基于形变长短期记忆网络的换道意图预测

混行交通下的自动驾驶车辆需具备换道意图预测能力来保障行驶安全。为尽早预测车辆换道意图，提出一种基于形变长短期记忆(mogrifier long short-term memory, M-LSTM)网络的换道意图预测模型。首先采用S-G (Savitzky-Golay)滤波器对自然驾驶数据集NGSIM(next generation simulation)进行降噪筛选，按向左换道、向右换道、直线行驶对不同时间长度的轨迹序列标注，选取车辆运动信息与环境信息输入模型，最后采用softmax函数进行意图分类。试验结果表明，在不同预判时间下，模型准确率均高于支持向量机(support vector machine, SVM)、LSTM模型，且越接近换道点预测准确率越高，在1.0、2.5 s时预测准确率分别为93.83%与81.30%。提出的模型具有良好的准确性与预判性，能为自动驾驶车辆尽早识别换道意图提供技术支持。

Lane-Changing Intention Prediction Based on MOGRIFIER LSTM

Automatic driving vehicles need to have the ability to predict the intentions of changing lanes to ensure driving safety in mixed traffic.In order to predict the intention as early as possible, a prediction model based on Morgrifier LSTM network is proposed. First, the S-G (Savitzky?Golay) filter is used to filter the noise reduction of the natural driving data set NGSIM (Next Generation Simulation), and the track sequence of different lengths of time is marked by changing lane to the left, right, and driving straightly.Select the input model of vehicle motion information and environmental information. Finally, the softmax function is used to classify the intention. The result shows that the prediction accuracy of the model is higher than SVM and LSTM under different prediction times, and the closer to the lane-changing point, the higher the prediction accuracy. At 1.0 s and 2.5 s, the prediction accuracy is 93.83% and 81.30% respectively. The proposed model has pleasurable accuracy and predictability. It can provide technical support for automatic driving vehicles to identify lane-changing intention as early as possible.