BH随机邻域嵌入在驾驶行为识别中的应用

针对驾驶系统处理大量驾驶数据时出现的效率和精度不足的问题,提出一种基于巴恩斯哈特随机邻域嵌入(BH-SNE)和径向基函数神经网络(RBFNN)的识别算法。从手机传感器中获取加速度数据、陀螺仪数据和磁强计数据,融合这三种传感器数据,经过预处理后使用BH-SNE完成降维处理,将降维数据输入到RBFNN中识别出驾驶行为。实验结果表明,BH-SNE的效率远高于t分布式随机邻域嵌入(t-SNE),并且可视化效果优于t-SNE,该模型的整体识别率为98.8%,分类效果优于传统的机器学习算法。

APPLICATION OF BH-SNE IN DRIVING BEHAVIOR RECOGNITION

In order to solve the problem of inefficiency and inaccuracy in handling large amounts of driving data in driving system,this paper proposes a recognition algorithm based on Barnes-Hut stochastic neighbor embedding(BH-SNE)and radial basis function neural network(RBFNN).The acceleration data,gyroscope data and magnetometer data were obtained from the mobile phone sensor,and the three sensor data were fused.After preprocessing,the BH-SNE was used to complete dimension reduction,and the dimension reduction data was input into RBFNN to identify driving behavior.The experimental results show that the efficiency of BH-SNE is much higher than that of t-distributed stochastic neighbor embedding(t-SNE),and the visualization effect is better than t-SNE.The overall recognition rate of this model is 98.8%,and the classification effect is better than the traditional machine learning algorithm.