Road Traffic Environment Detection And Data Fusion Technology

Intelligent vehicle traffic environment perception system consists of traffic environment and sensor hardware and software,and the interaction between traffic environment and sensor seriously affects the safety of traffic participants.Based on the main characteristics of the known traffic environment,the corresponding detection and recognition algorithms are designed,the logic detection and logic recognition algorithms are designed.Through the real vehicle experiment,the algorithm is verified and the information fusion algorithm is studied.Then the intelligent vehicle finite state machine model is established.According to the characteristics of the traffic environment,the characteristics and main influencing factors of the traffic environment perception system of the intelligent vehicle are studied by theoretical analysis and real vehicle observation.The main features of the environment are obtained,including the line and angle features,The zebra crossing rectangle,the angle,the striped characteristic,the front vehicle's rectangle,the symmetrical characteristic,the traffic signal lamp aspect ratio,the color,and the position characteristic.The algorithm of lane-line detection,zebra crossing detection,front vehicle identification and traffic signal identification are designed,and the real-time vehicle experiment scheme is studied and implemented to verify the real-time and accuracy of the algorithm.According to the environmental characteristics obtained in the real vehicle experiment,the basic rules of feature-level information fusion are obtained,and the traffic environment information fusion data are obtained.Based on the results of real vehicle experiment and information fusion,the intelligent vehicle's limited State machine model is established.