OPTIMIZATION OF DANGEROUS SECTION PASSAGE FOR UNMANNED VEHICLES

Abstract

Subject of research. The research is focused on optimization management tasks of the ground unmanned vehicles in difficult conditions. We showed that it is possible to minimize passage time of road sections with narrowing and maintain security requirements in case of the centralized management. Method. A method of the unmanned cars movement is based on the dynamic programming model. The method provides faster movement than by full search of all possible variants for dangerous road section passage. Such result is achieved by the reorganization of the control task to the class of extreme tasks and the solution of such tasks is based on the Bellman recurrence equation. According to the proposed method a road section is considered as a system with discrete time and finite set of the end positions. Optimal plan of the cars movement is prepared by the central computing device, which has gathered information about road infrastructure and vehicles. The central device is sending control commands to the cars about the change of speed and movement direction in such a way that given criteria and restrictions are satisfied. Main results. The efficiency of the method is demonstrated by the example of driving on a two-lane road. One of the road lanes is blocked for traffic. It is proved that there is a solution that enables all cars to leave the dangerous road section in the shortest time with safety measures. Practical relevance. The proposed method is applicable for control of unmanned vehicles communicating with the central control device by wireless radio. The central control device is a part of the road infrastructure in conditions of isolated environment when there are no any fully autonomous vehicles in the group.

Keywords

• unmanned vehicles
• dynamic programming
• traffic
• Bellman equation