UNMANNED VEHICLE CONTROL SYSTEM BASED ON FUZZY CLUSTERING. PART 1. VEHICLE MOVEMENT MODEL

Abstract

A control system for ground unmanned vehicles is presented, using fuzzy clustering methods for making decisions at an individual level. A new approach to the management of ground unmanned vehicles has been developed, taking into account the state of vehicles in a dense traffic, in particular, the presence of road accidents, the appearance of traffic congestion (high density clusters), etc. An important advantage of this approach is the description of the rules for the interaction of various agents with each other and the external environment within the framework of the final decision-making system of individual agents without the need for a complex computational procedure for identifying the potentials of various forces of the system as a whole. In particular, such rules can be described using systems of differential equations with a variable structure, taking into account all the variety of possible interactions and collisions (potential collisions) between different (moving or stationary) objects. A key feature of the proposed model is the use of the concept of the radius of the agent’s personal space, which explains the effects of turbulence and crush. In this case, the radius of the agent’s personal space is a function of the density of vehicles. As a result, a model of unmanned vehicle movement is developed.