Keywords: Vehicle Dynamics and Control, Vehicle control, Control of UAV/USV/UUVs
Abstract: This paper investigates the intelligent driving control problem for a class of automatic vehicles by using a developed policy iteration method. Firstly, we analyze the automatic vehicle's motion with respect to its linear and rotational velocities. By this way, the dynamic function based on kinematic equation is built, then, combining the desired reference and the automatic vehicle's trajectories, the tracking driving dynamic is constructed. Secondly, according to the optimal control theory, a performance index for the tracking error system is designed, which converts the tracking driving problem into an optimal control problem. Besides, by iterating the Hamiltonian function and control policy, the policy iteration method is developed to solve the optimal tracking control solution. Finally, the simulation is applied to verify the developed tracking control algorithm, and the results demonstrate the effectiveness of the new scheme.

Keywords: Decision Making, Intelligence of vehicle, Automated and Connected Vehicle
Abstract: In recent years, machine learning is widely used in many fields. Compared with the rule-based method, machine learning plays a more excellent role in the decision-making of the autonomous vehicle. Some complex situations are often met in our daily life. To this end, Safe reinforcement learning(RL) is introduced to ensure that safer actions are selected. Constant Turn Rate and Acceleration(CTRA) model is first used to predict the future trajectories of surrounding vehicles. Then Double Deep Q-Learning(DDQN) method is used to make decisions and ensure the autonomous vehicle can move at the desired speed as much as possible. In order to achieve a safer decision-making, some safety rules are introduced. Finally, the algorithm is demonstrated in Simulation of Urban Mobility(SUMO) and has been proved to have an outstanding performance on such a complex scenario.

Keywords: Electrification of vehicle
Abstract: LLC resonant converter is widely used in electric vehicle (EV) charger for the advantages of low switching loss and high power density. However, its dynamic performance and robustness are easily influenced by multiple disturbance factors. This paper proposes a nonsingular fast terminal sliding mode (NFTSM) control strategy for the LLC resonant converter to improve the dynamic performance and robustness. First, the second-order small-signal model is obtained by the linearized and simplified large-signal mathematical model which is established based on the extended description function method. Then, the NFTSM controller is designed based on the small-signal model. And the system stability is proved by Lyapunov's stability theorem. Finally, Simulation results verify the feasibility and effectiveness of the proposed control scheme.

Keywords: Vehicle Dynamics and Control, Hybrid Electric Vehicle
Abstract: This paper presents two learning-based approaches to solve the optimal energy management problem for hybrid electric vehicles. It will be shown that by applying a learning algorithm to the interpolation of value-function, which is an optimal approximate value-function in continuous state space, the discretization error can be rejected when performing dynamic programming. Extreme Learning Machine and Gaussian Process Regression are exploited as learning tools. Finally, numerical simulation results with a parallel HEV will be demonstrated to show the effort of value-function learning.

Keywords: Intelligence of vehicle, Automated and Connected Vehicle, Vehicle control
Abstract: With the development of Intelligent Connected Vehicles (ICVs), Cloud Control Platform is becoming an important part to compute driving strategies. However, when strategies are put to cloud, some of vehicle manufacturers’ private data must be sent to the cloud, like Engine Map, which are core data for vehicle manufactures. How to protect these data has been the largest obstacle for the ICVs. Thus, this paper proposes a new framework in which Fully Homomorphic Encryption (FHE) and Blockchain technology are combined to compute encryption data on the cloud and record trails of cloud request. In this framework, private data can be protected, the scope of data usage will be limited, and at the same time, the execution of specific type of computations with encryption data on the cloud are fulfilled. In the end, with the help of Simple Encrypted Arithmetic Library (SEAL) developed by Microsoft Research, and IBM blockchain framework Hyperledger Fabric, this framework is verified to be feasible to build a trustworthy ICVs cloud computing system.

Keywords: Automated and Connected Vehicle, Vehicle Dynamics and Control, Vehicle control
Abstract: To improve the regeneration energy of electric vehicle, an energy-optimal braking strategy is developed. First, the vehicle braking intention is accessed by using vehicle-to-everything communication, i.e., braking distance and terminal velocity. Then, an optimal control problem with consideration of braking intention is formulated for maximizing regeneration energy. The control problem is solved by distance-based dynamic programming algorithm to plan the energy-optimal braking velocity. Finally, the effectiveness of proposed strategy is evaluated by simulation. The results show the regeneration energy efficiency of proposed strategy achieves improvement is over 10% compared with the constant speed strategy. Furtherly, the energy-optimal braking suggestions is investigated based on several traffic scenarios, i.e., a larger braking force in a high-velocity range can reduce vehicle resistance and make full use of motor generation power; the braking force was adjusted in moderated-velocity range for reducing friction braking, and a larger braking force should be used for parking quickly.

Keywords: Vehicle Dynamics and Control, Advanced Driving Assistant System, Automated and Connected Vehicle
Abstract: Advanced chassis control systems are essential to improve vehicle handling characteristics. Tire-road friction coefficient (TRFC) is a key parameter for these systems. Instead of directly measuring this parameter, the indirect estimation of TRFC will provide a cost-effective way for the implementation of vehicle chassis control systems. In this article, a novel adaptive cubature Kalman filter (ACKF) is proposed to estimate TRFC. First, a nonlinear vehicle dynamics model is established. Then, an improved cubature Kalman filter is used to estimate TRFC, which process noise can be updated dynamically based on lateral acceleration information. To verify the effectiveness of ACKF, some virtual validation tests are carried out. The test results indicate that the estimation performance of ACKF is better than the cubature Kalman filter.

Keywords: Automated and Connected Vehicle, Intelligence of vehicle, Vehicle control
Abstract: A velocity planning strategy for the autonomous-rail rapid transit (ART) based on the pseudospectral (PS) method is proposed in this paper. The PS method is used as a re-planning algorithm provided with its real-time performance. The multi-particle model is adopted in the dynamics model of ART. The energy consumption of ART is chosen as the optimization goal, and the arrival time and arrival velocity are taken as constraints to ensure economic efficiency and punctuality of the ART. When encountering obstacles such as pedestrians and lower-speed vehicles, the velocity planning strategy based on the PS method is applied to re-plan velocity of the ART. Performance of the proposed strategy is evaluated by comparing with the strategy to track the original velocity planned offline by dynamic programming (DP) algorithm. The simulation results in Matlab/Simulink-Trucksim environment illustrate that the PS based method has better real-time performance than the DP based method. The proposed planning strategy also makes ART arrive at the next station punctually, as well as leading to 36.21% reduction of energy consumption compared with the DP based method. Results of the jerk of the ART with PS strategy also show better performance in passenger comfort.

Keywords: Fuel Cell Vehicle
Abstract: To improve the volume power density of proton exchange membrane fuel cell (PEMFC), a design of graphite bipolar plate straight channel characterized by narrow ribs is studied in this article. A three-dimensional multiphase model of PEMFC is employed to analyze the effects of the geometric parameters on the mass transfer characteristics and power density. The results show that smaller channel width and channel depth can enhance water removal and gas transport, which could increase the fuel cell performance. However, the influence of gas diffusion layer (GDL) thickness on fuel cell performance is not monotonous which means that there is an optimal value. The overall dimensions are optimized with volume power density as objective function. The best performance is obtained when the channel width, channel depth and GDL thickness are 0.1, 0.2 and 0.05 mm, respectively. Compared with the conventional channel design, the volume power density of optimal channel is significantly increased by 211.32% at 0.6 V.

Keywords: Engine control, Fuel Cell Vehicle, Vehicle Dynamics and Control
Abstract: Proton exchange membrane fuel cell (PEMFC) is an environmentally friendly and efficient power generation device. It offers promising advantages over conventional power sources in backup power supplies, distributed generation and vehicle power. A rapid response to the actual power required by load is of great significance to improve the economy and efficiency of the system. However, due to various uncertainties such as frequent disturbances and inaccurate model, the net power control has certain challenges. Therefore, a data-driven nonlinear subspace identification method is developed to build the model of net power. A segmented and consecutive step response of net power for PEMFC system are identified and analyzed, the models are verified by high-fidelity simulation data. Data-driven active disturbance rejection control (ADRC) algorithm is developed to control the model. Internal and external disturbances are considered as a total term, which is estimated and compensated by real-time input-output data and ADRC, respectively. Compared with the conventional proportion integral and proportion integral derivative control, the proposed ADRC can improve the performance of set-point tracking, disturbance rejection and robustness under different operating conditions.

Keywords: Fuel Cell Vehicle
Abstract: Vehicle fuel cells tends to operate at high pressure and high power. Performance tests show that increasing temperature and decreasing relative humidity can promote the voltage of fuel cell at high pressure and high load. In order to identify the impedance characteristics and explain the performance loss of fuel cell at high current, this paper designs temperature and humidity sensitivity tests. However, the internal state of the fuel cell is usually unstable at high current, which makes the electrochemical impedance spectroscopy results fluctuate greatly, especially at low frequency. And it cannot be directly analyzed. In this paper, the impedance spectrum is analyzed by the distribution of relaxation times (DRT) method. The diffusion impedance in low frequency range is quantitatively described. By utilizing DRT method, the impedances of fuel cell in different frequency ranges are quantitatively identified, solving the problem that the low-frequency impedance cannot be identified at high current. This technique can be utilized to analyze different polarization losses, and provides a new method for fuel cell fault diagnosis.

Keywords: Fuel Cell Vehicle
Abstract: This paper numerically compares the existing capillary pressure-liquid water saturation correlations of gas diffusion layer with a three-dimensional fuel cell model. The cell performance and liquid water distributions are calculated with different pc-s correlations under the same conditions. The results indicate that the applicability of these correlations are not consistent. Polynomials with higher orders predict the polarization curve better. Exponential correlations tends to overestimate the capillary pressure and water saturation. Therefore, the uniformity of oxygen concentration and fuel cell performance are underestimated.

Keywords: Fuel Cell Vehicle, Vehicle Dynamics and Control
Abstract: Air supply system is one of the most important auxiliary subsystems of proton exchange membrane fuel cell (PEMFC). The response speed of voltage and current in fuel cell system greatly depends on the dynamic performance of air supply system. In this paper, based on AMESim software®, a fuel cell air supply system model of 72kW stack is built. The influence of air compressor response speed, buffer tank and flow resistance on the dynamic response characteristics of air supply system is analyzed, and the influence mechanism is briefly analyzed according to the simulation results.

Keywords: Electrification of vehicle, Vehicle Powertrain Control, Engine control
Abstract: Bidirectional single-stage isolated matrix converter(BSMIC) has a great application prospect in battery testing and electric vehicles charging systems with the advantages of wide power transfer range and high efficiency. In battery testing systems, it also needs a fast dynamic response. This paper proposes a hybrid modulation and control strategy to realize the fast dynamic response. A hybrid modulation strategy is proposed, which consists of a seven-segments space vectors pulse wide modulation(SS-SVPWM) for the matrix converter semi-stage and a phase-shift modulation for the H-bridge semi-stage. Then, a dual-loop control strategy is proposed. Both modulation ratio and phase-shift angle are used to control the output current. Finally, simulation results verify the effectiveness of the proposed modulation and control strategy.

Keywords: Electrification of vehicle
Abstract: Lithium-ion batteries may be slightly overcharged due to the errors in the Battery Management System (BMS) state estimation when used in the field of vehicle power batteries, which may lead to problems such as battery performance degradation and battery stability degradation. Therefore, this paper conducts an experimental study on the influence of slightly overcharging cycles on battery performance degradation, and builds a semi-empirical capacity degradation model under slightly overcharging cycles on this basis. The experimental results show that the slightly overcharging cycle causes the capacity decay of the battery to be significantly accelerated, and its capacity decay will also cause the capacity "diving" phenomenon at the end of its life under normal cycle conditions. The slightly overcharging cycle has little effect on the internal polarization resistance of the battery. However, due to the thickening of the SEI film, it has a greater impact on the ohmic internal resistance.

Keywords: Vehicle control, Automated and Connected Vehicle, Intelligence of vehicle
Abstract: When comparing the environmental protection and economy of different cars, it is necessary for cars to run in the same driving cycle to obtain the pollutant emission and fuel consumption. However, in the actual driving process, the performance of the vehicle may be markedly different from the performance in test cycle. In order to generate the driving cycle that can represent the actual driving process, this paper adopts the driving data of an express truck with specific driving routes to construct the typical driving cycle of a city by combining Markov chain with Monte Carlo random sampling. The random response is added in the construction process, and the variation parameter is used to simulate the sudden traffic situation. CCPV and CPV parameters are set to evaluate the generated driving cycle. Through Simulink simulation, the reliability of the generated driving cycle is verified and the influence of different statistical characteristics is determined.

Keywords: Intelligence of vehicle, Human-Computer Interaction, Advanced Driving Assistant System
Abstract: In human-machine cooperative decision making and control of intelligent vehicle, the intelligent system needs to understand driver’s intention and desired vehicle trajectory in order to assist driver with safety driving in complex traffic scenes. In this paper, a vehicle trajectory prediction encoder-decoder model based on Gated Recurrent Unit (GRU) with attention mechanism is proposed. The proposed model is comprised of intention recognition module and trajectory prediction module. The intention recognition module was employed for recognizing driver’s intention and calculating the probabilities of turning-left, lane-keeping, turning-right. The trajectory prediction module predicts vehicle trajectory using GRU decoder with attention mechanism, which takes vehicle historical position as input and predicts future position. Both intention recognition module and the trajectory prediction module share one encoder to save time. The NGSIM dataset was employed for training and testing. The experimental results indicate, comparing with traditional methods, the proposed horizontal-longitudinal decoupling hierarchical trajectory prediction method based on GRU neural network can predict driver’s desired vehicle trajectory in a long prediction horizon and the attention mechanism improve the trajectory prediction accuracy at the same times.

Keywords: Vehicle Dynamics and Control
Abstract: The performance of vehicle chassis control systems relies on the accuracy of input information to the control systems. Some important vehicle states which are necessary for chassis control cannot be directly measured at low cost, such as the vehicle longitudinal and lateral velocities. In the existing literature, many vehicle state estimation solutions are designed based on vehicle dynamic models. These models inevitably involve the acquisition of tire forces which cannot be easily measured or estimated. In this paper, a vehicle state estimator is proposed based on a straightforward vehicle kinematic model, which does not rely on any tire force information. The complexity and computation load of the proposed state estimator is low. Besides, to ensure competitive estimation performance, the state transition model used in this estimator is designed to be adaptive to the on-board sensor measurements. In the simulation studies, the proposed estimator is able to provide accurate estimation results under different simulation conditions, which verifies the effectiveness of the proposed vehicle state estimator.

Keywords: Advanced Driving Assistant System, Intelligence of vehicle
Abstract: Recently, self-supervised method has become an increasingly significant branch of depth estimation task, especially in the realm of autonomous driving applications. However, current per-pixel depth maps yielded from RGB images still suffer from uncertain scale factor generated by the nature of monocular image sequences, which further leads to the insufficiency in practical use. In this work, we first analyze such scale uncertainty both theoretically and practically. Then we perform scale recovery utilizing geometric constraint to estimate accurate scale factor, RANSAC(Random sample consensus) outlier removal is introduced into pipeline to obtain accurate ground point extraction. Adequate experiments on KITTI dataset(dataset generated by an autonomous driving platform built up by KIT and TRINA comprising stereo and optical flow image pairs as well as laser data, distributed to train set and test set on account of deep learning), show that, using only camera height prior, our proposed method, though not relying on additional sensors, is able to achieve accurate scale recovery and outperform existing scale recovery methods.

Keywords: Advanced Driving Assistant System, Human-Computer Interaction, Intelligence of vehicle
Abstract: The non-driving related operation behavior in driving process has a significant impact on road traffic status and driving safety, but there is less systematic study on the main characteristics and influence mechanism of such behaviors. Aiming at this problem, four types of typical behaviors of normal and abnormal driving are monitored and recorded by real vehicle test. The cognitive distracted driving behavior is taken as the research object, and the influence mechanism and prediction method of distracted driving are studied by using the driver's physiological state and vehicle running state. This paper focuses on the changes and statistical characteristics of driver's physiological state parameters and vehicle running state parameters during distracted driving, and then explores the influence mechanism of different types of distracted driving tasks with different loads on driver's state. This paper analyzes the influence mechanism from two aspects of human and vehicle. Based on the comparison of behavior criterion and load criterion, the parameter system of cognitive distracted driving behavior considering driving load is obtained after cross analysis. The prediction model is established as the training sample of LSTM model, and the model is tested with the data collected from real vehicle test After 100000 iterations, the training accuracy is 90.2% on the training set and 74% on the test set. The results showed that the cross-comparison method is scientific and reasonable, and the prediction model of distracted driving behavior based on physiological state and vehicle running state has good accuracy.

Keywords: Intelligence of vehicle, Advanced Driving Assistant System, Automated and Connected Vehicle
Abstract: This paper is a review about calibration method based on the hand-eye calibration principle. This method is designed for the coordinate system calibration problem of the vehicle LIDAR and Inertial Measurement Unit (IMU). Firstly, the point cloud data is pre-processed: the LIDAR estimation of self motion is realized by DB-SCAN incremental segmentation, covariance matrix’s eigenvalue calculation, description and matching, factor-graph based optimization and loop closure detection; secondly, the self motion of IMU device is estimated according to the output of combined inertial navigation equipment; finally, based on the hand-eye calibration principle and the error minimization method, the calibration between the two coordinate systems is calculated. Coordinate conversion matrices are shown to be friendly and accurate for initial value requirements.

Keywords: Intelligence of vehicle
Abstract: The Visual-Inertial navigation system(VINS) has become a popular navigation approach in the field of unmanned aerial vehicles(UAV) or robotics. While its performance under autonomous driving scenario is not satisfactory due to the fact that autonomous driving scenario is more challenging and dynamic than the UAV scenario. Thus, the Visual-Inertial navigation system will collapse occasionally and thus undermine the navigation result. In this work, we propose a adaptive mechanism that could switch between three modes, only VINs, only GNSS and VINS&GNSS fusion. When Visual-Inertial component breaks down, our algorithm could only rely on the GNSS signal until VINS recovers. Similarly, when GNSS signal is not very accurate, our system could only rely on the VINS-Mono. We demonstrate our algorithm under challenging scenarios such as night sight and high speed road and do both qualitative analysis and quantitative analysis.

Keywords: Vehicle control, Vehicle Dynamics and Control
Abstract: Model uncertainties and external disturbances can inevitably affect vehicle dynamic control accuracy and even cause the vehicle system to be unstable and unsafe. Therefore, vehicle dynamic controller must be able to suppress the influence of model uncertainties and external disturbances on vehicle dynamic control performance. To this aim, we design a matrix inequalities (both bilinear matrix inequalities (BMIs) and linear matrix inequalities (LMIs) are involved) based robust model predictive controller for vehicle dynamic control. Robust positive invariant (RPI) set is used to guarantee the controller is robust and to construct the matrix inequality equations. We test the usefulness of the proposed controller via a numerical example.

Keywords: Advanced Driving Assistant System, Decision Making
Abstract: Intelligent driving functions, such as ACC (Adaptive Cruise Control) and ALC (Automated Lane Changes) , require lane assignment for objects. It needs an accurate traffic lane path estimation. This paper proposes a fused front lane trajectory estimation algorithm based on current common ADAS sensor configuration. This trajectory is generated by fusing information of lane markers, front object trails and host motion state. This algorithm uses a clothoid lane model and its coefficients is estimated by a Kalman Filter, which weighs predicted model state and current measurement. This approach is verified by a set of real road test data.

Keywords: Advanced Driving Assistant System, Intelligence of vehicle, Automated and Connected Vehicle
Abstract: In this paper, the vehicle state parameters during driving are obtained through simulated driving experiments, the corresponding parameter change rules are analyzed, the characteristic parameters describing the intention are selected, and a sample library is established. The driver intention recognition model is built based on BP neural network, and the model is trained based on the data samples in the sample library to obtain the driver intention recognition model. The performance of the model was then analyzed, and the single working condition and compound working condition were verified in the model verification stage. From the experimental results, it can be seen that the intention model can accurately identify the driver's intention under a single operating condition. Under composite operating conditions, the vehicle's deviating behavior from the center line of the lane is similar to the lane changing behavior, so the model recognition results have certain errors, but the model can be accurately identify the driver's intention.

Keywords: Automated and Connected Vehicle, Vehicle control
Abstract: A key problem to be solved for automated driving is to generate a feasible trajectory regarding the constraints imposed by vehicle physical limits and road surface. This paper presents a model predictive trajectory planning framework for automated driving, which takes the dynamic driving constraints into consideration. The constraints are calculated by a separate dynamic driving constraints computation module and represented in the form of inequalities. This module decouples the planner from the vehicle physics such that the planner only needs to consider the kinematics of the vehicle. Therefore, the planner can generate feasible trajectories without directly knowing the properties of vehicle or actuators. The proposed model predictive trajectory planning framework is evaluated in simulation with three different scenarios and the results demonstrate the effectiveness of the proposed framework even in critical use cases of automated driving vehicles.

Keywords: Vehicle control, Vehicle Dynamics and Control
Abstract: With the development of electric power steering becoming more and more mature, many companies and universities have begun to develop steer-by-wire systems. In the online steering system, the steering wheel end is disconnected from the steering rack, and the road feel simulation unit is required to provide the driver with a real-time road feel to ensure the driver's maneuverability of the vehicle. There are many challenges in the generation of analog road feel. The basic functions include central position road feel, active return function, low-speed lightness and high-speed stability. The first part of this article discusses the control method of the road sense feedback motor, the second part is the design of the road sense feedback algorithm, and the third part is the test results and analysis. The motor control algorithm uses PID and feedforward algorithms to achieve precise current control. The road sense feedback part is designed to change the steering force gradient and damping with the speed of the vehicle. It is verified by Carsim and Matlab/Simulink joint simulation. The test part verifies the road sense algorithm. Adapt to different vehicle speeds and working conditions, and the smoothness of the center position and feel meets the needs of road feedback.

Keywords: Advanced Driving Assistant System, Vehicle Dynamics and Control, Vehicle control
Abstract: This study proposes an integrated vehicle emergency collision avoidance control based on nonlinear moving horizon estimation (MHE) combined with real-time vehicle state and road friction coefficient. The MHE method is employed to estimate the real-time road adhesion coefficient and vehicle lateral velocity, and the real-time road parameters and process state of the vehicle are applied to the collision avoidance system. Furthermore, the stability and safety of the vehicle is greatly improved. A nonlinear vehicle model is established based on the dynamic change of the road friction coefficient and highly nonlinear characteristics of the vehicle. Experiments were conducted under varying road adhesion conditions to verify the effectiveness of the proposed integrated collision avoidance control based on MHE. Results indicate that the integrated collision avoidance control can accurately estimate different road-adhesion coefficients and improve the vehicle stability in the collision avoidance process.

Keywords: Automated and Connected Vehicle, Vehicle Dynamics and Control, Vehicle control
Abstract: Aim at rejecting uncertainty disturbance and actuator saturation, a path tracking method is proposed for autonomous driving vehicles, which is implement by active disturbance rejection controller (ADRC) with conditional integration. Firstly, a kinematic-dynamic vehicle model is deduced for describing path tracking process. Secondly, a nonlinear extended state observer is designed to observe the uncertainty disturbance, such as external disturbance and parameter uncertainties. Finally, in order to eliminate error and reject disturbance while resisting actuator saturation, a conditional integration is developed as feedback control low. The test results of lane changing scenarios show that the proposed algorithm can track the desired path quickly and accurately compared with PID and ADRC.

Keywords: Control of UAV/USV/UUVs, Vehicle Dynamics and Control, Intelligence of vehicle
Abstract: This paper proposes a path tracking and planning approach based on backstepping, neural network, and resistance network for autonomous driving. The improved diamond resistance network method is presented to avoid the discontinuous curvature of the generated path in this paper. The path tracking controller is designed based on radial basis function neural network and backstepping method, taking the failure of the actuator into account. The stability of the proposed controller is proven in this paper. The simulation results show that the designed controller can track the planned path accurately. Meanwhile, it is compared with the classic backstepping controller, which proves the advantages of the designed controller to deal with actuator failures.

Keywords: Advanced Driving Assistant System, Decision Making, Vehicle control
Abstract: To improve the control performance of the lane departure prevention system, a segmented coordinated control method of active steering and differential braking is proposed. Taking the longitudinal velocity and road curvature as input and considering the driving habits of the driver in different lanes, the fuzzy controller used to determine the virtual lane boundary is designed. The control decision of LDPS is determined by combining the steering behavior index, which characterizes the steering intervention of the driver with the virtual lane boundary width. Next, the segmented control strategy is designed. In the first stage of the assistance control, the intervening time of coordinated control mode based on active steering and differential braking and the additional yaw moment when working in coordinated control mode is determined. A single control mode based on active steering is adopted in the second phase of assistance control. Finally，the effectiveness of the proposed coordinated control algorithm is evaluated with numerical simulation and experiments on a human-in-the-loop platform. The obtained results show that this method not only can avoid lane departures effectively, but also provides a smooth ride quality to the passenger.

Keywords: Control of UAV/USV/UUVs, Automated and Connected Vehicle
Abstract: The autonomous underwater vehicle (AUV) technique, which was widely used in various scenes, has become one of the research hotspots in recent years. This paper applies a distributed model predicted control (DMPC) approach for cooperative diving control of multi-AUV systems with bounded control input and state constraints. Within the DMPC framework, each AUV exchanges assumed state trajectories with neighbors and solves a local open-loop constrained optimization problem to obtain the optimal control input. The effectiveness of the method is validated in three simulation scenarios.

Keywords: Hybrid Electric Vehicle, Vehicle control, Vehicle Dynamics and Control
Abstract: Abstract—In this paper, a model of vehicle torque demand prediction based on Long Short Term Memory networks (LSTM) is presented. The correlation analysis of the data used to train the network is carried out, and appropriate characteristic parameters including acceleration pedal, brake pedal, speed and torque are selected as the input of the network. The characteristic parameters of the past 2 steps are used to predict the torque demand of the future. The data collected from vehicle controller is divided into two sections which are training set to train the LSTM and testing set to verify the performance. The results compared with that of artificial neural network show that prediction model can effectively improve the prediction accuracy.

Keywords: Automated and Connected Vehicle, Decision Making, Advanced Driving Assistant System
Abstract: Most of the existing researches only consider vehicles and signals as control objects, and there are also problems of loss of space and time resources caused by unreasonable distribution of spatiotemporal-right. In this paper, an overall collaborative control model for intersections considering the distribution of spatiotemporal right, vehicle trajectory and signal timing was established. A solution algorithm for the assignment of spatiotemporal-rights based on decision tree C4.5 is proposed. A high-dimensional solution based on genetic algorithm and an enumerated low-dimensional solution for signal timing and vehicle trajectory optimization are proposed respectively. Finally, an overall control model including the phase and lane, signal timing and vehicle trajectory was established. The simulation program was developed with python3.7, and the effectiveness of algorithm proposed in this paper was verified by experiments. When flow intensity is 0.23, the algorithm has the best improvement effect, the high-dimensional and low-dimensional algorithms can reduce the delay by 57.6% and 44.8% respectively. It also verified that the algorithm has better adaptability to the change of traffic demand than the algorithm that only considers the vehicle trajectory or signal timing.

Keywords: Automated and Connected Vehicle, Intelligence of vehicle, Vehicle control
Abstract: Abstract—The self-driving forklift, as a promising technology to reduce the labor intensity of workers, can also improve the efficiency of logistics freight transportation. In this paper, a path-following controller that combines cascaded active disturbance rejection controller and geometry-based feedforward controller, is proposed. The cascaded controller, designed based on a kinematic model, minimizes the lateral error via the outer-loop by mitigating the desired heading direction, and then achieved by the inner loop through adjusting the steering angle. The deviation between the simplified kinematic model and the actual forklift motion is lumped as a total disturbance, to be observed by the extended state observer (ESO). In order to enhance the transient response, a geometry-based feedforward controller is developed, computing the desired steering angle through preview. The proposed method effectively improves the response speed and reduces the overshoot. The effectiveness of the algorithm is quantitatively evaluated in experiments.

Keywords: Automated and Connected Vehicle, Hybrid Electric Vehicle, Intelligence of vehicle
Abstract: For electric vehicle or hybrid electric vehicles, the regenerative braking is one of the important means to realize energy saving, for which braking ahead of a traffic light intersection is a representative scenario. The uncertainty in driver behavior and future traffic flow, however, make it challenging to achieve optimal dynamic energy recovery through conventional braking operation by drivers. Therefore, in this paper, an energy recovery optimization-oriented vehicle speed planning algorithm ahead of traffic lights intersection is proposed, for autonomous vehicle or driving assistance system. First, the reward function is designed, taking the energy recovery amount, traffic efficiency and driving smoothness into consideration. Then, the information of traffic lights at intersections is obtained in advance through V2I (vehicle to infrastructure) communication. Finally, the q-table and neural network are trained in the framework of reinforcement learning, deriving optimal vehicle speed profile. Simulation results on a high-fidelity model show that the amount of recovered electrical energy using q-learning algorithm is 45.08% higher than that of uniform deceleration. The amount of electrical energy using DQN (Deep Q-network) algorithm is 2.24% higher than q-learning, showing to be a better candidate in terms of comprehensive optimality than q-learning.

Keywords: Automated and Connected Vehicle, Intelligence of vehicle
Abstract: Autonomous driving is one of the three major innovations in automotive industry. Autonomous vehicles promise to revolutionize human mobility and vehicle safety. This promise can’t be realized without the ability to constantly make the right decisions even in the complex situations. This paper proposed a new decision-making system including a new way using the long short-term memory neural network to predict the states of the vehicles nearby in the short future using the history of which got from the cognitive ability. Based on the future states predicted by the neural network, this paper also proposed some statistics methods to give a classification criterion to judge a vehicle is dangerous, attentive or safe.

Keywords: Decision Making, Vehicle control, Automated and Connected Vehicle
Abstract: More precise decisions enable the autopilot driving system to ensure safety performance as well as respect human willingness. A learning-based parameter decision-making method for automated vehicles is proposed to achieve it in the lane-change scenario. The rationality of such parameter-based precise decisions is shown in three aspects. First, microcosmic behaviors in decision parameters,such as lane-change time and expected acceleration, will influence planning, which provides a flexible space for decision-making. Secondly, based on the analysis of the real traffic data, NGSIM, changeable lane-change time and expected acceleration when drivers change lane are revealed， which is seldom considered in the decision layer of current researches. Also, we find some potential emergencies in NGSIM are promoted to be safer by optimizing part of decision parameters. The reinforcement learning method is designed, which adds lane-change time and expected acceleration to the action space and considers safety, current driver's willing and average human driving style as reward signal. The problem is solved by kernel-based least-squares policy iteration reinforcement learning (KLSPI) with a customized basis function. Simulation results demonstrate that using RL to learn parameter decisions can realize more precise decisions that promote safety performance and imitate human driver's behavior in the lane-change scenario.

Keywords: Automated and Connected Vehicle, Decision Making, Vehicle Dynamics and Control
Abstract: Accurate intention prediction can help intelligent vehicles make safe decisions and have a better understanding of the environment, thus improving the safety of automatic driving and promoting cooperative driving. In this paper, an interactive intention prediction method is proposed. Firstly, the Hidden Markov Model integrated with Gaussian Mixture Model is modeled for current behavior recognition and its parameters are trained through NGSIM dataset. Then, a trajectory prediction method based on Frenet frame is used to predict the future traffic situation, considering which future behavior reasoning is realized by maximum expected utility theory. The final intention prediction result is a combination of historical trajectory recognition and future behavior reasoning. The simulation results show that the proposed method has the ability of reasonably reflecting the interaction process between vehicles and the prediction performance is good.

Keywords: Advanced Driving Assistant System, Automated and Connected Vehicle
Abstract: With the deep research on autonomous driving, the target detection algorithms based on 2D images have become a hot topic in recent years. In this paper, we mainly study the six mainstream deep learning detection algorithms, namely YOLO, YOLOv2, YOLOv3, YOLOv4, SSD and RetinaNet, which are used as the representative algorithms of one-stage detection methods. This paper is a review of the working principles of these six detection algorithms in deep learning. Firstly, this paper explains and compares the detection network structure, loss function and improvements of these six detection algorithms in detail according to the order in which they were presented. Especially the YOLOv3, YOLOv4, SSD and RetinaNet algorithms, which are the most common used algorithms at presented with a very high detection speed and detection precision for real-time detection of the autonomous driving. Secondly, by comparing the detection accuracy and speed of the YOLOv3, YOLOv4, SSD and RetinaNet algorithms, this paper gives their respective suitable application scenarios.

Keywords: Electrification of vehicle, Intelligence of vehicle, Decision Making
Abstract: Single-phase neutral point clamped rectifier (SP-NPCR) is getting much interest in 10 kV electric vehicle (EV) charging DC systems due to its low voltage stress on power switches. In order to achieve good tracking performance and anti-disturbance ability, a robust command-filtered backstepping control strategy of the SP-NPCR is proposed in this paper. Firstly, based on the SP-NPCR model in dq frame, the command-filtered backstepping controller is constructed, which compose of the output voltage controller and the reactive current controller. The command filter is used to meet the requirement of analytic differentiation of the virtual controller and simplify the controller structure. Then, the global stabilization of the SP-NPCR is conducted and proved by the Lyapunov stability theory. Simulation results show that the proposed method has better tracking performance and stronger robustness compared with the conventional proportional-integral method.

Keywords: Vehicle Powertrain Control, Vehicle control
Abstract: Multiple speed transmissions are applied to electric vehicles gradually. A reverse gear mechanism using dog clutch is proposed for the inverse Automated Manual Transmission (I-AMT), and the coordination controller of the driving motor and the dog clutch is designed. Considering the characteristic of time delay in the motor control system, a control strategy based on Smith predictor is derived to increase the tracking ability and further improve the dynamic performance of the closed-loop control system. The experiment shows that compared with PID control strategy, those with Smith predictor is better in shifting comfort and reducing the machine wearing of dog clutch.

Keywords: Advanced Driving Assistant System, Intelligence of vehicle
Abstract: This paper proposes a road risk assessment(RRA) algorithm based on lane and obstacle information, including obstacle selection based on lane line equations and risk assessment based on time to collision (TTC). The algorithm can integrate the functions of collision warning and lane change risk assessment, and can be implemented on actual vehicles. RRA can provide collision warning for vehicles in the lane and also give lane-changing suggestions, and can conduct risk assessment of the lane to be changed according to the needs of lane changing. In addition, the alarm logic is designed, that is, when multiple risks appear at the same time, it is determined which HMI signal to prompt the driver. This method has been tested on the AEOLUS D53 vehicle, and the test results show that the system can effectively evaluate the driving risk.

Keywords: Automated and Connected Vehicle, Decision Making, Intelligence of vehicle
Abstract: Motion planning in dynamic environment requires proactive response to other road users' intentions. While much progress has been made in intention-awareness methods, challenges remain for the motion planner to consider the uncertainty of intention estimation as well as other risk factors (e.g. the type and motion state of the obstacle). In this paper, we propose a novel motion planner based on safety-field and model predictive control (MPC). Driving-risk assessment is firstly conducted through safety-field based on probabilistic intention-inference with dynamic Bayesian network (DBN), the optimal controller then generates a feasible trajectory within the prediction horizon. Through simulations based on a cut-in situation and an intersection-passing scenario, the effectiveness of this method is demonstrated in enhancing driving safety with considerate motion planning strategy.

Keywords: Intelligence of vehicle
Abstract: For autonomous vehicle, ground segmentation is one of the most basic and essential capabilities. In this paper, to further improve the performance of LiDAR-based ground segmentation approaches, other than the commonly used `occupancy' information of the point clouds, we try to utilize the `free' information naturally contained in any point clouds. We choose to use the stixel as the processing primitives. By utilizing the intrinsic relationship between the 3D voxel grid and the 2D range image, we can efficiently calculate the stixel features that contain both the occupancy information and the free information. Using the stixel features as inputs, we design a deep learning network for ground segmentation. The designed network consists of a feature extraction net and an ERF (Efficient Residual Factorized) semantic segmentation net. Experiments on SemanticKITTI dataset show that our proposed approach outperforms state-of-the-art approaches.

Keywords: Vehicle Dynamics and Control, Vehicle control, Advanced Driving Assistant System
Abstract: Under extreme driving conditions, the tire lateral forces saturate easily, which should be considered for better performance in vehicle stability control, as well as the safety constraints and real-time response. To address the above problem, a real-time model predictive controller for four wheel independent motor-drive electric vehicles is proposed to improve the lateral stability. First, considering the saturation characteristic of the tire dynamics on slippery road, the tire model is developed into the piecewise form of linear and saturation regions, which extracts the main nonlinearity of tire. Second, the additional yaw moment is determined to achieve the control objectives of lateral stability and handling performance. Then, the additional yaw moment is distributed into torques acting on each motor by optimizing the tire load rates. Finally, co-simulations with MATLAB/CarSim and hardware-in-the-loop simulation are performed, and the fast solution of optimization problem is realized based on C-language. The results show that lateral stability and handling performance are efficiently improved, and the real-time performance can be ensured with a sampling time as 5ms.

Keywords: Vehicle control, Intelligence of vehicle, Human-Computer Interaction
Abstract: Image-based autonomous driving control is one of the important research directions in the field of autonomous driving. Most of the existing image-based control algorithms use end-to-end mapping from image to vehicle control amount, which is not explanatory enough, and the control amount is not intuitive enough to effectively implement human-machine collaborative control and incremental learning of models. This paper proposes an incremental learning algorithm for driving vehicle control based on fast pursuit point estimation. We establish a model to calculate the mapping of image to the pursuit point, and then get the actual control amount of the vehicle throttle value and front-wheel rotation angle value by the pursuit point. Combining the features of pursuit point which can be observed intuitively and has obvious physical meaning, we propose an incremental model updating method based on man-machine collaborative control, which can incrementally improve the model performance in the actual driving process of vehicles. Finally, the experiment of automatic control is carried out on the Carla simulation platform. The experimental results show that the algorithm can incrementally improve the performance of the automatic control model, with the average calculation speed over 50fps. The autonomous driving system realizes automatic cruise in the real campus environment.

Keywords: Electrification of vehicle, Integrated Chassis Control, Vehicle Dynamics and Control
Abstract: Brake by wire is a key technology in the motion control level of autonomous vehicles. In order to provide a flexible control interface for the decision-making level, the hardware of the brake-by-wire controller is designed, and the active hydraulic boosting law is implemented using model-based-design method. The application verification is carried out on the experimental bench. The results show it is feasible to use the model-based-design method for the brake-by-wire controller development.

Keywords: Vehicle Dynamics and Control, Vehicle control
Abstract: This paper proposes a novel control strategy adjusting damper force of semi-active suspension to improve vehicle performance, including comfort performance considering both vertical vibration and roll motion during a gentle turn on uneven road, yaw tracking capability during a shaper turn, and rollover avoidance during a fierce turn. The coupled roll and yaw dynamics model and quarter suspension model are firstly established. Considering road unevenness is the main factor which causes vertical vibration and discomfort, a simple method to evaluate road unevenness with vertical acceleration of sprung mass is proposed. The coupled roll and yaw dynamics model is simplified to a prediction model with lower computational cost, and then an MPC controller is designed. Three different cost functions of comfort, yaw tracking and rollover avoidance respectively are designed, and their switching strategy is proposed according to priorities. Simulation results show that control strategy proposed in this paper is effective to reduce discomfort, overshoot of yaw rate and risk of rollover.

Keywords: Advanced Driving Assistant System, Automated and Connected Vehicle, Intelligence of vehicle
Abstract: Crossroad recognition is an important part of unmanned navigation, which provides crossroad location information. Most existing works treat scene recognition as a classification problem and achieve good results. However, it is difficult for them to cope with scene changes, such as occlusions and seasonal changes. In this paper, we propose a human-inspired crossroad recognition method that detects some key local regions of the scene to recognize crossroads. The key local regions contain the unique features of the intersection, which are extracted by visualizing the features of the pre-trained crossroad classification network. Based on the detected key local regions, an inference model calculates the score of crossroad classification and output the recognition result. The proposed method can effectively recognize the crossroad under occlusion and other changing scenes. We collected the crossroad recognization dataset on our campus, and evaluate the performance of our method on this dataset. Extensive evaluation shows that our method achieves higher accuracy and satisfies the real-time requirement.

Keywords: Intelligence of vehicle, Vehicle Powertrain Control, Automated and Connected Vehicle
Abstract: with the development of intelligent and connected vehicles, a novel approach for energy-saving, i.e. eco-driving control, has attracted much attention from relative researchers. The combination of eco-driving control and plug-in hybrid electric vehicles provide an opportunity to achieve further energy-saving for transportation. In this paper, an optimal eco-driving control strategy is proposed for plug-in hybrid electric vehicles based on the neural network. In order to mitigate the huge computational cost of velocity optimization and powertrain control, an efficient hierarchical optimal control strategy is proposed. An artificial neural network is constructed for the modeling of optimal energy cost. This optimal energy cost model is applied as objective function in the solving of the optimal eco-driving problem. The simulation results show that the proposed method can improve fuel economy by 4.29-12.71%, compared with conventional eco-driving control strategy. The neural network based optimal energy cost model significantly heightens the computational efficiency, with small sacrifice for fuel economy compared to optimal bench mark.

Keywords: Vehicle Powertrain Control, Vehicle control
Abstract: In this study, a nonlinear generalized predictive control (NGPC) strategy in a cascaded structure, combining with sliding mode disturbance observer (SDMO), is proposed to control the permanent magnet synchronous hub motor (PMSHM) with uncertainties and disturbances. The NGPC is designed on the basis of the Taylor series expansion to approximate the predictive response in finite time domain. Since NGPC cannot thoroughly remove the deviation in the load torque variation and parametric uncertainties, an improved SMDO is exploited to estimate and compensate the deviation of controller. The developed controller can fulfill the performance of strong robustness and fast dynamic response with easy regulation characteristics. The simulation results manifest the feasibility and effectiveness of the designed control strategy applied to the PMSHM drive.

Keywords: Intelligence of vehicle, Vehicle Powertrain Control, Automated and Connected Vehicle
Abstract: Energy management strategy is a crucial factor impacting the fuel efficiency in hybrid electric vehicles. In this paper, an energy management strategy settled by reinforcement learning is proposed for a plug-in hybrid electric vehicle (PHEV). First, the architecture of the target PHEV is modeled and analyzed in detail. As one of the effective state-action value estimation method in temporal-difference learning, Q-learning is then employed to solve the energy management problem. The main mechanism of Q-learning and specified training process considering battery state of charge planning is articulated, and the corresponding simulation validations are implemented sequentially. Comparing to global optimal control strategies, the proposed algorithm performs near-optimal power allocation with the state of charge planning and constraint.

Keywords: Vehicle control, Fuel Cell Vehicle, Hybrid Electric Vehicle
Abstract: The fuel cell hybrid electric vehicle (FCHEV) is a new type of vehicle with the advantages of high efficiency and environmental protection. As the government and society pay more and more attention to environmental and energy issues, the development of FCHEV has entered an important stage. The control algorithm of FCHEV is a key technology of new energy vehicles and requires research. This research mainly focuses on the power system modeling and the longitudinal dynamics control and simulation of FCHEV. Based on the tire model, a new slip ratio estimation strategy was proposed. The target drive torque control algorithm and the anti-slip control algorithm adopt the feedforward control and Proportional-integral feedback control. The hydraulic braking force and the regenerative braking force were distributed to ensure that the motor exerts the maximum regenerative braking capability, while the braking force distribution meets the requirements of the ECE braking regulations. On the MATLAB/Simulink software platform, a FCHEV power system model and a coordinated driving and braking control model were established. Through simulations in different working conditions, this paper proved the performance of the new slip ratio estimation algorithm and the feasibility of the dynamics control algorithm.

Keywords: Vehicle Powertrain Control, Hybrid Electric Vehicle
Abstract: For plug-in hybrid electric vehicle, an optimal energy management strategy can maximize its potential to achieve high efficiency. However, energy management strategy without condition information cannot achieve optimal fuel economy in real-time. In order to obtain higher efficiency and adapt to unexpected situation, we develop an energy management strategy based on velocity prediction using digital map information. The detailed model of the hybrid powertrain system such as engine, battery pack and vehicle model are established. The typical driving cycles are constructed to minimize the fuel consumption with equivalent consumption minimization strategy. To adapt to sudden congestions, a realtime strategy based on velocity prediction is proposed. Results indicates that equivalent consumption minimization strategy with velocity prediction is more efficient than the traditional equivalent consumption minimization strategy.

Keywords: Hybrid Electric Vehicle, Vehicle control, Vehicle Powertrain Control
Abstract: Based on the key components and controller models of the hybrid electric vehicle, the energy flow of each component of the vehicle under different power of the A/C (Air-Conditioning) system is analyzed and calculated. The control method adjusts the torque distribution strategy between the engine and motor. In this paper, the energy flow of each component is analyzed under the condition of different power of the air conditioning accessories in the standard cycle condition. Compared with the traditional control method, the improved vehicle control method can be used in different accessory powers to reduce the overall energy consumption of the vehicle. When the power of air conditioning accessories are 3kW and 5kW, the energy consumption of the new method can improve by 3.8% and 8.6%.The results show that the proposed control method can significantly reduce the comprehensive energy consumption of the hybrid electric vehicle under the different air conditioning powers.

Keywords: Vehicle Powertrain Control, Vehicle Dynamics and Control, Vehicle control
Abstract: A vehicle model equipped with wet dual clutch transmission based on Matlab/Simulink is established and the filling influence on the vehicle’s gear shifting quality is simulated by the model. The simulation results show the friction work increases when the wet clutch is under fill and the jerk increases when the clutch is over-fill. So a two-parameter fuzzy control method is proposed to adjust the filling oil pressure and the built model is used to verify the effectiveness of the algorithm. Furthermore, rapid prototype simulation verification is carried out based on dSPACE hardware. The simulation results show that the adaptive control can reduce the jerk and friction work of the vehicle shifting process, and improve the shifting quality and driving comfort of the vehicle.

Keywords: Vehicle Powertrain Control, Hybrid Electric Vehicle, Automated and Connected Vehicle
Abstract: This paper proposes a real-time energy optimization algorithm for hybrid electric vehicles that operates in a platoon using constant headway control strategy. The driving power of a hybrid electric vehicle can be distributed between the electric motor and the combustion engine. The power distribution makes evaluable energy efficient improvement. Combining the control strategy proposed by a classic cooperative adaptive cruise control (CACC) algorithm, an optimal energy management strategy is evaluated for the host vehicle in a platoon. Through the establishment of simulation experiments, the string stability and real-time energy optimization performance of the control system are verified.

Keywords: Hybrid Electric Vehicle, Automated and Connected Vehicle, Vehicle Powertrain Control
Abstract: An adaptive energy management control strategy is proposed for a commuter plug-in hybrid electrical vehicle (PHEV) during car-following process in this paper. The proposed energy management strategy (EMS) is an instantaneous optimization control strategy integrating car-following behavior performance index into adaptive equivalent consumption minimization strategy (A-ECMS). In order to achieve better fuel economy and safety performance under different carfollowing scenarios, the equivalent factor (EF) of ECMS and the weight factor of car-following performance in the instantaneous optimization cost function are designed as adaptive forms of Map tables about battery state of charge (SOC) and travel distance. The Mapping tables are established offline by utilizing historical traffic data of the commute road and particle swarm optimization (PSO) method. The effectiveness and practicality of the designed EMS are verified through the co-simulation of MATLAB/Simulink and GT-Suite simulator.

Keywords: Electrification of vehicle, Vehicle control
Abstract: The angular velocity and rotor position of the motor are computed by PI observer. Clark and Park transformations is utilized to establish a dynamics model of the motor. Space vector pulse width modulation algorithm is used for voltage modulation. The main control chip of the hardware board selects TI’s TMS320F28035, and the power device selects Mitsubishi’s IPM(PS22A79). The hardware board is designed to test and verify the PI observer. The test results are analyzed and indicated that the PI observer is a good candidate for sensorless field oriented control of permanent magnet synchronous motor.

Keywords: Engine control
Abstract: High-pressure injectors are the key actuators of a GDI engine. However, manufacturing deviation leads to the inconsistency of flow characteristics in the ballistic zone, which affects the accuracy of fuel control. Based on the feedback signal of driving voltage of high-pressure injector, the algorithm of recognizing the opening and closing action characteristics of needle valve is studied, and the self-learning method of flow characteristics and the compensation method of injection driving pulse width are proposed. The test results show that the method can effectively improve the consistency of flow characteristics in the ballistic zone for different fuel injectors, reduce the deviation from 25% to 10%, and effectively improve the fuel injection accuracy, so that fuel rail pressure can be increased, injection splitting can be adopted or injection splitting times can be increased under more engine conditions, so as to improve emissions.

Keywords: Hybrid Electric Vehicle
Abstract: For plug-in hybrid electric vehicle, an objective evaluation index for ride comfort during mode switching is proposed and theoretically analyzed. A test is designed to obtain the subjective and objective evaluation results of mode switching. Based on the test data, analysis and research are conducted, and Pearson correlation coefficient is used to verify objective ride comfort consistency of indicators and subjective comfort evaluation scores, and the establishment of a model to switch between subjective and objective evaluation models through neural network to improve the efficiency and reliability of evaluation during vehicle development.

Keywords: Vehicle Dynamics and Control, Automated and Connected Vehicle
Abstract: In this paper, a fast nonsingular terminal sliding mode (FNTSM) control scheme is developed for autonomous vehicles' collision-avoidance steering control. First, a vehicle kinematic-and-dynamic model is established to describe the vehicle's fundamental lateral dynamics in path-tracking behavior. Afterwards, an FNTSM controller is designed, where the control system's stability is proved by means of Lyapunov. Finally, MATLAB-Carsim co-simulations are carried out for the aim of testing the control performance. Simulation results illustrate that the proposed FNTSM control algorithm possesses remarkable superiority reflected in higher tracking precision, faster convergence rate, and stronger robustness in comparison with a nonsingular terminal sliding mode (NTSM) controller and a conventional sliding mode (CSM) controller.

Keywords: Vehicle Dynamics and Control, Automated and Connected Vehicle
Abstract: In this paper, a fuzzy nonsingular terminal sliding mode (FNTSM) control strategy is proposed for the trajectory-following control problem of a Mecanum-wheeled automated guided vehicle (MWAGV). Initially, a plant model with 4 inputs and 3 outputs is identified to describe the kinematics and dynamics of the MWAGV's trajectory-tracking behavior. Then, an FNTSM controller is designed for the MWAGV, and the control system's stability is verified via Lyapunov. Lastly, simulations are executed to test the control performance in the cases of lateral motion and circular motion with an initial offset. The simulation results indicate that compared with conventional sliding mode (CSM) control, the developed FNTSM control algorithm owns remarkable superiority reflected in higher tracking accuracy, stronger robustness and a better balance between the tracking precision and control smoothness.

Keywords: Decision Making, Human-Computer Interaction
Abstract: As more and more electrical equipment is used in coal mines, ensuring the normal operation of mine explosion-proof mobile substations has become a top priority for coal mine safety managemen. We must ensure the mine explosion-proof mobile substations run without hitches for extended periods of time, and all the electrical equipments work safely and reliably, which finally guarantee that various tasks of coal mine can be completed well. This paper introduces the structure and basic principles of mine explosion-proof mobile substations, and reviews the common faults and solutions for mine explosion-proof mobile substations.

Keywords: Vehicle Dynamics and Control
Abstract: Aiming at the complex coupling between the various subsystems in the current automobile suspension, an active suspension control method based on the multi-agent prediction algorithm is proposed; Taking the nonlinear two-degree-of-freedom active suspension model as the research object, a suspension multi-agent model is established, with the goal of improving vehicle ride comfort, a distributed predictive control optimization algorithm based on Nash optimization is proposed;Simulation results show, Two different road conditions are selected: docking road and separated road. The active suspension control method based on multi-agent prediction algorithm maintains good suspension control performance and improves the ride comfort of vehicle suspension to a certain extent.

Keywords: Vehicle Dynamics and Control, Advanced Driving Assistant System, Electrification of vehicle
Abstract: Distributed drive electric vehicles have drawn more attention owing to their flexible control. This paper proposed a deep reinforcement learning (DRL) control for the torque distribution among four in-wheel motors. In detail, torque distribution is incorporated into Markov Decision Process and the actor-critic networks are employed to approximate the action-value and policy functions. The policy function networks guarantee the continuous solution of desired external yaw moment. The proposed torque distribution strategy is implemented with Carsim/Simulink platform. Numerical results illustrate the effectiveness of proposed torque distribution. Compared with the typical MPC and LQR strategies, the DRL based torque distribution strategy fully considers the understeering characteristics of DDEV and depicts a better handling stability performance.

Keywords: Vehicle control, Vehicle Dynamics and Control
Abstract: Magnetorheological (MR) controllable damping is promising in suspension control and almost commercialized in luxuries. However, the development of MR semi-active control for vehicles is complicated because of the messed inter-disciplinary process both in the suspension control and the MR damper control. In this paper, a new scheme of driving control based on BP neural network is proposed to package the MR damper as a black box implementing the strong nonlinearity mapping between the excitation current and damping force by the embedded driver. The sensor also embedded in the MR damper for integrated solution, and a mechanism for tackling the sedimentation problem of the MR damper are also pointed out.

Keywords: Electrification of vehicle, Vehicle control
Abstract: In order to calculate the iron loss resistance of the permanent magnet synchronous motor (PMSM) in real time, this paper presents an identification method of iron loss resistance based on model reference adaptive control (MRAC), and designs the adaptive mechansim by Popov’s hyperstability theory. According to the equivalent phase d-q circuit model of PMSM considering the iron loss resistance, the simplified PMSM model in the original Simulink motor library is modified. The MRAC-based identification method is simulated on the modified PMSM system under steady and dynamic conditions. The simulation results show that the estimated value of iron loss resistance can effectively converge to the real value. By adjusting the PI parameters and adding a low-pass filter, the robustness and dynamic characteristics of the identification system are improved.

Keywords: Engine control, Vehicle control
Abstract: The precise and efficient control of permanent magnet synchronous motor (PMSM) relies on real-time and accurate position and speed information. The traditional motor position detection needs to be realized by position sensor. However, for PMSM used in rail transit, the sensor scheme is complex and high cost. Therefore the paper takes position and speed sensorless detection of PMSM as the research goal, and systematically summarizes the principles, applicable environment and improved methods from zero-low speed and medium-high speed, which provides reference for sensorless control of PMSM for rail transit.

Keywords: Vehicle Dynamics and Control, Advanced Driving Assistant System, Vehicle control
Abstract: A new electro-hydraulic brake system with the structural characteristics of dual master cylinders is presented in this paper and considering the disadvantage of functional backup of the conventional vehicle stability control system, three vehicle yaw stability control strategy are presented. Firstly, the three-closed-loop pressure following PI control algorithm of the new electro-hydraulic brake system is studied to make it quickly follow the target pressure value; Secondly, based on analyzing the two-degree-of-freedom(2-DOF) vehicle dynamic model, the upper-layer, lower-layer controller were designed respectively using the hierarchical control strategy. The upper-layer controller adopted PID, Fuzzy and PID + Fuzzy three controls for the front wheel, rear wheel and front wheel + rear wheel of the vehicle respectively to calculate the additional yaw moment; Then the additional yaw moment is distributed to the single action wheel by lower-layer controller, and then motor control command is calculated by the target braking torque value to ensure that the additional yaw moment generated by the brake actuator tracks the desired yaw moment value of the upper-layer controller in real time. Finally, in order to verify the feasibility of the control strategy and the effectiveness of the algorithm, a co-simulation experiment of CarSim and MATLAB/Simulink is established. The results show that the control algorithm can match the characteristics of the electro-hydraulic brake system and has a good failure backup function and yaw stability control efficiency.

Keywords: Vehicle Dynamics and Control, Vehicle Powertrain Control, Vehicle control
Abstract: In this paper, an adaptive fuzzy control scheme is proposed for one-quarter automotive active suspension system with full sate constraints and stochastic disturbance. In the considered active suspension system, to further improve the driving security and comfort, the problems of stochastic perturbation and full state constraints are considered simultaneously. In the framework of backstepping, the barrier Lyapunov function is proposed to constrain full state variables. Consequently, by combing the differential formula and stochastic control theory, an adaptive controller is designed to adopt the uneven pavement surface. Ultimately, on the basis of Lyapunov stability theory, it proves that the designed controller not only can constrain the bodywork, the displacement of tires, the current of the electromagnetic actuator, the speeds of the car body and the tires within boundaries, but also can eliminate the stochastic disturbance.

Keywords: Vehicle Dynamics and Control
Abstract: Autonomous emergency braking (AEB) system enables vehicles to avoid collisions or mitigate the damage. Due to large braking distance, a safety distance model on the basis of the vehicles driving state and the process of vehicle braking is established. The driver’s reaction time and early warning vehicle speed following controller are established, and the predictive AEB algorithm based on the driver’s operation intention is established. Finally, TruckSim-Matlab/Simulink co-simulation for the AEB control strategy is conducted. The simulation result shows that the proposed control strategy can hierarchically warn and automatically brake, which is effective.

Keywords: Vehicle Dynamics and Control, Vehicle control, Integrated Chassis Control
Abstract: Unmanned tracked vehicle mostly use remote control. Communication delay and unclear images are easy to cause operator errors in operation, especially when starting on large uphill, safety accidents such as slipping and sideslip often occur. Aiming at the problem of uphill assist safety, on the basis of analyzing the longitudinal dynamics of the vehicle, a feedforward and feedback control method is proposed, the evaluation index of the big uphill assist performance is designed, the target driving force is obtained according to the uphill resistance and braking force, and the feedforward is designed The compensator calculates the feedforward driving force through the braking force, and then completes the feedback closed-loop control through the quasi-sliding mode controller. Through model simulation, this design method can help the unmanned tracked vehicle to start safely on the uphill, and the vehicle speed tracking effect is good, reducing the driver's control difficulty in uphill starting, and meeting the design requirements of the unmanned tracked vehicle's safe starting control on the uphill.

Keywords: Vehicle control, Vehicle Dynamics and Control, Vehicle Powertrain Control
Abstract: Abstract— In order to improve the safety performance of heavy commercial vehicles when the brake fails under cornering conditions, a strategy for differential compensation of initial braking force when single-wheel braking fails is proposed and a fuzzy PID controller based on yaw rate control is designed. The method of hardware-in-the-loop is used to verify the control effect of the brake. The results show that: the control algorithm is still effective and feasible to control the remaining braking performance of the vehicle under each braking condition.

Keywords: Vehicle Dynamics and Control, Vehicle control, Advanced Driving Assistant System
Abstract: In this paper, a reduced-horizon model predictive control strategy is proposed for autonomous parking. Given the planned trajectory, the linear time varying model is obtained by discretizing and linearizing along the reference trajectory. Then the linear time varying model is represented in an incremental form. A standard quadratic programming problem is formulated which can be solved online. The joint simulation in Simulink and CarSim in three parking scenarios shows the effectiveness of the proposed method.

Keywords: Vehicle Dynamics and Control, Advanced Driving Assistant System
Abstract: This paper presents an adaptive approach to simultaneously estimate the angles of vehicle roll and road bank with off-the-shelf vehicle inertial sensors. Measured signals are firstly processed through a kinematic model based adaptive complementary filter, and then fused in a dynamic model based Kalman filter. Adaptive law is designed to suppress the undesired effect caused by transient motion and integral drift. Suspension displacement sensors were installed to accurately measure the reference value of vehicle-body roll angle, and on-vehicle experiments on uneven ground were conducted to evaluate the performance of the proposed method. The effectiveness of the estimator was approved by comparing the estimating results and the reference.

Keywords: Vehicle Dynamics and Control, Vehicle control, Decision Making
Abstract: This paper mainly studies a harzard escaping strategy for high speed vehicle during tire blowout from the viewpoint of interference. Firstly, a harzard escaping strategy is proposed, mainly concerning with three stages, such as lane keeping, lane changing and emergency braking. Secondly, a vehicle stability controller is designed based on the model predictive control (MPC), which can deal with multiple constraint problem. Thirdly, external interference is employed to simulate the tire blowout of the vehicle at first time. Finally, the effectiveness of the escaping strategy and controller proposed is verified by the Simulink/CarSim cosimulation platform.

Keywords: Vehicle Dynamics and Control, Advanced Driving Assistant System, Vehicle control
Abstract: Platoon can travel with a small distance between vehicles, effectively reducing air resistance, reducing fuel consumption, and significantly improving traffic conditions. In order to maintain a safe and stable intra-vehicle distance, the paper proposes a distributed model predictive control method considering the state of the leader vehicle. The leader vehicle adapts an adaptive cruise control, and the following vehicle uses a controller with the same parameters to maintain a constant time gap. Vehicles need to join or exit the platoon in order to perform specific tasks in driving, so the paper proposes a scheduling strategy for the three scenarios of rear exiting, rear joining, and middle exiting. This paper uses Simulink and TruckSim to verify the effectiveness of the algorithm under platoon merge and split conditions.

Keywords: Vehicle Dynamics and Control, Advanced Driving Assistant System, Vehicle control
Abstract: The automatic parking scene of freight trucks is an important part of the "smart port". The trajectory tracking effect in the automatic parking scene directly determines the success of parking. Therefore, the development of stable and reliable trajectory tracking control algorithms has great practical applications value. This paper aims to develop a reversing trajectory tracking algorithm for semi-trailer trucks that meets the needs of actual engineering applications, so as to achieve the purpose of accurately and stably following the reversing planned trajectory. In this paper, under the input of trajectory planning based on kinematics model, a two-layer path following controller is designed. The first-level controller superimposes the feedforward control based on planning and the feedback control based on the trailer trajectory error to obtain the expected yaw rate of the tractor as the control output. The second-level controller outputs the steering wheel angle command to the EPS angle tracking system to realize the following control of the desired tractor yaw rate. The simulation verification under TruckSim/Simulink shows that the controller designed in this paper has good following performance and stability.

Keywords: Engine control
Abstract: It is a challenging problem in diesel engines to control throttle, variable geometry turbine(VGT) and exhaust gas recirculation (EGR). A effective method is model predictive control (MPC), which has been successfully applied to typical multi-input multi-output (MIMO) system with fast dynamics, actuator constraints, and strong couplings, such as diesel engines. In MPC controller design, the choice of output variables has a direct impact on the resulting control performance. Through investigating and discussing different selections of outputs, we propose that it is beneficial to select EGR-fraction and boost pressure as output variables while setting the oxygen fuel ratio as a constraint. Besides, equipping an integral action on the oxygen fuel ratio can improve the control performance.

Keywords: Vehicle Dynamics and Control, Vehicle control
Abstract: Phase plane is a visualized method to analyze the vehicle lateral behavior. The envelope control based on the sideslip angle and yaw rate phase plane can effectively restrict the vehicle states in the safe operation region for ensuring the stability of the vehicle. However, In most of such control schemes, many stable regions are not included in the envelope such that the dynamic performance of the vehicle is largely constrained. To solve this problem, this paper proposes a novel envelope control scheme based on saddle nodes position. It improves the overshoot dynamics by incorporating more stable regions in yaw rate direction and thus, improving the dynamic performance and steering agility. The simulation results indicate that compared to the previous envelope schemes, the proposed one can effectively improve the dynamic performance of the vehicle under the condition of ensuring stability.

Keywords: Vehicle Powertrain Control, Decision Making
Abstract: In order to shorten the power interruption time of the automobile AMT shift process, and restrain the shift shock, aiming at the control problem of the shifting actuator, the AMT variable parameter shift process control strategy is proposed. This method divides the shift process into three stages according to the synchronizer working characteristics, and adopts different control strategies. According to the fuzzy control theory, the fuzzy criterion rules are established to automatically determine the stage boundary point of the synchronizer shift process, and adopt different motor shift control strategies at each stage. It monitors whether the shift process is normal by means of redundant judgment. The test results show that this method can effectively shorten the shift time.