基于确定性策略梯度算法的机械臂控制模型构建及仿真  

Construction and Simulation of Manipulator Control Model Based on Deterministic Strategy Gradient Algorithm

作　　者：贾红涛[1] 胡文娟[1] Jia Hongtao;Hu Wenjuan(Shangluo Vocational and Technical College,Shangluo 726000,China)

机构地区：[1]商洛职业技术学院,商洛726000

出　　处：《粘接》2021年第9期151-154,共4页Adhesion

摘　　要：为更好的实现对工业制造领域中机械臂的控制,结合当前的深度学习算法,提出一种改进奖励函数的DDPG机械臂控制方法。在该方法中,通过引入多奖励参数等方式,增强机械臂控制的灵活性,提高目标抓取的准确率。最后通过参数设置和DDPG网络模型构建,对改进方案进行验证。结果表明,该改进方式在目标抓取方面更具有稳定性。In order to better control the manipulator in the field of industrial manufacturing,combined with the current deep learning algorithm,a DDPG manipulator control method with improved reward function is proposed.In this method,multi reward parameters are introduced to enhance the flexibility of manipulator control and improve the accuracy of target grasping.Finally,through parameter setting and DDPG network model construction,the improved scheme is verified.The results show that the improved method is more stable in target capturing.

关 键 词：DDPG算法 机械臂控制 仿真 奖励参数 

分 类 号：TM359.9[电气工程—电机]