微型电脑应用2026，Vol.42Issue(1)：34-38,5.
基于深度学习的机器人移动路径规划算法
Robot Moving Path Planning Algorithm Based on Deep Learning
摘要
Abstract
In response to the low efficiency of traditional deep Q network(DQN)applied to mobile robot path planning,an im-proved DQN algorithm is proposed.By combining the environmental perception ability of deep learning with the decision-mak-ing capability of reinforcement learning,intelligent path planning for mobile robots in complex environments is achieved.The improved DQN algorithm can be used to record obstacles encountered during the learning process,enhance the reward function,and design a dynamic exploration factor function.The proposed mobile robot path planning algorithm is compared with A*,rapidly-exploring random tree(RRT),and traditional DQN algorithms,and the results show that the improved DQN algorithm effectively enhances the path planning efficiency of mobile robots.关键词
深度学习/强化学习/移动机器人/路径规划Key words
deep learning/reinforcement learning/mobile robot/path planning分类
交通工程引用本文复制引用
尉粮苹,汪可盈..基于深度学习的机器人移动路径规划算法[J].微型电脑应用,2026,42(1):34-38,5.基金项目
山东省科技厅科学支持项目(21SD09865402) （21SD09865402）
青岛恒星科技学院项目(HX2021YYYLD-PY3) （HX2021YYYLD-PY3）
