基于机器学习的采摘机器人动作执行效率优化研究  

Research on the Optimization of Action Execution Efficiency of Picking Robot Based on Machine Learning

作　　者：杨玉霞[1] Yang Yuxia(Jiyuan Vocational and Technical College, Jiyuan 454650, China)

机构地区：[1]济源职业技术学院,河南济源454650

出　　处：《农机化研究》2022年第6期223-226,231,共5页Journal of Agricultural Mechanization Research

基　　金：河南高等教育教学改革研究与实践项目(2017SJGLX160)。

摘　　要：为了提高采摘机器人动作的执行效率,提高动作执行的控制精度,将机器学习算法引入到了采摘机器人控制系统的设计上,利用机械学习和归纳学习设计了采摘机器人的机器学习方法,并利用PID反馈调节对学习过程进行了优化,从而得到了效率较高的机器人采摘动作执行系统。以相同时间内果实的采摘量为作业任务,对不同机器学习方法的机器人采摘果实数目进行了仿真模拟,结果表明:采用机械与归纳学习方法和PID反馈调节方式,相同时间内可以采摘更多的果实,有效地提高了采摘机器人动作的执行效率。In order to improve the execution efficiency and control precision of the action of the picking robot,it introduced the machine learning algorithm into the design of the control system of the picking robot.And it designed the machine learning method of the picking robot by using the mechanical learning and inductive learning,and the learning process is optimized by using the PID feedback regulation,thus it obtained the robot with higher efficiency with removing the action execution system.Taking the picking amount of fruits in the same time as the task,it carried out the simulation of the number of fruits picked by robots with different machine learning methods.The simulation results showed that more fruits can be picked in the same time by adopting the mechanical and inductive learning method and PID feedback adjustment method,which can effectively improve the execution efficiency of the picking robot action.

关 键 词：采摘机器人 动作执行 机器学习 PID反馈调节 仿真模拟 

分 类 号：S225.39[农业科学—农业机械化工程] TP242[农业科学—农业工程]