Jabar RRT* Path Planning

Aiming at the problems of slow convergence, path redundancy and fixed step size of the traditional rapid expansion random tree (RRT), this paper proposes a Jabar RRT ^* algorithm, which integrates three strategies: the artificial potential field guided sampling, adaptive expansion step size and greedy path smoothing: the artificial potential field guides the sampling points to the low potential energy region by dynamically adjusting the gravity coefficient and optimizing the repulsion calculation, so as to reduce invalid sampling; The adaptive step size is dynamically adjusted according to the distribution of environmental obstacles, target distance and node density to balance the exploration efficiency and obstacle avoidance safety; Greedy path smoothing eliminates redundant inflection points through piecewise optimization, and adapts to robot motion constraints. The results show that compared with the basic RRT and the standard RRT, the improved algorithm reduces the number of iteration convergence by 57%, the planning time by 36%, the path length by 30%, and the proportion of invalid sampling to 12%. Through the simulation deployment of gazebo, the practicability and superiority of the improved algorithm in the smart factory scenario are verified, which can meet the requirements of efficient and safe path planning of the transport robot in the dynamic environment.