Indoor Mobile Robot Path Planning Based on Improved A* Algorithm
Qijie Lu, Dongcheng Wei, Liu Yang · 2025
To address issues such as data redundancy, low execution efficiency, and the difficulty for mobile robots to adjust their posture at turning points encountered by the traditional$A^{*}$algorithm in path planning, this paper proposes an improved$A^{*}$path planning algorithm. It can effectively calculate turning points, rotation directions, and minimum rotation angles, and smooth the path by optimizing the selection of key points. The algorithmic refinement phase initially optimizes computational throughput through enhanced jump point prioritization, strategically pruning redundant node expansions in$A^{*}$path exploration. Complementing this structural optimization, a dynamic heuristic weighting mechanism was implemented to refine search directionality. Finally, parametric Bézier interpolation was applied to ensure geometric continuity in derived trajectories, achieving second-order continuity essential for kinematically feasible robotic navigation. To verify the feasibility and effectiveness of the improved algorithm, comparative simulation experiments were conducted on grid maps of different sizes and mobile robot platforms. Comparative simulations across varied grid maps and mobile robots demonstrate about 36.25 % fewer expanded nodes, 31.25 % faster computation, and 21.03 % improved path smoothness versus conventional$A^{*}$. The optimization framework significantly enhances angular adaptation at inflection points and navigation efficiency, validating its effectiveness in autonomous systems. The integrated approach balances computational economy with trajectory quality, establishing a practical advancement in robotic path planning paradigms.