基于支持向量机的车辆制动距离研究

Research on vehicle braking distance based on support vector machine

车辆的制动距离对汽车的主动安全性尤为重要,以制动时的初始速度和路面附着系数作为影响制动距离的主要因素,在TruckSim中搭建了某款客车的模型后进行制动仿真实验,获得了176组在不同的路面附着系数和不同制动初始速度下的制动距离,在MATLAB中基于支持向量机进行分析,利用其中的140组数据组成训练集对模型进行训练,用剩余的36组数据组成测试集进行交叉验证,结果显示预测的相对误差在8.3×10^(-5)以内,拟合优度判定系数在0.99978以上,表明预测精度较高。因此,基于支持向量机可以对车辆的制动距离进行精确预测。

The braking distance of the vehicle is very important to the active safety of the car. The initial speed of braking and the adhesion coefficient of the road are the main factors that affect the braking distance.A model of a passenger car was built in TruckSim, the braking simulation experiment was carried out. The braking distance of 176 groups under different adhesion coefficient and different initial speed of brake was obtained. In MATLAB, based on support vector machine （SVM）, the training set of 140 groups of data is used to train the model, and the remaining 36 sets of data are used to cross validation. The results show that the relative error of the prediction is less than 8.3×10-5, the goodness of fitting coefficient is more than 0.99978, which indicates that the prediction accuracy is higher. Therefore, the braking distance of vehicles can be accurately predicted based on support vector machine.