论文部分内容阅读
为更准确地对驾驶行为进行预警,进一步提高驾驶人换道意图的辨识准确率,借助驾驶模拟器采集数据,建立基于支持向量机(SVM)理论的换道意图辨识模型。对比分析不同人-车-路系统参数组合在换道意图和车道保持期间的差异性,选取最佳特征参数组合,运用网格和遗传算法-支持向量机(GA-SVM)寻优方法优化模型参数,并对优化模型进行验证。结果表明,以纵向加速度、方向盘转角、车辆偏离车道中心线的距离、驾驶人头部运动横坐标变化值作为表征换道意图的人-车-路系统特征参数,优化模型惩罚参数c为58.642 3、核函数参数g为222.732 6时,该模型对驾驶人换道意图的辨识准确率为90%,误警率为5%,基本实现准确辨识换道意图。
In order to more accurately predict the driving behavior and further improve the recognition accuracy of drivers’ lane changing intent, the driving simulator was used to collect the data to establish the lane changing intent identification model based on Support Vector Machine (SVM) theory. The differences of lane-changing intent and lane keeping between different people-vehicle-road system parameters were compared and analyzed. The optimal combination of characteristic parameters was selected, and the optimal model was optimized by using genetic algorithm and GA-SVM Parameters, and verify the optimization model. The results show that the vertical-acceleration, the steering wheel angle, the distance of the vehicle from the lane centerline and the change of the driver’s head movement abscissa are the characteristic parameters of the car-car-road system, the penalty parameter c is 58.642 3 , And the kernel function parameter g is 222.732 6, the accuracy of the model is 90% and the rate of false alarm is 5%, which can basically identify the lane change intention.