Line Crossing Assistance Based on Situation Modulated Potentials Using Stereo Camera Detection

抄録

Designed to help the driver avoid hazardous situations, driving assistant technology can be seen as an intermediary step between cars controlled by humans and automated cars. Its capacity to reduce the number of casualties on the road and its better acceptance from users compared to autonomous technology make it an interesting field of research. One crucial aspect of a safe driving behavior is the avoidance of involuntary lane changes. This situation can occur if the driver does not pay attention to his task or in the case of high-speed driving. However, lane changing can also be desired, and the driving support should then help the driver. This paper describes a proposal for a line crossing active assistant. After realizing road surface marking extraction, virtual potential fields are created from the lines defining the road lane. Assistive torques are then applied to the steering wheel and pedal of the controlled car. The intensity of the effect of this support is modulated as a function of the situation by a fuzzy logic algorithm. Finally, results obtained under real conditions tests are discussed and analyzed.