Keywords: Intelligence of vehicle
Abstract: In order to improve the accuracy of the expression of the design concept of ICV, this paper puts forward a metaphor mapping method between ICV design concept and multi-sensory design elements, analyzes the metaphor mapping principle between the design concept and multi-sensory design elements, establishes a mathematical model based on the "entropy method" for the metaphor mapping between the design concept of ICV and multi-sensory design elements, and designs the vehicle design concept and multi-sensory design elements The system and device of metaphorical mapping of multi-sensory design elements verify the feasibility of the proposed method with the mapping process of design concept and multi-sensory design elements in the forward design of an intelligent Internet connected SUV.

Keywords: Intelligence of vehicle, Vehicle Dynamics and Control
Abstract: Aiming at the problem that UGV can easily fail to identify invisible water obstacles on unstructured roads due to image recognition method, this paper proposes an algorithm for recognizing and avoiding invisible water obstacles in the field based on vehicle dynamics. First, the vehicle model and tire model are established, the road adhesion coefficient identification framework is built, and the identification algorithm is designed. Secondly, a three-dimensional information grid obstacle map containing road adhesion coefficient information is established to define the original planned path coordinates, and the mapping processing method on the grid map and the search direction of modified path are defined according to different obstacle shapes in the field. Finally, an obstacle avoidance algorithm is designed and evaluated. By comparing the recognition value and setting the threshold value, it can judge whether to drive to the edge of the obstacle. If the point is not suitable for walking, the feasible grid point is searched again; otherwise, the grid point is kept. The simulation results show that this method can be used to plan a safe obstacle avoidance path for unmanned ground vehicle in the field under the rainy bog obstacle scenario.

Keywords: Vehicle Dynamics and Control, Integrated Chassis Control, Automated and Connected Vehicle
Abstract: This paper proposes a novel observer-based sliding mode control (SMC) to enhance the performance of autonomous vehicles (AVs) rollover behavior under various road profile input. The model of half-car system is first established to describe the AVs rollover behavior by considering nonlinear dynamics of tire force and controllable suspension force under various movement conditions. Moreover, an unscented Kalman Filter (UKF) algorithm is proposed to identify the sprung mass. Combing with the interacting multiple model (IMM) approach and Markov Chain Monte Carlo (MCMC) theory, a novel interacting multiple model unscented Kalman Filters (IMMUKF) observer based is developed to estimate the movement state of AVs system. Then, an adaptive observer-based sliding mode control (AOSMC) strategy is proposed to constrain the AVs roll performance under the various external input. The stability of the proposed algorithm is proved by using Lyapunov function. Finally, simulations and validations are performed on a high-fidelity CarSim® software by using J-turn scenario under various road excitation, to validate the proposed algorithm for AVs system, and the results illustrate that the improved roll states are more than 15% compared with the traditional SMC algorithm.

Keywords: Advanced Driving Assistant System, Vehicle Dynamics and Control
Abstract: Adaptive cruise control (ACC) systems have been implemented successfully in production vehicles primarily for improving safety and reducing workload. As the core of all ACC systems, the mode switching strategy directly affects the performance of the ACC system. This paper presents a brief comparison of mode switching strategies for existing ACC solutions in the literature. The current mode switching strategies can be classified into three major types according to their required information, including the distance-based switching strategy, distance-speed-based switching strategy, and distance-speed-acceleration-based switching strategy. Their characteristics are carefully reviewed and explained. Besides, future works for improving the existing mode switching strategies are also presented.

Keywords: Vehicle Dynamics and Control, Vehicle control, Intelligence of vehicle
Abstract: A motion planning framework for autonomous vehicles in highway considering position uncertainty of surrounding vehicles is proposed in this paper. The extended Kalman filter (EKF) is applied to predict the future trajectory of surrounding vehicles and determine its position uncertainty in a confidence ellipse, which will ensure generate a safe trajectory in real time. Besides, a load transfer ratio (LTR) index is designed to describe whether the vehicle has reached the critical rollover state in highway. Furthermore, the position uncertainty and artificial potential field are filled into the model predictive controller objective to achieve obstacle avoidance while guaranteeing safety and stability. Finally, the simulation results implemented in the Prescan-Simulink-Carsim co-simulation platform show the effectiveness of the presented framework.

Keywords: Human-Computer Interaction, Intelligence of vehicle, Advanced Driving Assistant System
Abstract: Drivers play an important role in the transportation system. Novice drivers have insufficient driving risk awareness due to lack of driving experience, which has become a potential hazard in the traffic system. The automotive driving assistance system (ADAS) can more or less help the novice driver to avoid danger. In order to further improve the ADAS control strategy for drivers with different driving experience, it is necessary to identify novice drivers and experienced drivers. In this study, a twelve-kilometer two-way straight highway was designed as the driving scenario. Electroencephalogram(EEG) data generated in the frontal region was recorded as an indicator to evaluate the driver’s perception of danger. We aim to identify novice drivers and experienced drivers by using beta waves extracted from collected EEG data when facing dangerous situations. The results indicate that the EEG features (PSD value of beta wave) extracted from the frontal region can effectively recognize the novice driver and the experienced driver through the BP neural network, and achieve a relatively high accuracy at nearly 88%.

Keywords: Decision Making, Intelligence of vehicle
Abstract: In this paper, we focus on a self-adaptive network reconstruction problem in a distributed multi-agent system. In this system, the relationship among agents forms a complex network, in which each agent plays two roles, task execution and communication. As agents move, the topology of the relationship changes, which could reduce the system’s performance. To guarantee the performance, we design a local centrality for evaluating the vitality of nodes in the network, and propose a distributed re-construction algorithm (DRA) based on the local centrality for re-constructing the dynamic network. To test the efficiency of DRA, we integrate DRA into our former proposed dynamic task allocation algorithm to form a new dynamic task allocation algorithm. The experiment shows that our new improved algorithm performs more efficiently when compared to other dynamic task allocation algorithms. The results illustrate the efficiency of reconstructed dynamic complex networks, which also indicates that the local centrality preforms efficient. Finally, we replace the local centrality with degree centrality and LocalRank in DRA and execute a series of experiments for performance evaluation. The results demonstrate the high efficiency of LC.

Keywords: Vehicle control, Intelligence of vehicle, Vehicle Dynamics and Control
Abstract: The affine formation tracking problem for fixedwing unmanned aerial vehicles (UAVs) is considered in this paper, where fixed-wing UAVs are considered as unicycletype agents with asymmetrical speed constraints. A group of UAVs are required to generate and track a time-varying target formation obtained from transforming a nominal formation affinely. To handle this problem, a distributed control law based on the stress matrix is proposed under the leader-follower control framework, which only uses the relative position information. It is theoretically proved the followers can converge to desired positions which is determined by leaders’ positions while tracking different trajectories. Furthermore, two numerical simulations are executed to produce evidence verifying the strengths of our proposed affine formation control strategy.

Keywords: Vehicle control, Automated and Connected Vehicle, Railway Train Control
Abstract: In this brief, the containment problem of double-integrate discrete-time agents network is investigated with control input and velocity constrains. A nonlinear projection algorithm is proposed to converge all follower agents into a convex hull formed by leaders, where a scaling factor is proposed to solve the nonlinear constrains such as saturations and nonconvex constrains. Based on model transformation and Lyapunov function, the distance between between follower agents and the convex hull is proved to be nonincreasing under suitable assumption. Finally, after convex analysis, the containment problem is solved by the proposed algorithm with bounded time delays as long as the union of the topology graphs has spanning trees. A simulation result is employed to illustrate the validity of theoretical results.

Keywords: Intelligence of vehicle, Automated and Connected Vehicle, Advanced Driving Assistant System
Abstract: It is a difficult challenge for humans to carry out environmental perception work at night and in low-light scenes. Depending on its extraordinary working performance in the dark, starlight camera is widely used in night driving assistance and various surveillance missions. However, the starlight camera images are lack of colorful information, which prevents users from understanding. This paper proposes a novel approach for colorizing starlight images using Generative Adversarial Network (GAN) architecture. The proposed method overcomes the time-space asynchronism of traditional heterogeneous data acquisition. We firstly introduce starlight-RGB image pairs generation. Inspired by 3D perspective transformation, we use LiDAR, camera and IMU data to create generated visible images. We collect synchronous visible images, lidar points data and IMU data in the daytime and acquire LiDAR, starcam and IMU data at night. Such image pair generation method overcomes the difficulty of obtaining pairs of data and image pairs are aligned at pixel-level. As there are no reflection LiDAR points in the sky, the perspective projection images have no content in the sky areas. Based on supervised image-to-image translation GAN architecture, we use daytime RGB images as unpaired data, which is in order to restore the texture and color of the sky. We use KITTI dataset as validation, and get good experimental performance on our datasets.

Keywords: Decision Making, Control of UAV/USV/UUVs
Abstract: Unmanned ground vehicles(UGV) autonomous control technology contains lots of methods, which autonomous decision-making is the most important one. The reliability of its strategy determines the result of UGV missions. To improving the environmental adaptability of traditional methods such as Bayesian network and dynamic Bayesian network, this paper proposes a parameter estimation method of particle filter based on the parameter time-varying discrete dynamic Bayesian network (PTVDDBN) for the situation where the parameters change smoothly with time, and applies it to UGV decisionmaking tasks. First, formally describe the general model of the discrete dynamic Bayesian network with time-varying parameters under the condition of invariable structure. Secondly, a novel method of parameter estimation and inference decision framework of PTVDDBN is put forward. Thirdly, a parameter estimation method of PTVDDBN based on particle filter is proposed. Finally, with UGV cooperative decision-making in battlefield as the application background, time-varying parameter estimation and PTVDDBN model inference and decisionmaking experiments were carried out. By comparing static Bayesian network and dynamic Bayesian network models, the analysis showed that the PTVDDBN model algorithm can estimate the time-varying parameters more accurately, and the decision-making inference is more reliable.

Keywords: Automated and Connected Vehicle, Vehicle control
Abstract: In the process of unmanned vehicle path tracking control, in view of the important problem that the traditional PID controller is difficult to adjust its control parameters in real time when the control object changes. This paper proposes a fuzzy adaptive PID control method based on vehicle kinematics and dynamics. First, the next driving path is planned according to the preview theory;then the positional relationship is used between the center of mass of the vehicle and the desired path preview point to calculate the lateral deviation and heading deviation; finally, the influence of deviation on path tracking is comprehensively considered, and the adjustment effect of the fuzzy adaptive PID controller on the error is used to adjust the front wheel angle. In order to compare the effect of traditional PID and adaptive fuzzy PID, simulation experiments are carried out respectively. The simulation results show that fuzzy adaptive PID improves the dynamic performance of the controller and has better adaptive ability.

Keywords: Intelligence of vehicle, Advanced Driving Assistant System
Abstract: An encoder-decoder convolutional neural network architecture is presented integrating multi-class semantic segmentation and multi-object detection to improve the efficiency and depth of scene parsing of intelligent vehicle. The encoder of the network is designed as a multi-scale structure to improve real-time performance while ensuring the accuracy. The decoders of the network comprise the semantic segmentation and object detection subnetworks, which share encoder feature maps to improve computational efficiency. During the training process, we use FPS (Frames Per Second) and MIoU (Mean Intersection over Union) as the evaluation metrics of semantic segmentation, while the mAP (mean Average Precision) and FPS are used as the performance evaluation indexes of object detection. We conduct separate and joint training on the network to evaluate its performance. Experimental results show that the proposed network can realize multi-class semantic segmentation and multi-object detection simultaneously with better real-time performance and richer feature information, making it highly possible for implementation on real vehicles.

Keywords: Intelligence of vehicle, Advanced Driving Assistant System, Automated and Connected Vehicle
Abstract: The object detection based on deep learning is an important application in the field of vehicle environment perception, which has been a hot topic in recent years. We propose a novel improved Yolov3-tiny to implement more accurate object detection for the objects in traffic scenes. We employ K-means algorithm to cluster the common objects in traffic scenes to obtain a suitable size and numbers of anchor box. In addition, we modify modifying detection scale and the backbone network structure of Yolov3-tiny, improving the detection accuracy for small object. The stereo vision is also introduced to improve the accuracy of boundary location. Experiments results demonstrate that the improved yolo-tiny has higher accuracy than the original algorithm and it also meet the requirement of real-time performance

Keywords: Automated and Connected Vehicle, Advanced Driving Assistant System
Abstract: Abstract：Vehicle target recognition technology is an important technology in the auxiliary safe driving system, which greatly improves Vehicle safety assist driving. This paper proposes a method to identify vehicle target recognition with lidar point cloud data and machine learning; This method first establishes an ROI (region of interest), and uses voxel grid filter to downsample the lidar point cloud data in this area to reduce the amount of processed data，then use RANSAC (random sampling consensus) to remove ground points that are useless for the recognition process，and then use Euclidean clustering for clustering.A rough classifier is set to initially eliminate obstacles that cannot be vehicles, then the features of each cluster is extracted.SVM (Support Vector Machine) is used as an accurate classifier, the parameters of SVM is optimized through cross-validation and grid search to achieve the best classification effect.Finally,the optimized SVM is used to identify each cluster. Experiments show that this method can effectively detect the target vehicle in the ROI and has a good recognition accuracy.

Keywords: Advanced Driving Assistant System, Electrification of vehicle
Abstract: Aiming at the functions and characteristics of the EPS power steering system, power model of electric power steering is established, the ideal power-assisted characteristic curve is analyzed, the EPS steering system test is designed based on the bench, and the driving simulator hardware-in-loop test is carried out. The effects of EPS power steering in stationary and motion state are tested. The results show that the experimental results are accord with the ideal assist characteristic curve, and the designed driving simulator hardware-in-loop EPS experiment has a good assist effect, which is beneficial to improve the operating sensitivity and steering stability of electric vehicles.

Keywords: Vehicle Dynamics and Control, Vehicle Powertrain Control
Abstract: Abstract—Distributed drive vehicles adopt multi-motor distributed drive, which makes it show advantages that traditional vehicles do not have, at the same time, it results in new problems to the reliability of the system. This paper takes four-wheel independent driven vehicle as the research object. Aiming at the failure of the in-wheel motors of distributed drive vehicles,a distributed motor drive vehicle model is built based on Carsim and Simulink. Based on the simulation verification of the established model, the simulation experiments for failure conditions are designed ,and the results are analyzed. The results show that the established model basically satisfies the experimental requirements, and the failure analysis can provide a basis for subsequent fault-tolerant controller design.

Keywords: Electrification of vehicle, Vehicle control
Abstract: Although pure electric vehicles have prominent advantages in environmental protection and motor technology has become more and more perfect, the competitive disadvantage of pure electric vehicles still lies in their lack of endurance. For lack of pure electric vehicle battery life of this problem, this paper analyzes the basic theory of pure electric vehicle braking energy recovery, put forward a kind of pure electric vehicle braking energy recovery based on flywheel energy storage and optimize management strategy, further studied the braking energy recovery system under this policy, and established the simulation model, simulation software for pure electric vehicle model and the design of pure electric vehicle model under the three typical driving cycle simulation. The simulation results were analyzed. The analysis results show that under the condition of ensuring the safety and stability of the pure electric vehicle in the braking process, the energy consumption rate of the pure electric vehicle set up in this paper is reduced by 4.1%, which improves the energy utilization rate of the vehicle, recovers more braking energy, and improves the endurance capacity of the pure electric vehicle to a greater extent.

Keywords: Decision Making
Abstract: Clean and efficient fuel cell(FC) power systems have shown great potential as an alternative to distributed energy resources. Fuel cell interconnection can relieve the pressure on the grid and meet emergency power needs. A strategy of fuel cell energy management based on evolutionary game is proposed. In the game, the fuel cell energy scheduling problem is treated as a multi-population scenario. Each part of the population has its own mixing strategy. On the other hand, there is a corresponding relationship between pure strategy and mixed strategy. Thus, the strategy here can flexibly meet different demands of power grid. In order to verify the feasibility of this method, the performance of the proposed approach is tested on real data measured on a distribution transformer from the SOREA utility grid company in the region of Savoie, France. The simulation results are compared with the dynamic programming results to further verify the effectiveness of the control strategy.

Keywords: Hybrid Electric Vehicle, Vehicle Powertrain Control, Electrification of vehicle
Abstract: Aiming at the special needs of heavy-duty hybrid electric vehicles(HEVs), an adaptive energy management strategy based on frequency domain power distribution is proposed. This article uses MATLAB/Simulink to establish a dynamic model of a heavy-duty HEV. First, the NARX neural network is used to predict the vehicle speed. Secondly, according to the predicted vehicle speed, principal component analysis and K-means clustering method are used to classify the working conditions, the corresponding control parameters are adjusted adaptively according to the working conditions category, and the power is distributed in the frequency domain. A piece of real vehicle driving cycle data of the vehicle is used as the simulation condition to verify and analyze the strategy. The simulation results show that this strategy can quickly restore the deviated battery state-of-charge (SoC) to the target value and maintain it stably. The battery's charge and discharge current amplitude are effectively reduced, at the same time, the transient working conditions of the engine are reduced too, and therefore the engine can work on the optimal efficiency curve. It is verified that this strategy is an effective real-time energy management strategy for heavy-duty HEVs.

Keywords: Fuel Cell Vehicle
Abstract: In this paper, the advantages and disadvantages of compressed gaseous hydrogen system and liquid hydrogen system are compared and analyzed from the perspective of hydrogen storage density and energy consumption. It is proved that on-board liquid hydrogen system is the most economical and efficient and environmentally friendly choice for heavy-duty trucks which are running continuously at high speed and long distance. This paper also conducted a joint test of a large-capacity liquid hydrogen system and a fuel cell system. The results show that the on-board liquid hydrogen system can provide relatively stable pressure for the fuel cell stack at dynamic and steady-state conditions. The liquid hydrogen system is highly compatible with fuel cell system and can be applied to high-power fuel cell system. This study makes it promising to develop a new generation of heavy-duty fuel cell truck with long driving range, high load capacity and high power.

Keywords: Fuel Cell Vehicle
Abstract: Being a promising alternative to internal combustion engines, the power output of a fuel cell stack is strongly dependent on its operating conditions. The fuel cell stack is integrated with auxiliary components to form a system that ensures optimal performance at the current layout. Here, we evaluated the key requirements of different components, designed and integrated an 80kW fuel cell system for automotive applications. The fuel cell system performance, with emphasis on the control accuracy of hydrogen and air sub-systems, is demonstrated. The system has peak power of 89.6 kW and peak energy efficiency of 57%. For the air subsystem, the air pressure and flow rate are controlled, both of which could achieve <5% control error. For steady state process, the control errors of the air flow rate and air pressure are <3% and <1%, respectively. The control error of the hydrogen subsystem is within 2%. This paper discusses the status of these system development activities and summarizes the results obtained to date.

Keywords: Fuel Cell Vehicle, Vehicle Powertrain Control
Abstract: Precise control of cathode pressure and flow rate is critical to the performance and durability of proton exchange membrane fuel cell systems. This study presents a model of the cathode subsystem of fuel cell engines, and the degree of air flow rate-pressure coupling in different working areas of air compressor is analyzed. A decoupling control algorithm based on the active disturbance rejection control is then designed to realize precise regulation of air pressure and flow rate. Finally, experiments are conducted on a domestic 80kW fuel cell engine, and the effectiveness of the control algorithm is validated. The experimental results indicate that the designed control algorithm has strong flow rate-pressure decoupling ability with fast response and high control accuracy.

Keywords: Fuel Cell Vehicle, Hybrid Electric Vehicle, Vehicle Dynamics and Control
Abstract: Fuel cell vehicles are a trend in the development of new energy vehicles in the future.However, the degradation of PEMFC has always been one of the reasons restricting its development.This paper studies a PEMFC degradation model based on ECSA to determine the appropriate working state to reduce the rapid degradation of PEMFC.This model mainly simulates the dissolution and maturation of platinum clusters in the catalytic layer and explains the decline of ECSA. The simulation results show that the model can well correspond to the phenomenon of accelerated aging of the fuel cell due to variable load under vehicle operating conditions.It is suggested that the model facilitates the development of a hybrid energy management system for vehicles aimed at fuel cell degradation.

Keywords: Fuel Cell Vehicle, Hybrid Electric Vehicle
Abstract: In order to assess the vehicle emissions and energy consumption in actual driving, the accurate vehicle driving cycles are extremely necessary. On the basis of the previous driving cycle’s construction methods, the innovation of this paper is proposing a method for constructing urban driving cycle based on simulated annealing algorithm. The major task is the data processing and optimizing. For data processing, the characteristic parameter of the micro-trips is selected according to the theory of micro-trips analysis, then this paper performs principal component analysis to reduce the dimensions of motion characteristic parameters and the K-means clustering method is used to classify kinematics segments. In the selection of fragments, this paper adopts the simulated annealing algorithm to optimize. The final analysis results show that the error is largely reduced and the accuracy of the operating conditions is further improved.

Keywords: Intelligence of vehicle, Vehicle control
Abstract: In this paper, we propose an end-to-end controller for self-driving vehicles based on visual attention. Attention strategy is used to weight the high-dimensional feature information extracted by convolutional neural networks (CNNs), and then the vehicle's velocity and steering wheel angle are predicted by different recurrent neural networks (RNNs). The end-to-end controller is trained on Comma.ai dataset and can effectively reduce the mean absolute error (MAE). The result shows that compared with other models, the end-to-end control model based on visual attention can achieve better control effects of vehicle’s speed and steering wheel angle.

Keywords: Intelligence of vehicle, Vehicle Dynamics and Control, Integrated Chassis Control
Abstract: In order to improve the accuracy of trajectory tracking and handling stability for high-speed autonomous vehicle in off-road conditions, a novel trajectory tracking layered coordinated control strategy based on future driving state prediction for autonomous four-wheel independent drive electric vehicle (A-4WID-EV) is proposed, For the upper controller, a driving state prediction algorithm based on the variable-order Markov model with dynamic window is proposed to predict the driving state in the future. For the lower controller, an active front wheel angle control strategy based on multi-scale model predictive control (MPC) is designed to provide vehicle front wheel angle. Meanwhile, a coordinated four-wheel drive torque control strategy based on the future driving state is proposed to ensure the lateral stability during the trajectory tracking. Finally, through the CarSim-Matlab/Simulink co-simulations, the results show that the proposed controller can effectively improve accuracy trajectory tracking and lateral stability of high-speed A-4WID-EV in off-road conditions.

Keywords—high-speed A-4WID-EV, trajectory tracking, handling stability, future driving state prediction

Keywords: Intelligence of vehicle
Abstract: In the field of automated driving, the accurate and efficient trajectory estimation is the basis for real-time state estimation of surrounding vehicles, which is the prerequisite for subsequent tasks such as correct decision-making, proper planning and control. At present, the exploration of trajectory estimation task focuses on complex observation models and ignores further optimization of the search strategy itself. This paper proposes a deep particle filter based trajectory estimation framework, which combines the distribution estimation ability of conventional particle filter and the powerful data-driven expressive ability of neural networks.

Its state transition model is based on the Long-Short-Term-Memory (LSTM) neural networks, while its observation model is made differentiable by the sequential regression method based on a soft resampling trick, which forms a unified learning framework. Experiments on the NGSIM dataset demonstrate the proposed approach with both a good trajectory estimation accuracy and a noise robustness.

Keywords: Electrification of vehicle, Intelligence of vehicle, Automated and Connected Vehicle
Abstract: In this paper, the optimization of flux-intensifying interior permanent magnet motor with the reverse salient rotor for electric vehicles is considered and explained. Firstly, the size parameters of an initial motor are selected and then the finite element model is established based on parametric variables. Secondly, to avoid the frequent usage of finite element analysis, a well-trained back propagation neural network model is used to replace the finite element model. Thirdly, the sequential unconstrained minimization technique and non-dominated sorting genetic algorithm-II algorithm are combined together to solve the multi-objective optimization solution with inequality constraints. Finally, the electric machine is reconstructed based on the optimal parameters extracted from Pareto front. The effectiveness of proposed approach is verified by the simulation results.

Keywords: Intelligence of vehicle, Control of UAV/USV/UUVs
Abstract: This paper presents a VPX-based computing platform for aircrafts with an artificial intelligence module built in. VPX bus connects together a serial computing module, a parallel computing module and an AI computing module, providing all the computing resources for the platform. These modules can communicate with varying topological types through SRIO connection, with a maximum data transfer rate of 2.5Gbps. The AI module, which consists of an ARM-based CPU and a machine learning unit focusing on accelerating neural networks, has been tested to show the ability to handle ImageNet classification tasks up to 2214 frames per second when using the TensorFlow framework.

Keywords: Intelligence of vehicle, Advanced Driving Assistant System, Automated and Connected Vehicle
Abstract: This paper proposes an improved global path planning algorithm to generate the optimal global path that satisfies the kinematic constraints of parking robots. The estimation function is improved through BP neural network, which improves the planning efficiency of finding the shortest path. Improve the drivability of the planned route by setting up the prohibited area and the route backtracking. A simulation platform is built based on ROS, and the path planning effect of the traditional A* algorithm is compared with the effect of the improved global path planning algorithm. The results show that the improved algorithm has a shorter path length and better drivability. The overall deviation of the simulated trajectory driving along this path is small. The improved algorithm is used to conduct multiple terminal path planning experiments. The results show that the total length of the path generated by the algorithm is close to the global optimum, the path is smooth and easy to track.

Keywords: Intelligence of vehicle, Vehicle control
Abstract: The active anti-collision technology can ensure vehicles to avoid car accidents in a short time, since vehicle collision accounts for a large proportion of most traffic accidents. It has become an important research content of active vehicle safety. In this paper a double-layer controller of trajectory planning and tracking is designed to improve driving safety for intelligent vehicles. Firstly, 2-DOF(degree-of-freedom) vehicle dynamics model is established based on the intelligent vehicle platform with front wheel steering. Secondly, the point mass vehicle model is applied to design the collision avoidance function and the trajectory re-planning function. The problem of anti-collision trajectory planning is transformed into a nonlinear quadratic planning problem with the constraints of speed, angle and angle increment. Based on MPC algorithm, the anti-collision trajectory in dynamic environment is re-planned, and the trajectory planner of active collision avoidance model for intelligent vehicle is established. Thirdly, a fuzzy adaptive PID trajectory tracking controller is employed as the trajectory tracking layer of this anti-collision model. Finally, the collision avoidance effect is simulated in Matlab/Simulink environment. The results show that the model has better collision avoidance effect and helps to track the original driving trajectory at 36km/h speed. The vehicle tests show that the dynamic trajectory planning algorithm based on MPC can ensure the vehicle to plan the anti-collision trajectory in real time when sensing obstacles, and the fuzzy PID algorithm as the trajectory tracking layer can ensure the tracking accuracy of the original trajectory after collision avoidance.

Keywords: Intelligence of vehicle, Advanced Driving Assistant System
Abstract: The active safety technology of vehicles has become a hot research topic nowadays, among which the intelligent tire is an important field. The existing tire pressure monitoring system (TPMS) however, cannot provide a comprehensible supervision for the tires. This paper proposed an intelligent tire information system, which can estimate the tire abrasion depth, based on triaxial accelerometer data and strain gauge data. The system consists of different sensors in the data acquisition section, a controlling chip in the car terminal and the display section on mobile devices. Based on the data analysis from the sensors and multiple useful features extracted from the data, a BP neural network is established to training an abrasion depth prediction model in MATLAB. At last, we conduct experiments to verify our models with 943 collected data and compare the model with the multivariable linear regression model. The absolute errors of randomly chosen testcases can all keep within ±0.5mm and about 80% of them lies in the range of ±0.2mm, which shows the efficiency and reliability of our system and prediction model.

Keywords: Intelligence of vehicle, Automated and Connected Vehicle, Advanced Driving Assistant System
Abstract: Aiming at the problems of high cost and limited installation of traditional unmanned vehicle environment perception methods, this paper proposes a method of personnel identification and distance measurement based on the fusion of YOLOv4 and binocular stereo vision. Through the annotation of the data set, the Darknet deep learning framework is used to train and recognize the personnel, and the binocular camera disparity data is used for personnel distance detection. The experimental results show that the recognition accuracy of this method is 0.941 and the distance error is less than 5%, which can meet the task requirements of unmanned vehicle and provide technical support for solving the environment perception problems of autonomous driving vehicle.

Keywords: Intelligence of vehicle, Vehicle Dynamics and Control, Vehicle control
Abstract: combined simulation; automatic guided vehicle; ROS; model predictive control; path tracking control

Keywords: Intelligence of vehicle, Automated and Connected Vehicle, Advanced Driving Assistant System
Abstract: Unmanned vehicle can be used as a transport tool for teams and groups to accompany and follow soldiers, reduce the load of team members and identify team members accurately in real time. It is a prerequisite for the realization of control algorithm and one of the core technologies for automatic control of military vehicles. Aiming at the problem of personnel identification under the fusion perception of lidar and camera, especially the problem of multi-sensor space and time synchronization, this paper proposes a solution based on multi-sensor fusion, and designs fusion criteria of space scale, time scale and personnel identification. The experimental results show that a designed personnel identification algorithm based on multi-sensor space and time fusion can accurately identify personnel targets in complex environment, and the intersection ratio of lidar and camera fusion algorithm exceeds 95%.

Keywords: Intelligence of vehicle, Automated and Connected Vehicle
Abstract: The roller is a typical articulated multi-body vehicle with multi-degree of freedom in motion. Accurate and reliable position and heading angle measurements are important foundations for the accurate path-following of unmanned rollers. Due to the poor operation environment of the roller, the positioning signal often drifts or jumps, which affects the reliable operation of the system. To achieve reliable fault diagnostic in the positioning system, in this paper, a novel solution that combines total disturbance observation and support vector machine (SVM) classification, is proposed. A multi-body kinematic model is established with steering wheel angle and vehicle speed as inputs, and with the longitude, latitude and heading angle as outputs. The discrepancy of model estimates from the measured value is treated as total disturbance, to be estimated by the extended state observer. Then the estimated total disturbance, together with the measured position and heading angle are input into the support vector machine for faults classification. Experimental results show that the fault diagnosis accuracy is 95%, the improvement in accuracy and computational time is 9% and 12% respectively, compared with the conventional solution that only based on SVM.

Keywords: Intelligence of vehicle, Control of UAV/USV/UUVs, Vehicle control
Abstract: Urban air mobility(UAM) is an emerging way of air transportation by using self-pilot electric vertical takeoff and landing (eVTOL) vehicles to transport people and cargo in an urban area. The final approach is the critical stage in the operation. Accurate location information and vertical height guidance are needed in this stage. In addition to the common GNSS/INS integrated navigation system, this work proposes an independent vision-based method to provide redundant location information in the final approach for eVTOL vehicles. The simulation results show that the positioning accuracy of the visual method is within 1.5 meters when the vehicles are no more than 25 meters away from the reference place.

Keywords: Intelligence of vehicle, Advanced Driving Assistant System, Automated and Connected Vehicle
Abstract: Accurate vehicle speed prediction has important practical value to enhance fuel economy, drivability, and safety of intelligent vehicles. Current research on vehicle speed prediction mainly focuses on adapting to the dynamics, random and complex driving environment, while rarely takes drivers' driving preferences into account. In this paper, a learning-based prediction model consisted of an oriented Hidden Semi-Markov model (Oriented-HSMM) and an optimal preference speed prediction algorithm is proposed to leverage drivers' driving preferences into vehicle speed prediction. The Oriented-HSMM is developed to learn the spatial-temporal coherence of drivers' driving preference states under different traffic conditions and infer its long-term sequences in position domain. Based on these preference states, the optimal speed prediction algorithm using preference dynamics features is designed to retrieve the speed trajectory with maximal likelihood. To show its effectiveness, the proposed method is tested with the Next Generation Simulation (NGSIM) data on the US101 dataset comprising with the Hidden Markov model (HMM) and HSMM without considering driving preferences. Experiment results indicate that the proposed algorithm obtains the best performance with the mean absolute error (MAE) of 4.15 km/h and the root mean square error (RMSE) of 0.7603 km/h at 200 m prediction horizon.

Keywords: Vehicle control, Automated and Connected Vehicle, Decision Making
Abstract: Cooperative adaptive cruise control (CACC) has important significance for the development of the connected and automated vehicle (CAV) industry. In this paper, a learning control method combined Deep Deterministic Policy Gradient and Proportional-Integral-Derivative (DDPG-PID) controller is proposed. The main contribution of this study is automating the PID weight tuning process by formulating this objective as a deep reinforcement learning (DRL) problem. Based on the Hardware-in-the-Loop (HIL) simulation platform, the DDPGPID controller is compared with the conventional PID controller under the test condition. Experiment results indicate that on 38.95% stability time in vehicular platooning system is decreased by utilizing the proposed method. The performance of maximum distance error is also improved efficiently, which is reduced by 60.94%. The research in this paper is a further development of learning control method and provides a new idea for the practical application of DRL algorithm in industrial field.

Keywords: Advanced Driving Assistant System, Vehicle Dynamics and Control, Vehicle control
Abstract: The strong transient crosswind has huge impact on the driving state of the autonomous vehicles, which requires the control system with good handling stability. Here, a predictive compensation-based handling stability control system (PCHSCS) is developed. The PCHSCS includes three parts: a steering controller, a speed controller, and a predictor. The steering controller is adopted to control the course for resisting the lateral acceleration caused by the crosswind. The speed controller is utilized to maintain the vehicle speed, decreasing the undesired longitudinal acceleration. The predictor is applied to predict the control error and the coupling interference. The predicted information, as compensation, is combined and fed to the speed controller and the steering controller. Simulation proves the PCHSCS can improve the vehicle handling stability under strong transient crosswind conditions.

Keywords: Automated and Connected Vehicle, Vehicle Dynamics and Control, Vehicle control
Abstract: This study addresses the conundrum of the trajectory tracking control for a connected vehicle (CV) platoon on the straight road and curved road. To be specific, the communication topology of bidirectional leader-follower is used to describe the vehicles’ communication connection in the vehicle-vehicle/vehicle-to-infrastructure (V2X) environment. Then, a nonlinear state feedback control algorithm for follower vehicles is put forward through merging the phenomenon of car-following interactions, the spacing error ， velocity difference and angular difference in relation to the leader vehicle. The proposed algorithm’s stability is verified by means of Lyapunov technique. And then, simulation experiments under two scenarios: the leader vehicle runs on the straight road and curved road, respectively, and the followers run after the leader vehicle. The numerical simulation results prove the effectiveness of the proposed control algorithm.

Keywords: Hybrid Electric Vehicle, Vehicle Powertrain Control
Abstract: In view of the nonlinear characteristics of the hybrid energy storage system (HESS), a nonlinear control strategy based on feedback linearization theory is proposed. The control objectives of the strategy are battery/super capacitor current and DC bus voltage. Firstly, by analyzing the state equation of the circuit, the affine nonlinear model of the circuit is established. Then, the input-output feedback linearization and sliding mode surface are designed to realize the current tracking control of the battery and super capacitor. Finally, the classic voltage closed-loop control is used to realize the bus voltage stability. The simulation model is established on Simulink, and the simulation results show that the proposed control strategy is effective.

Keywords: Intelligence of vehicle, Automated and Connected Vehicle
Abstract: The unmanned roller, as an important technical means to reduce manual labor intensity, can improve the operation efficiency and quality. Due to the change of road surface conditions, it is difficult to drive in a straight-line accurately. As a result, some area that needs to be rolled missed.So it is necessary to select appropriate repeated rolling width (referred to as overlapping distance) for adjacent strips according to real-time rolling conditions. Therefore, in this paper, a self-learning online optimization method is proposed, to optimize the overlapping distance in the cloud computing center. To this end, a multi-objective cost function is proposed, considering impacts of overlapping distance on missed rolling percentage, over-rolling percentage and operation efficiency, which turns the problem of overlapping distance online adjustment into a convex optimization problem. In the cloud side, with the minimum cost function as the goal, the best overlapping distance is searched in real-time by using the mountain-climb searching algorithm, and then sent to the real-time control system of unmanned roller. Finally, the proposed algorithm is validated in experiments. Results show that compared with the fixed overlapping distance algorithm, the over-rolling ratio can be reduced by 4.05% and 5.06%, the fuel consumption can be reduced by 4.10% and 1.83% in the flat transition zone and the large-size rockfill zone respectively.

Keywords: Vehicle Dynamics and Control, Vehicle control, Automated and Connected Vehicle
Abstract: Trajectory tracking problem of autonomous vehicle under lane change is considered in this paper. To improve trajectory tracking performance in real scene, an adaptive tracking controller with finite time convergence performance is proposed to deal with the uncertain factors in the lane change scenario. The effectiveness and advantages of the proposed adaptive finite time control strategy (AFTCS) are illustrated by the simulation comparison results with the conventional PID tracking controller.

Keywords: Advanced Driving Assistant System, Human-Computer Interaction
Abstract: A belt-type passive exoskeleton equipped with multiple working modes has been designed to support the back muscles during manual lifting tasks. Our concept is to develop a wearable assistive device that can provide motion-based assistance like the existing passive devices in the down phase of lifting, but in the up phase, can supply more load-based power. To achieve this goal, we designed two purely mechanical control mechanisms that can preload the load-based assist force and release it when the wearer intends to erect the trunk. This paper presents mathematical proof using a linked segment model and moment balance equations in the sagittal plane. Simulation proof is also provided based on a simple musculoskeletal model which executed sagittal plane lifting with a 20kg load in hands in stoop posture under three different conditions (no assist, classic mode, enhanced mode). The subsequent results suggest that the device substantially reduces the muscle force and lumbar moment by 27.7-43.5%, and in the enhanced mode, the efficacy is better than the classic mode. No such a quasi-passive back-support exoskeleton was found at the time of writing. This design concept is promising to help reduce the risk of back injuries in heavy load lifting and works that need workers to keep a forward bending and static holding postures such as vehicle assembly.

Keywords: Advanced Driving Assistant System, Human-Computer Interaction
Abstract: Highly automated vehicle has the possibility in getting stuck with edge scenarios where the automation cannot handle. Under this circumstance, sending out a takeover request and dragging the driver back into the control loop are required to avoid traffic accidents. Among various possible modalities for alerting drivers about take-over requests, vibrotactile alerts provide significant advantages. A driver-in-the-loop and hardware-in-the loop driving simulator was designed for the investigation of take-over performance. In this simulator, take-over signal was provided the vibration motors embedded in the vibrotactile seat. Moreover, body pressure mapping test illustrated that the vibration motors fixed in the vibrotactile seat would not reduce seating comfort. Twentyfour vibration patterns were generated via the vibration motors embedded in the backrest and cushion of the vibrotactile seat. Besides, Eighteen participants were recruited to take part in the experiment, which consisted of three sessions: 1) baseline (no driving task), 2) HAD (driving a highly automated vehicle and getting ready for the respond to the take-over request), 3) N-back (performing the same task with mental demanding task added in). Specifically, in baseline session, participants need the only answer regarding the type of vibration pattern. However in HAD and N-Back session, participants had to perform the maneuver (steering left/right or braking) according to the coding directional information of vibration patterns. Correct response rate and reaction time of each participant in each session were recorded and analysed. The results indicated that dynamic patterns yielded significantly higher correct response rate than static patterns. In addition, reaction times for dynamic patterns were faster than those for static patterns, but the effect was not statistically significant. Moreover, ANOVA tests illustrated that mental-demanding non-driving task had no significant effect on take-over performance.

Keywords: Automated and Connected Vehicle, Human-Computer Interaction, Intelligence of vehicle
Abstract: Automated vehicle testing and evaluation is an important guarantee for vehicle safety and reliability. The current L3 automated vehicle evaluation and evaluation procedures are not yet perfect. For the field test of L3 automated vehicles, we proposed to establish a comprehensive evaluation index system from the five dimensions of safety, intelligence, experience, energy consumption, and efficiency. A scientific method was designed to select and screen indicators in each dimension, and to preprocess behavior indicators based on effect size. The analytical hierarchy process and entropy method were used to determine the index weight, and the BP neural network and grey relation analysis were used to establish two comprehensive evaluation models for automated vehicles. Taking the comprehensive evaluation of the safety of automated vehicles in highway conditions as an example, two comprehensive evaluation models were established to verify the effectiveness of the models.

Keywords: Decision Making
Abstract: With the consideration of personalized driving for automated vehicles (AVs), this paper presents a human-like decision making framework for AVs. In the modelling process, the driver model is combined with the vehicle model, which yields the integrated model for the decision-making algorithm design. Three different driving styles, i.e., aggressive, normal, and conservative, are defined for human-like driving modelling. Additionally, motion prediction algorithm is designed with model predictive control (MPC) to advance the effectiveness of the decision-making approach. Furthermore, the decision-making cost function is constructed considering drive safety, ride comfort and travel efficiency, which reflect different driving styles. Based on the decision-making cost function, a noncooperative game theoretic approach is applied to solving the decision-making issue. Finally, the proposed human-like decision making algorithm is evaluated with an overtaking scenario. Testing results indicate different driving styles cause different decision-making results, and the designed algorithm can always make safe and reasonable decisions for AVs.

Keywords: Vehicle Dynamics and Control, Integrated Chassis Control, Human-Computer Interaction
Abstract: In order to obtain the accuracy of vehicle parameters and occupant injury information after collision, and reduce the high cost of the sample vehicle crash test and the uncertain factors in the collision process,build rigid body model and multi-rigid body model by PC-Crash platform of traffic accident reconstruction software. Set the parameters and initial motion state of the model, and carry out the simulation and reconstruction of the traffic accident collision, which can obtain instantaneous motion state parameters of the collision vehicle and occupant. To simulate and demonstrate the process of the traffic accident, comparing the demonstration results with the actual traffic accidents, and the reduction rate of the real motion pattern is more than 95%,so that the effectiveness of this method is verified. It provides a reference for the better use of PC-Crash software to realize the simulation analysis of occupant's frontal injury, and provides a theoretical basis for accident treatment.

Keywords: Integrated Chassis Control, Vehicle control
Abstract: Most vehicle active suspension control systems assume that the dynamic system model descriptions are accurate. However, there may exist modeling error and external disturbances for real world applications. While extensive research in robust model predictive control has been considered to handle such issues, the control performance may degrade due to the conservation of the prior uncertainty set. In this work, a vehicle active suspension control problem with modeling error and external disturbances is studied. We propose an adaptive tube-based model predictive controller to identify parameter uncertainty set and optimize reformulated quadratic optimization problem (QOP) for increasing control performance. The recursive feasibility and stability analysis of the proposed method is presented, and simulation results are demonstrated to indicate the effectiveness of the proposed algorithm.

Keywords: Automated and Connected Vehicle, Intelligence of vehicle, Advanced Driving Assistant System
Abstract: Vehicle chain collision is considered to be one of the main causes that lead to serious casualties and reduce traffic efficiency. In this paper, the dynamic process of chain collision is investigated based on the virtual mass-spring-damper (MSD) model, and the ideal values of virtual spring constant and damping coefficient are determined for collision avoidance by optimization method. Simulations have been performed when a vehicle slows down suddenly on a single-lane road, and the causes that lead to chain collision have been discussed. It can be seen from the simulation results that connected vehicles break the trend of decreasing inter-vehicle spacing as it propagates down the line of platoon, and the probability of chain collisions in the vehicular platoon is reduced and the overall safety of the platoon is improved.

Keywords: Decision Making, Automated and Connected Vehicle, Advanced Driving Assistant System
Abstract: Lane change is receiving attention in academia. Most existing lane changing models are rule-based lane changing models which only assume one-direction impact of surrounding vehicles on the lane-changing vehicle. In fact, lane change is a process of mutual interaction between vehicles due to the complexity and uncertainty of the traffic environment. In this paper, we proposed a multi-vehicle cooperative control approach with a distributed control structure to control model. The innovation of this paper lies in that we proposed a multi-vehicle cooperative lane changing controller which combines game theory and model predictive control (MPC) based on vehicle to vehicle (V2V) communication; Moreover, we designed a multi-lane vehicle ordering method, and decided the optimal time and acceleration of lane change by considering the mutual interaction between vehicles. Typical scenarios were tested to verify that a lane changing vehicle could interact with other vehicles and change lanes without collision. We verified this approach of lane changing through CarSim and MATLAB co-simulation, and compared it with the conventional rule-based lane change decision controller. Test results show that the controller is capable of changing lanes in a smarter manner and guaranteeing the safety and efficiency of the autonomous vehicle.

Keywords: Advanced Driving Assistant System, Intelligence of vehicle, Vehicle Dynamics and Control
Abstract: Driving comfort, which is mainly influenced by vibration and shock, is an essential factor to evaluate the performance of intelligent vehicles. The evaluation methods of driving comfort mainly contain subjective and objective evaluation. Subjective evaluation is time-consuming, expensive and sensitive to personal feelings. And objective evaluation is difficult to exactly define the relationship between objective parameters and driving comfort. In order to combine the advantages of subjective and objective evaluation, a neural network that adopt objective indicators as input and subjective ratings as output was established for evaluating driving comfort. First, a road test with about 9000 km was conducted and key parameters of vehicle status were recorded, as well as subjective ratings. Secondly, 25,165 segments were extracted from the naturalistic driving data. Then, total weighted root-mean-square accelerations of all segments were computed according to ISO 2631-1997 Standard. And the result shows that the comfort levels calculated by weighted root-mean-square accelerations cannot match the subjective ratings given by professional evaluators very well. Finally, a 20-128-256-256-128-6 BP neural network was established and trained. And the accuracy of evaluation based on neural network is better than evaluation based on weighted root-mean-square value. The result reveals that it is feasible to establish a neural network model based on collected naturalistic driving data to evaluate the driving comfort of vehicles.

Keywords: Automated and Connected Vehicle
Abstract: For the application of autonomous ground vehicle (AGV) operating in unstructured environment, a path planning method based on an improved goal-biased Rapidly-exploring Random Trees (bias-RRT) is proposed. The algorithm combines random sampling with numerical optimization to achieve fast convergence speed and satisfy constraints. KD-Tree and potential field of the environment are implemented to increase the sampling efficiency, and cubic B-splines are used to smooth the path for better tracking performance. The algorithm improves the efficiency of searching while guarantee safety and quality of the planned path. Simulation results verify the effectiveness of the proposed method.

Keywords: Automated and Connected Vehicle, Decision Making, Intelligence of vehicle
Abstract: Traffic complexities in no-signal intersections lead to amounts of accidents, among of which are due to inappropriate decision based on inconsiderate judgements of the other traffic users. Focusing on an example intersection driving scenario, this paper analyses the decision-making behaviour of two crossing vehicles at intersections without traffic lights, while considering the influence of safety factor, traffic efficiency and drivers’ irrationality, etc. We propose a corresponding utility model to treat the whole dynamic process as finite repeated games. Nash Equilibrium approach is adopted to solve the decision-making problem at intersections. The effectiveness of the proposed decision algorithm is validated by both simulation and human-in-the-loop experiments.

Keywords: Hybrid Electric Vehicle, Vehicle Powertrain Control, Vehicle control
Abstract: HIL test is a typical application of semi-physical simulation. It is an important function verification and test link in the development process of automotive electronic control systems. It has become a necessary means in the standardized development process, and it has been increasingly affected by various automotive OEMs and component manufacturers. Wide attention. This article focuses on the HIL test environment of the new energy vehicle TCU control system, introduces the HIL test system architecture, software and hardware components, and the establishment process of the test environment. The test requirements are formulated according to the application scenarios of the TCU in the new energy vehicle, and the test requirements are based on the test requirements. Set up the HIL test environment. Considering that the motor speed regulation in practical applications is controlled by the vehicle controller, in order to more realistically simulate the vehicle use environment, this paper adopts the VCU and TCU dual-in-the-loop method for HIL testing, aiming at the functions of the electronic control unit controllers of the power system. Simulate the vehicle environment for testing and analyze the test results.

Keywords: Vehicle Powertrain Control, Vehicle control, Hybrid Electric Vehicle
Abstract: By analyzing the principle and method of the direct vector control torque control method for induction motor, an analysis method of the influence of motor parameter changes on the accuracy of motor electromagnetic torque control under steady-state conditions is proposed, and the application of four common flux observations is derived. In direct vector control, the error function of the electromagnetic torque is estimated. The influence of the change of induction motor parameters on the torque control accuracy is theoretically analyzed, and the simulation is verified by Simulink. The simulation proves that the method can correctly analyze the torque control accuracy under the condition of induction motor vector control without complicated calculations.

Keywords: Vehicle Dynamics and Control, Vehicle control
Abstract: For the problem of ecological cruise control integrating the terrain information of the road ahead, in order to input the actual slope changed with time over the prediction horizon, a linear time-varying model predictive controller is designed in this paper based on the traditional theory of model predictive control. The analysis of vehicle driving on flat road and hilly road is simulated by MATLAB/Simulink, results show that it has a good effect on energy saving at the cost of shorter distance. In addition, the energy consumption under different control targets and the influence of weight coefficients on vehicle economy under the same target are also analyzed and compared.

Keywords: Automated and Connected Vehicle, Vehicle control, Intelligence of vehicle
Abstract: Driver-automation co-piloting, a driving mode under which autonomous driving systems and human drivers accomplish driving tasks cooperatively is expected to be widely used to reduce driver workload in future driving. The work presented in this paper focuses on safety evaluation of the transition mechanism between autonomous system and human drivers. A group of two-factor experiments, in which two factors are: (1) advance responding time for drivers:15s,45s, (2) notification modes to drivers: audio, visual, audio/visual, were performed to quantitatively measure driver workload by using eye tracking data, which is highly relevant to driving safety. The results of these experiments indicate that drivers’ workloads increased more smoothly when given audio notification and more responding time during transitions. The research has brought about a solution to ensure a good level of driving safety in co-piloting.

Keywords: Vehicle Powertrain Control, Electrification of vehicle, Integrated Chassis Control
Abstract: This paper designs a novel collaborative control approach, including the longitudinal and lateral motion control, to guarantee the vehicle stability by the estimated vehicle states of electric vehicles. A nonlinear observer is developed to observe the lateral velocity and tire-road friction coefficient by a Dugoff's tire model. Moreover, a Lyapunov-based method is utilized to prove that the estimated errors converge to zero. The collaborative control is converted into a tracking problem by establishing a reference model. According to the estimated vehicle states and reference model, a passivity-based control strategy based on the port-Hamiltonian model is adopted to follow the referenced vehicle states and ensure the stable planar motions, and the asymptotic stability of the proposed controller is proved. In addition, a wheel torque distribution considering the transfer of vertical loads is designed to maximize the utilization of tire adhesive forces. Finally, simulation cases demonstrate the effectiveness of the designed nonlinear observer and controller.

Keywords: Human-Computer Interaction, Decision Making
Abstract: The pupil diameter is one of the most important assessing indicators for the mental state of drivers. On the other hand, it is still a topic to quantitatively assess the effect of vehicle speeds on pupil diameters of drivers. Therefore, this study recorded pupil diameters under different vehicle speeds of 40, 60, and 80 km/h in the driving simulator experiment. Furthermore, this paper analyzed pupil diameters on the straight and curve road sections, and the difference of pupil diameters between left and right eyes. The results indicate that there is a positive association between the pupil diameter and the vehicle speed, and there is a difference for pupil diameters in the straight and curve driving. In addition, the difference of pupil diameters between left and right eyes is higher in the straight section than that in the curve section.