| 摘 要: 针对传统路径规划方法在大状态空间下计算效率低的问题，提出一种基于改进层次结构的机器人路径动态规划方法。该方法通过将环境建模为多层马尔可夫决策过程，在高层抽象空间中聚类状态以降低计算复杂度，并结合动态规划策略优化路径。实验结果表明，该方法在人工数据和真实机器人数据上均显著减少了值更新次数，提高了路径规划效率，适用于大规模环境下的实时导航任务。 |
| 关键词: 机器人路径规划 层次结构 多层马尔可夫决策 动态规划策略 |
|
中图分类号:
文献标识码:
|
|
| Dynamic path planning for robots based on improved hierarchical structure |
|
CHEN Xiaolong, LI Jiangjiang
|
Zhengzhou University of Science and Technology
|
| Abstract: To address the issue of low computational efficiency of traditional path planning methods in large state spaces, a dynamic path planning method for robots based on an improved hierarchical structure is proposed. This method models the environment as a multi-layer Markov decision process, clusters states in the high-level abstract space to reduce computational complexity, and combines dynamic programming strategies to optimize the path. Experimental results show that this method significantly reduces the number of value updates and improves the efficiency of path planning on both artificial data and real robot data, making it suitable for real-time navigation tasks in large-scale environments. |
| Keywords: Robot path planning hierarchical structure multi-level Markov decision dynamic programming strategy |
