阳471000

出　　处：《农机化研究》2024年第7期257-261,共5页Journal of Agricultural Mechanization Research

基　　金：洛阳市社会科学规划项目(2022B382)。

摘　　要：为了提升无人驾驶农用车辆的经济性和安全性,介绍了卷积神经网络的理论,搭建了障碍物模型和障碍物碰撞判断方法,并建立了基于卷积神经网络的无人驾驶农用车辆路径学习与优化模型,实现了无人驾驶农用车辆路径学习和优化。实验结果表明:系统可以实现对无人驾驶农用车辆的路径优化,实现了无碰撞的最优路径规划,可以满足无人驾驶农用车辆的经济性和安全性需求。In order to improve the economy and safety of driverless vehicles,it first introduces the theory of convolutional neural network,then builds the obstacle model and obstacle collision judgment method,and establishes the path learning and optimization model of driverless vehicles based on convolutional neural network,which realizes the path learning and optimization of driverless vehicles.The experimental results show that the system can realize the path optimization of driv-erless vehicles.From the perspective of walking path length and turning conditions,the system realizes the collision free optimal path planning,which proves that the system can meet the economic and security requirements of driverless vehi-cles.

关 键 词：无人驾驶农用车辆 卷积神经网络 障碍物 路径优化 无碰撞 

分 类 号：S219.89[农业科学—农业机械化工程] TP274[农业科学—农业工程]