基于跨道时间的车道偏离报警方法

Lane Departure Warning Algorithm Based on the Time Across Road

针对车道偏离报警系统的精确度和鲁棒性不足造成车辆事故频发的问题,提出一种基于跨道时间的车道偏离报警方法,以提升车道偏离报警系统的整体性能。建立稳定可靠的车辆运动模型,根据运动模型预测车辆的行驶轨迹,计算出车轮接触到车道边缘所需要的时间;当此时间小于某一特定阈值时,车辆迅速进行车道偏离报警。仿真与实车实验均证明:基于跨道时间的车道偏离报警方法的平均报警正确率能够达到91.8%,确保车道偏离报警系统准确有效的提示驾驶者,对于减少车辆行驶过程中安全事故的发生具有重要作用。

For lane departure warning system has low accuracy and robustness caused the problem of traffic accidents occur frequently. Put forward a lane departure warning algorithm based on the time across road to enhance the overall performance of the lane departure warning system. Established a stable and reliable vehicle movement model and predicted vehicle trajectory.Calculated the time of wheel contacts with the lane edge. When the time is less than a certain threshold,the lane departure warning system working quickly. Simulation and real vehicle experiment proved that the average accuracy of lane departure warning algorithm based on the time across road can reach 91.8%. Ensured the lane departure warning system prompts driver effectively and accuratly. It will also has a great significant to reduce traffic accidents.