计算机工程与应用2025，Vol.61Issue(24)：116-133,18.DOI:10.3778/j.issn.1002-8331.2506-0280
SN-BI-RRT*:基于动态梯度和人工势场的双向探索随机树算法
SN-BI-RRT*:Bidirectional Exploratory Random Tree Algorithm Based on Dynamic Gradient and Artificial Potential Field
摘要
Abstract
Aiming at the problems of low convergence efficiency and randomness of search direction of RRT*(rapidly-exploring random tree star)algorithm in complex obstacle scenarios,which lead to poor path generation,this paper proposes a bidi-rectional fast exploratory random tree algorithm(SN-BI-RRT*)based on dynamic gradient sampling and artificial poten-tial field.A stepwise dynamic gradient sampling strategy is used to optimize the sampling process and explore the configura-tion space more effectively.An improved artificial potential field method is introduced in the expansion to improve the convergence speed of the algorithm.In addition,the generated new nodes are optimized with an improved reconnection parent node strategy to reduce the total cost of the path.In order to improve the smoothness of the paths,a fusion path smoothing strategy of path pruning,linear interpolation and B-spline smoothing is used for post-processing.Through simulation experiments,the SN-BI-RRT*algorithm is compared with several other sampling-based path planning algo-rithms in different obstacle environments and narrow environments,and the results show that the algorithm has good per-formance in different environments,and it can be an effective solution to the problem of efficient path planning for robots in complex indoor environments in robot path planning.关键词
路径规划/动态梯度采样策略/人工势场法/节点优化Key words
path planning/dynamic gradient sampling strategy/artificial potential field method/node optimization分类
信息技术与安全科学引用本文复制引用
HUANG Yourui,ZHU Zhongtao,HAN Tao..SN-BI-RRT*:基于动态梯度和人工势场的双向探索随机树算法[J].计算机工程与应用,2025,61(24):116-133,18.基金项目
安徽省高校协同创新项目(GXXT-2023-068). （GXXT-2023-068）
