采摘机器人自主采摘学习能力研究—基于LM-BP神经网络

Research on Learning Ability of Picking Robot for Autonomous Picking Based on LM-BP Neural Network

为了让采摘机器人更加快速和准确地识别目标水果的成熟度,提出了一种基于BP神经网络的自主学习方法。由于BP神经网络容易陷入局部最优值且训练效率较低,因此进行了改进,实现了LM-BP神经网络算法。测试结果表明:与BP神经网络算法相比,LM-BP神经网络算法训练学习速度更快,测试精度更高,能够满足采摘机器人对目标水果成熟度识别精度的要求,具有一定的应用价值。

In order to make the picking robot recognize the ripeness of the target fruit more quickly and accurately,this paper proposes an autonomous learning method based on BP neural network,but because BP neural network is easy to fall into the local optimal value and the training efficiency is low,this paper makes some improvements and realizes the LM-BP neural network algorithm.The test results show that compared with BP neural network algorithm,LM-BP neural network algorithm has faster training speed and higher testing accuracy,and can meet the requirements of the picking robot for the target fruit maturity recognition accuracy.It has a certain practical value in future applications.