The path planning problem is one of the most researched topics in autonomous vehicles. During the last decade, sampling-based algorithms for path planning have acquired signiﬁcant attention from the research community. Rapidly exploring Random Tree (RRT) is a sampling-based planning approach, which is a concern to researchers due to its asymptotic optimality. However, the use of samples close to obstacles in path planning and the path with sharp turns does not make it efficient for real-time path tracking applications. For the purposes of overcoming these limitations, this paper proposes a combination of RRT and Dijkstra algorithms. The RRT-Dijkstra guarantees a shorter path planning to the optimum and collision-free solution. The optimality is measured by various factors such as path length, execution time, and the total number of turns. The aim here is review and performance comparison of these planners based on metrics, i.e., path length, execution time, and the total number of turning points. The algorithms are tested in complex structured with obstacles environments. The experimental performance shows that RRT-Dijkstra requires less turning point and execution time in 2D environments. These are advantages of the proposed method. The proposed method is suitable for off-line path planning and path-following.
Optimal criteria, Path planning, RRT, Dijkstra, Sampling-based algorithms