Target-biased informed trees: sampling-based method for optimal motion planning in complex environments

Aiming at the problem that the progressively optimized Rapidly-exploring Random Trees Star (RRT*) algorithm generates a large number of redundant nodes, which causes slow convergence and low search efficiency in high-dimensional and complex environments. In this paper we present Target-biased Informed Trees (TBIT*), an improved RRT* path planning algorithm based on target-biased sampling strategy and heuristic optimization strategy. The algorithm adopts a combined target bias strategy in the search phase of finding the initial path to guide the random tree to grow rapidly toward the target direction, thereby reducing the generation of redundant nodes and improving the search efficiency of the algorithm; after the initial path is searched, heuristic sampling is used to optimize the initial path instead of optimizing the random tree, which can benefit from reducing useless calculations, and improve the convergence capability of the algorithm. The experimental results show that the algorithm proposed in this article changes the randomness of the algorithm to a certain extent, and the search efficiency and convergence capability in complex environments have been significantly improved, indicating that the improved algorithm is feasible and efficient.

Automated summary

This paper proposes an improved RRT* path planning algorithm based on the target-biased sampling strategy and heuristic optimization strategy, which combines target bias and heuristic sampling to efficiently generate paths and optimize convergence capability. The experimental results demonstrate that the algorithm has high search efficiency and convergence capability in complex environments.