基于MIDE个体寻优的采摘机器人轨迹规划仿真
Simulation of Harvesting Robot Trajectory Planning Based on MIDE Individual Optimization
王腾 1皮大能2
作者信息
- 1. 长江大学文理学院信息与机电工程学院,湖北 荆州 434020
- 2. 湖北师范大学电气工程与自动化学院,湖北 黄石 435002
- 折叠
摘要
为了提高移动式采摘机器人的作业效率,实现高效率高精度的采摘作业,提出改进DE算法下移动式采摘机器人轨迹规划方法.建立移动式采摘机器人的动力学模型,获取采摘机器人的动力学特性;根据该特性,建立基于双目视觉的针孔成像模型完成采摘机器人果实采摘经过的路径点收集;将得到的路径点作为种群个体,对DE算法实施改进,选取一种基于多种群移民的差分进化算法(MIDE)展开个体寻优并输出最优解,实现移动式采摘机器人的最优轨迹规划.实验结果表明,所提方法具有较高的收敛能力和搜索效率,能够有效提高机器人的工作效率和稳定性.
Abstract
In order to improve the operational efficiency of mobile harvesting robot and achieve efficient and high-precision harvesting operations,this paper put forward a trajectory planning method for mobile harvesting robots based on improved DE algorithm.Firstly,we constructed a dynamical model of mobile harvesting robot to obtain the dynamic characteristics.Based on this characteristic,we built a pinhole imaging model based on binocular vision to collect the path points that the harvesting robot passed through.Using these path points as population individuals,we improved the DE algorithm,and selected a differential evolution algorithm based on multiple immigrations(MIDE)to perform the individual optimization and output the optimal solution,thus achieving the optimal trajectory planning of mobile harvesting robots.The experimental results show that the proposed method has high convergence ability and search efficiency,and can effectively improve the work efficiency and stability of robot.
关键词
移动式采摘机器人/动力学模型/双目视觉/轨迹规划Key words
Mobile harvesting robot/Dynamical model/Binocular vision/Improved DE algorithm/Trajectory planning引用本文复制引用
出版年
2024
NETL
NSTL
万方数据