Driving intention identification is a key technology which can improve the adaptability of the intelligent driver assistance systems and the energy efficiency of electric vehicles. This article proposes a novel method for identi-fying the driver braking intention. In order to improve the identification accuracy of driving intention, a braking inten-tion identification model based on Long Short-Term Memory (LSTM) Network is constructed. The data of slight brak-ing, normal braking and hard braking that can use for offline training are obtained through tests on real vehicle at Chang’an University vehicle performance testing ground. Support vector machine - recursive feature elimination (SVM-RFE) algorithm is used to select the characteristic parameter of braking intention identification model. The ran-dom search is subsequently used to optimize the hyper-parameters of LSTM. LSTM-based and General Hidden Markov Model (GHMM)-based model under different time window are used to identify braking intention of slight braking, normal braking and hard braking respectively. The results show that the Precision, Recall, F-measure, Accuracy of the braking intention identification model which propose in this paper based on LSTM are better than that of the braking intention identification model based on GHMM. Moreover, the Recall and Accuracy of the LSTM-based braking inten-tion identification models are above 0.95, indicating the good ability of intention identification.

Pedestrian tactical choices and operational movement in evacuations essentially pertain to decision-making under risk and uncertainty. However, in microscopic evacuation models, this attribute has been greatly overlooked, even lacking a methodology to delineate the related decision characteristics (bounded rationality and risk attitudes), let alone their effects on evacuation processes. This work presents an innovative two-layer floor field cellular automaton model framework, where three intertwined sub-modules respectively dedicated to modelling the exit choice, the locomotion movement and the exit-choice changing behaviours are proposed and integrated as an entity. By introducing various decision-making elements computed by the proposed algorithm, Cumulative Prospect Theory (CPT) is proposed for the first time to model the exit choice and locomotion decision-making under risk and uncertainty. In the exit-choice changing module, attractive and repulsive forces are invented to jointly describe the tendency to revisit the routing decision. Each sub-module and the whole framework are validated in manifold indoor environments. The simulation results of the modules with CPT accord with the empirics from the evacuation experiments and are superior over those from the state-of-the-art models. The degree of rationality and risk attitudes are proven to have significant impacts on tactical and operational decisions. Furthermore, irrational behaviour in decision-making is not variably detrimental to locomotion efficiency of pedestrians. The proposed framework can serve as an elegant tool to predict pedestrian dynamics. The behavioural findings shed new light on understanding and modelling the tactical and operational decisions in evacuations.

Reliable motion prediction of surrounding vehicles is the key to safe and efficient driving of autonomous vehicles, especially at urban intersections with complex traffic environments. This study models driving intentions and future driving speeds at urban intersections and improves model prediction performance by considering traffic environment characteristics. Key feature parameters including environmental characteristics are first extracted through driving behavior analysis and existing research experience. Then models with different input combinations are constructed to explore the effectiveness of different factors in predicting driving intention and future speed. In particular, in vehicle speed modeling, a target detection algorithm is used to identify traffic participants. Based on the identified traffic participant and vehicle position information, a new method for speed prediction that can reflect the dynamic interaction characteristics between the driver and the traffic environment is proposed. Models are trained and tested using natural driving data from China. Finally, the models with the simplest input and the best effect are determined. The driving intention recognition model can accurately predict the driving maneuvers of straight-ahead, stopping, turning left and right 4 seconds before reaching the intersection. The speed prediction model can significantly improve the speed prediction accuracy, and shows stronger robustness and adaptability than existing models. This research provides important technical support for developing intelligent driving systems suitable for complex urban traffic environments.

A large amount of traffic crash investigations have shown that rear-end collisions are the main type collisions on the freeway. The purpose of this study is to investigate the rear-end collision risk on the freeway. Firstly, a new framework was proposed to develop the rear-end collision probability (RCP) model between two vehicles based on Generalized Pareto Distribution (GPD). Secondly, the freeway rear-end collision risk (F-RCR) was defined as the sum of the rear-end collision probability of each vehicle and divided into three levels which was high, median, and low rear-end collision risk. Then, different machine learning algorithms were used to model F-RCR under the condition of an unbalanced dataset. The result of the RCP model showed continuous change and can identify the dangerous condition quickly compared to the traditional models even when the speed of the leading vehicle is faster than the following vehicle. When the vehicle distribution was unbalanced on road and the speed difference between adjacent lanes and the traffic volume was large, F-RCR will increase. Multi-Layer Perceptron (MLP) was found to be more suitable for modeling F-RCR. The framework provided in this research was transferrable and can be used in the freeway proactive traffic safety management system.

Vehicle stability control should accurately interpret the driving intention and ensure that the actual state of the vehicle is as consistent as possible with the desired state. This paper proposes a vehicle stability control strategy, which is based on recognition of the driver’s turning intention, for a dual-motor drive electric vehicle. A hybrid model consisting of Gaussian mixture hidden Markov (GHMM) and Generalized Growing and Pruning RBF (GGAP-RBF) neural network is constructed to recognize the driver turning intention in real time. The turning urgency coefficient, which is computed on the basis of the recognition results, is used to establish a modified reference model for vehicle stability control. Then, the upper controller of the vehicle stability control system is constructed using the linear model predictive control theory. The minimum of the quadratic sum of the working load rate of the vehicle tire is taken as the optimization objective. The tire-road adhesion condition, performance of the motor and braking system, and state of the motor are taken as constraints. In addition, a lower controller is established for the vehicle stability control system, with the task of optimizing the allocation of additional yaw moment. Finally, vehicle tests were carried out by conducting double-lane change and single-lane change experiments on a platform for dual-motor drive electric vehicles by using the virtual controller of the A&D5435 hardware. The results show that the stability control system functions appropriately using this control strategy and effectively improves the stability of the vehicle.