arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2107.11245cs.ROcs.AI
收藏

基于双DQN的复杂环境下机器人路径规划改进算法

An Improved Algorithm of Robot Path Planning in Complex Environment Based on Double DQN

Shanghai Jiao Tong University(上海交通大学);Northwestern Polytechnical University(西北工业大学)

Fei Zhang, Chaochen Gu, Feng Yang

摘要:

根据我们的实验,深度Q网络(DQN)在应用于具有许多困境的环境中的路径规划时存在一些局限性。奖励函数可能很难建模,并且在经验回放中很难找到成功的经验转换。在此背景下,本文提出了一种改进的双DQN(DDQN)来解决这个问题,参考A* 和快速探索随机树(RRT)。为了实现丰富的经验回放实验,基于RRT策略重新定义了机器人在每轮训练中的初始化。此外,根据A* 中位置成本的定义,为加速学习过程,专门设计了对空闲位置的奖励。仿真实验结果验证了改进的DDQN的有效性,机器人能够成功地学习避障能力和最优路径规划能力,而DQN和DDQN对避障能力和最优路径规划能力没有影响。

英文摘要:

Deep Q Network (DQN) has several limitations when applied in planning a path in environment with a number of dilemmas according to our experiment. The reward function may be hard to model, and successful experience transitions are difficult to find in experience replay. In this context, this paper proposes an improved Double DQN (DDQN) to solve the problem by reference to A* and Rapidly-Exploring Random Tree (RRT). In order to achieve the rich experiments in experience replay, the initialization of robot in each training round is redefined based on RRT strategy. In addition, reward for the free positions is specially designed to accelerate the learning process according to the definition of position cost in A*. The simulation experimental results validate the efficiency of the improved DDQN, and robot could successfully learn the ability of obstacle avoidance and optimal path planning in which DQN or DDQN has no effect.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑