Mobile robot path planning for complex dynamic environments

The last few years, research in the area of path planning for mobile robots has been focusing on dynamic environments. Most of methods proposed in this topic need to re-plan the remaining path of the robot, every time when new information come in which significantly increase the computation time and make real-time implementation of these methods difficult and sometimes impossible. The proposed approach consists of two different modules (static and dynamic) that combine two navigation methods to exploit prior information of global static map and local information coming from sensors. The robot uses global information about his environment, and plans the optimal path using genetic algorithms (GA) combined with Dijkstra algorithm through static obstacles. The dynamic phase is done while the robot is moving. The algorithm is able to avoid a moving obstacle by wait/Accelerate concept (W/A C), or by producing a new optimal local path using Dijkstra algorithm. A particularly new aspect of the work is to use the prior information about the environment for searching a global path efficiently and the incorporation of Dijkstra algorithm in both off-line and online phases, which leads to a significantly computation time reducing, and increases the accuracy of the global trajectory. The introduction of the accelerate action is also a new aspect. The simulation results confirm the efficiency and effectiveness of this approach in complex environments.