Research of graph-analytical methods for a vehicle motion planning

In this paper we present the results of research on the efficiency of graph-analytical methods for vehicle motion trajectory planning under uncertainties. The three most common graph-analytical methods were considered: potential fields, algorithms based on Voronoi diagrams, and A*-algorithm combined with the method of dynamic window. We proved the applicability of the considered methods to the problems of trajectory planning for complex dynamical objects on the example of the quadrotor flight control. The testing of the considered graph-analytical methods efficiency was carried out by simulation of the quadrotor horizontal flight through four maps with different number and location of obstacles. Moreover, initially the control system of the vehicle does not have any information about obstacles. The trajectories are created during the motion on the basis of new information about the map. As a result, it was concluded that all methods can be used in uncertain environments for the control of complex dynamical objects of quadrotor type.