<![CDATA[In the field of wheeled mobile robots (WMRs), path planning is a critical concern. WMRs employ advanced algorithms to find out the feasible path from a starting point to a specific destination. The paper proposes efficient and optimal path planning for WMRs, integrating collision avoidance strategies and smoothed techniques to determine the best route during navigation. The proposed hybrid path planning consists of improved RRTstar algorithm and reinforcement learning method. Therefore, the RRT* algorithm employs random sampling in conjunction with a reinforcement learning model to purposefully guide the sampling process towards areas that demonstrate an increased likelihood of successful navigation completion. The proposed RRTstar-RL algorithm generates significantly shorter trajectories compared to the traditional RRT and RRTstar methods. Specifically, the path length with the proposed algorithm is 11.323 meters, while the lengths for RRT and RRTstar are 15.74 and 14.40 meters, respectively. Moreover, the optimization of computation time, especially when using pre-trained data, greatly speeds up the path-finding calculation process. In particular, the time needed to generate the optimal path with the RRTstar-RL algorithm is 2.02 times faster than that of RRTstar and 1.6 times faster than RRT. Finally, the proposed RRTstar-RL algorithm has been successfully verified for feasibility and effectively meets numerous objectives established during simulations and validation experiments.]]>
