Paradigms of driving a highly automated vehicle

Abstract

The vehicle moving in space without direct human participation, ceased to seem something supernatural or fantastic. Today, there are cases of fairly successful implementation of a motor vehicle autonomous driving. In many cases this was caused by a sharp leap in the development of intellectual transport systems at the turn of the first two decades of the 21st century. However, until some significant problems are solved talking about the widespread appearance of autonomous vehicles is premature. The purpose of the work is to formulate the main problems arising in the transition to the higher levels of automation. The article lists three main problems which are to be solved while implementing autonomous vehicle driving: legal (responsibility for road traffic accidents), technological (adaptability of algorithms, self-organization and survivability of the driving system) and social (hedonistic approach to driving). The author calls the widely spread application of mobile gadgets one of the main social problem and notes the negative impact of this phenomenon on the level of road safety in the country. The transition to fully automatic control seems to be the only possible solution to this problem. Modern classifications of automation levels have been considered and a number of remarks was made concerning them. The author developed a new classification of driving paradigms depending on the kind of signal between the elements of the system “driver - vehicle - road - objects - environment” (DVROE), which included traditional, active, intelligent, remote, autonomous and isolated control. The main problem arising in the transition to an autonomous and isolated level of automation is the adaptation of the driving process to the elements of the DVROE system. The author proposes some solutions to this problem. A diagram of the application of various control paradigms, depending on the automation level of the vehicle is suggested.