Adaptive weight optimization based jump point Theta* algorithm in mobile robot path planning for intricate environments

Conventional path planning methods often struggle to efficiently find optimal solutions in complex and dynamic environments, where the terrain and obstacle distributions vary significantly. This paper presents a novel path planning approach based on the Adaptive Weighted Jump Theta* (AWJ-Theta*) algorithm, which synergistically combines the any-angle path-shortening capability of Theta* with the search efficiency of Jump Point Search (JPS). Key innovations include an adaptive weighting mechanism for the heuristic function that dynamically adjusts based on the distance to the target and local environmental complexity, and a novel node screening strategy that evaluates visibility between nodes and their parent nodes, coupled with cost-based prioritization to reduce computational overhead. The algorithm is systematically evaluated through experiments on standard grid maps, randomly generated maps with varying obstacle densities, and maze environments. Results demonstrate that AWJ-Theta* achieves superior performance in terms of execution time, runtime stability, number of expanded nodes, and path cost compared to traditional Theta*, JPS, and J-Theta* algorithms, particularly in complex scenarios. These findings highlight the algorithm’s potential for efficient and robust real-time path planning in obstacle-rich environments.