计算机应用与软件2025，Vol.42Issue(4)：279-288,10.DOI:10.3969/j.issn.1000-386x.2025.04.040
机械臂深度强化学习降维快速训练方法
FAST TRAINING METHOD OF DEEP REINFORCEMENT LEARNING DIMENSIONALITY REDUCTION FOR MECHANICAL ARM
摘要
Abstract
Aimed at the problem that the training cycle of the deep reinforcement learning algorithm is too long when it performs full degree of freedom training for manipulator in 3 D environment,a fast training method of deep reinforcement learning for manipulator is proposed.By decomposing the grasping task,the training of the lateral steering gear and the longitudinal steering gear of the manipulator was decoupled,and the solution space was compressed by dimensionality reduction,which simplified the training process while ensuring the execution accuracy of the action.The deep deterministic policy gradient(DDPG)algorithm was improved,and the secondary value estimation was performed on the same batch of samples to delay the updating of the strategy network,supplemented by preferential experience replay,which effectively improves the training efficiency of DDPG algorithm.Experimental results show that the proposed method has the characteristics of low training complexity,high speed and low cost,and the success rate of grasping can reach 98％,which is beneficial to the application and promotion of industrial occasions.关键词
深度强化学习/机械臂/深度确定性策略梯度/目标抓取/降维Key words
Deep reinforcement learning/Mechanical arm/Deep deterministic policy gradient/Target capture/Dimension reduction分类
信息技术与安全科学引用本文复制引用
王敏,王赞,李珅,陈立家,范贤博俊,王晨露,刘名果..机械臂深度强化学习降维快速训练方法[J].计算机应用与软件,2025,42(4):279-288,10.基金项目
国家自然科学基金项目(61901158) （61901158）
河南省科技厅重点研发与推广专项(202102210121) （202102210121）
河南省科技发展计划项目(科技攻关)(212102210500) （科技攻关）
开封市重大专项(20ZD014) （20ZD014）
开封市科技项目(2001016) （2001016）
开封平煤新型炭材料科技有限公司项目(2021410202000003). （2021410202000003）
