Rapid path planning algorithm for mobile robot in dynamic environment

Abstract

Searching the lowest-cost path through a graph is central to many problems, including path planning for a mobile robot. By combining Dijkstra's algorithm, A* algorithm, and rolling window principle, a new rapid path planning algorithm for a mobile robot in dynamic environment is proposed. First, Dijkstra's algorithm is applied to find an initial path from the initial state to the goal. As a robot moves along the path, if a possible collision is predicted, a local optimal target state within the detection range of the sensors is selected using the rolling window principle. Then, an optimal path from the robot's current location to this local target is searched through A* algorithm and a new path which leads the robot to move from current location to the goal is obtained. Compared to other algorithms, such as ant colony optimization algorithm, A* algorithm, and D* algorithm, the proposed algorithm can always find an optimal path during re-planning and at the meantime greatly reduce the re-planning time. The simulation results prove the feasibility and effectiveness of the proposed algorithm.

Improved Dijkstra Algorithm for Mobile Robot Path Planning and Obstacle Avoidance

“…Although the results were reported to be promising, the time-consuming character of neural networks were considered as a limitation. Other researches have tackled the path planning of MRs using Q-learning algorithm [34] and a combination of Dijkstra and A * algorithms [35]. As compared to the state-of-the-art algorithms for path planning, the improved Dijkstra algorithm has shown its superiority at least in three levels.…”

Quantitative Performance Review of Wheeled Mobile Robot Path Planning Algorithms

“…Hence it is use to solve mobile robot path planning problem. It principle is to expand outward from the start coordinate s to the goal coordinate g , calculate the optimal path costs from s to g through all the free states, and store optimal path from s to g until all states between s to g have been traversed [8].…”

“…It emphasizes metric like, start and goal coordinate, static or dynamic workspace, static or dynamic obstacles, computational time and local minimum problem. In [7,8], path planning algorithms are subcategorized as global and local. Global path planning algorithms are frequently applied to static workspace having static obstacles and robot has comprehensive knowledge of this workspace.…”

“…Global path planning algorithms are frequently applied to static workspace having static obstacles and robot has comprehensive knowledge of this workspace. This algorithm generates the traverse path off-line before the movement of the robot within the workspace [8,9]. However, these algorithms are very ineffective in real-world workspace, where the workspace is unpredictable and clustered with static and dynamic obstacles.…”

A Modification of Shortest Path Algorithm According to Adjustable Weights Based on Dijkstra Algorithm

“…In our future research effort, we will use a changing target state to calculate the trail planning issue, and we can increase the algorithm's efficiency further. Moreover, we will study the procedures regulating transient system performance [9]. In 2018 Sunita did a study on this subject.…”