Toward More Safety оn the Roads: Development of a Lane Keep Assistance System for Public Transport

this paper discusses the development of Lane keep assistance system on the level 2 of Advanced Driver Assistance Systems (ADAS) for the public transport on Russian roads (KAMAZ vehicles). ADAS systems improve the safety on the road by warning the driver about any potential collision on the zero level and make control for the higher levels of automation. The article discusses the question of architecture for the proposed system. Random Sample Consensus (RANSAC) algorithm is introduced in general, then it is presented as a new approach for lane detection. The lane coordinates can be received as a result of vision algorithm. Moreover, the lateral vehicle dynamics is introduced, and control algorithm is used for controlling the vehicle lateral position. The conducted results can help to keep the vehicle on the lane by warning the driver and controlling the vehicle steering. The results conducted in the Simulator give a good performance of the proposed system.