Bio-inspired Algorithm for Path Planning of Terrestrial Robot Using Aerial Images

Path Planning is one of the most important tasks in the navigation of a mobile robotics. In this paper, we focus on global path planning, Using the grid method to present the workspace. The division of the grid size is critical, the bigger the grid size it will be more accurate but it will take up a lot of time the search through the map. Otherwise, it will be very imprecise if grid size is set too small. Then we use both grid sizes, assuming we count with a high-resolution aerial image of the terrain, The A-Star algorithm is applied to the small-sized grid to find the path to the target, this minimizes the graph created by this algorithm and avoid the problem with local minimums. The result is translated to the high-resolution grid to be optimized by a bio-inspired algorithm called particle swarm optimization to finally get a free collision path. We use different fitness functions to get the smoothest path to the target.