Abstract
One of the challenges in autonomous road vehicles is designing a system that is applicable in all weather conditions. An original hybrid navigation system with RTK DGPS is used for this purpose. The challenges of affordability and driving efficiency are solved with a novel approach of precisely compiling track data for autonomous driving (geodata, inclines, speed limits, coefficients of fuel consumption and ecological efficiency, weather limitations, etc.). The notion of a “basetrack” is introduced, which is a high-accuracy spatial driving route that contains additional datasets for the vehicle’s optimal control with respect to the requested transport task, road, and traffic conditions. The basetrack concept conditionally transforms an autonomous car into rail transport with a virtual railway and simplifies all the challenges of the tactical level of vehicle control. The advantages of the basetrack data structure include the predetermined variability of control actions on the route, depending on the tasks of time saving, energy efficiency, or safety; increased reliability; reduced requirements for on-board computing power; and improved performance. The proposed approach provides autonomous highway driving and supports lane-keeping assistance systems in cases of failure to visually recognize road markings. Fuel consumption was optimized for autonomous driving mode on a hilly test road over a distance of 200 km and showed up to 4% of fuel savings for a passenger car in comparison with constant-speed driving and the same trip time.