Finding optimal local path for mobile robot with kinematic constraints: How to find the safest path for nonholonomic robot

The paper deals with the optimization of local path for the mobile robot, whereas the safety of the route is the major criterion. The robot equipped with differential chassis with kinematic constraints is considered for the task. The optimization is based on genetic algorithm (GA), optimizing the polyline representation of the path. The core idea of the method presented is to sequentially increase the number of key points in the path during the optimization. This way the computational expenses are kept reasonable and resulting path quickly converges to the optimal one, with respect to the actual number of parameters to be optimized. The cost function is defined as the lowest distance from the obstacle along the path. Sequential refinement of key path proved to speed up the search.