Search

When the linear Model Predictive Control (MPC) algorithm is used for trajectory tracking control of autonomous vehicles in high-speed turning conditions, the tracking accuracy is usually decreased due to the low fidelity of the predictive model. To deal with this problem, a trajectory tracking controller based on Neural Network Predictive Control (NNPC) is designed in this paper. The data collected from driving simulation experiments is used for the Back Propagation Neural Network (BPNN) training process to obtain the vehicle state predictive models. To reduce the iteration times of the rolling optimization module of NNPC, Particle Swarm Optimization (PSO) with Fitness Allocating Inertia Weights (PSO-FAIW) algorithm is proposed, which can allocate inertia weights in light of the distance between each particle’s fitness and the best fitness of the swarm. Finally, the performance of the controller designed in this paper is verified based on MATLAB /Simulink. The simulation results demonstrate that the controller proposed in this paper can give consideration to tracking accuracy and real-time performance.

This paper analyzes the lateral control technology of autonomous semi-trailer trucks. Existing researches on the lateral control algorithm of semi-trailer trucks focus on making the head-truck or trailer follow a track well while ignoring the motion characteristics during the turning process, leading to specific security issues. Meanwhile, it is difficult to cope with the complex and uncertain factors influencing lateral control effect, such as the curvature of the desired trajectory, the load, and the velocity of the semi-trailer truck. This paper proposes a parametric self-learning model predictive control (MPC) based on the Proximal Policy Optimization of One Step (OSPPO) method to solve these problems. After modeling the kinematics of the semi-trailer truck, a lateral motion controller for the relationship between the head-truck and trailer based on the MPC method is established. The traditional MPC method has difficulty in adapting to the changeable influencing factors. Thus, a deep reinforcement learning algorithm named OSPPO is introduced to improve the flexibility of the MPC method. OSPPO establishes the nonlinear mapping relationship between the critical parameter of the MPC method and the factors influencing control effect by self-learning, avoiding a large amount of labeled data for training. In simulations, Trucksim and Matlab were used to conduct co-simulation to verify the usefulness of the control method. The method was implemented on an autonomous semi-trailer truck for many outdoor scenes. The actual experimental results showed the validity and advantages of the proposed method.

In the field of vehicle system dynamics, revealing the dynamic characteristics of nonlinear vehicle models with high degrees of freedom is a key point in current research. In the classic 2-DOF (degrees of freedom) model, the dynamics is analysed and represented in two-dimensional state space. However, it is necessary to reveal the three-dimensional state characteristics of the system when analysing high-DOF vehicle models. The concept of equilibrium points in two-dimensional space will be extended to curves in three-dimensional space, including stable curves and unstable curves, which can qualitatively reveal the evolvement characteristics of the system under different control parameters. In addition, because the dissipation of energy method can quantitatively reveal the global dynamic characteristics of the vehicle, this paper combines the (un)stable curves in the three-dimensional space with dissipation of energy to better analyse the global dynamics characteristics and mutually verify the correctness of conclusions. In this paper, first, the connection between (un)stable curves and region boundary is established. Second, the theoretical analysis of relationship between dissipation of energy and phase trajectories convergence characteristics is perfected. Third, the static and dynamic characteristics of vehicle system are analysed. Therefore, this paper proposed the concept of stable and unstable curves to intuitively reveal the global dynamics characteristics and system evolvement characteristics in three-dimensional state space.

Rubber mounts are the vital component to reduce the vibration translated from the powertrain to the vehicle body. The dynamic stiffness of the conventional rubber mount increase with the increase of the excitation frequency, and form a peak under high-frequency excitation in the battery electric vehicle (BEV). So the double isolation rubber mount has gradually been used in the BEV powertrain to improve the vibration isolation performance in the high-frequency range. Compared with conventional rubber mounts, the double isolation rubber mount has smaller dynamic stiffness in the high-frequency range. However, the traditional simulation model of the dynamic characteristics cannot consider the influence of the metal bracket of the double isolation rubber mount on the dynamic characteristics. Therefore, it is necessary to carry out the modeling and analysis of the dynamic characteristics simulation model of the double isolation rubber mount. In this paper, a general dynamic model of double isolation rubber mounts is proposed to describe the high-frequency dynamic characteristics. In this general dynamic model, the complex stiffness element representing the viscoelasticity of rubber is expressed using the fractional derivative model and the Maxwell model, respectively, and two high-frequency dynamic models are established. Two model parameter identification methods are proposed for high-frequency dynamic models of double isolation rubber mounts, and the identification results are compared and analyzed. Then, two high-frequency dynamic models are used to simulate the high-frequency dynamic characteristics of double isolation rubber mounts. The simulation results show that both high-frequency dynamic models can calculate the high-frequency dynamic characteristics of double isolation rubber mounts. Among them, the dynamic model based on fractional derivative elements has relatively high accuracy, and the dynamic model based on Maxwell elements has relatively high stability. The high-frequency dynamic characteristic modeling method of double isolation rubber mounts presented in this paper is helpful to the vibration and noise analysis of electric vehicle powertrain systems under high-frequency excitation.

Semi-empirical tire model considering turn slip input is of great significance for the force feedback of vehicle at low speed or small turning radius conditions. In this paper a refined discrete model considering contact patch and the deformation of the belt/carcass which are from the finite element model is established. New force and moment calculation matrix equations are constructed and different iterative methods are compared and Richardson iteration has been chosen because of its best iteration speed. The tire dynamics characteristics under sideslip and turn slip inputs are analyzed based on the discrete model simulation results. It shows the PAC2002 influence coefficients of turn slip on peak value of side force and cornering stiffness can not be well expressed under different loads. And at small loads, the peak value and stiffness of aligning moment relative to turn slip under different sideslip angles can not well be expressed. According to above problems, the PAC2002 model is improved. After improvement, the average error of lateral force relative to sideslip angle under different turn slip is reduced from 3.7% to 1.4%, the average error of lateral force relative to turn slip under different sideslip angle is reduced from 9.9% to 3.6%. And the error of aligning moment relative to sideslip angle under different turn slip is reduced from 8.5% to 5.1% and the error of aligning moment relative to turn slip under different sideslip angle is reduced from 8.9% to 3.2% at small loads. The improvement proposed in this paper have a good result.

The steering feel is mostly indirectly adjusted by directly modifying the assist’s magnitude of the electric power steering system. The adjustment is complex, and workload is large, and the steering feel is highly dependent on the performance of the hardware. It has poor robustness to vehicle performance degradation and external disturbance, so providing a consistent and controllable steering feel is not easy. This paper proposes an EPS control strategy for steering feel control based on closed-loop control of steering motion to solve the above problems. Firstly, take the driver’s desired steering motion intensity and pinion angle position as intermediate variables to decompose the EPS control strategy into three parts: steering style, the desired pinion angle calculation, and the desired steering resistance moment calculation. This decomposition method makes the steering feel control independent of the vehicle dynamics characteristics and reduces the difficulty of vehicle characteristic calibration and correction. In addition, a steering style model based on Stevens’ law and considering hysteresis characteristics is proposed. The steering feel can be directly controlled based on calibrating the vehicle's dynamic characteristics and the designed steering style. The test results driven by ACSL driving simulators of Jilin University show that adopting the control strategy proposed in this paper can improve the vehicle’s return-to-center performance, realize the designed driving style, and provide consistent and controllable steering feel with the design hysteresis characteristic. In addition, it has good robustness against the deterioration of vehicle hardware performance and external disturbance.

The Brakes contact surface temperature rises sharply when vehicle braking by friction, so vehicles are generally equipped with ventilated brake discs for cooling. However, the recirculation zone in the ventilated brake disc passage severely blocks fluid flow. In order to improve the convective and heat dissipation performance of the ventilated brake disc, this paper aims to introduce an optimized slotted vane geometry to improve the heat dissipation performance of ventilated brake discs. Therefore, three structural parameters of slot width, slot inclination and number of vanes were selected as optimization variables, and the optimization objective was to maximize the mass flow rate and convection heat transfer coefficient. And then, combined with the RSM and NSGA-II for optimization. The results showed that the optimized slotted vanes effectively suppressed the recirculation zone in the flow passage and made the airflow distribution more uniform. The optimal characteristic parameters after optimization are N?=?36,??=?56.38°, H?=?12.7?mm. The CFD calculation results are consistent with the experimental results, and the Nu error obtained is about 2.6%. Within the given speed range, compared with experimental data of the straight vane brake disc, the total heat dissipation rate of the optimized brake disc is increased by 9%, the heat dissipation rate in the channel was increased by 5.6%, and the pumping capacity is increased by 12%.

Misunderstanding the driver behavior in the next short time is the primary reason of the false warning for the lane departure warning system. This paper proposes a learning-based evaluation to predict whether the driver notices the deviation of the vehicle and takes corrective actions. First, statistical Gaussian model and K-means clustering method are utilized to classify driving style of drivers and determine warning areas based on key driving parameters extracted in driving scenarios. Then, according to the vehicle trajectory in the warning area and the time of lane crossing (TLC) value of the two warning area boundaries, an advanced horizontal dual-area warning (HDAW) model that is trained by bi-direction long short-term memory (BiLSTM) originated from recurrent neural network (RNN) is applied to predict the lane departure and corrective behavior of driver. The personalized warning strategy is finally developed by considering driver characteristics, which allows the warning system to adapt to different driving styles of drivers. Natural driving data from 57 drivers in the experimental driving simulator are collected to train personalized prediction and verify proposed evaluation method. The recent directional sequence of piecewise lateral slopes (DSPLS) and traditional TLC are also researched and compared. Experimental results show that the proposed approach has as low as false alarm rate of 3.97% and can improve prediction accuracy approximately 41.39% over DSPLS method.

This paper proposes a novel integrated path following control scheme for a 4-Wheel Independent Drive (4WID) autonomous vehicle that can adaptively change its mechanism according to the driving conditions. The proposed integrated system handles the lateral steering controller, longitudinal speed, and yaw moment controls considering tire force capacity of each corner. For the lateral controller, the cornering stiffness uncertainties and the transient performance are considered and combined into an H8 robust controller based on linear matrix inequality (LMI) theory. A super-twisting sliding mode controller (STSMC) based longitudinal controller is designed to deal with disturbances and suppress chattering. When encountering extreme conditions, the active yaw moment controller with hierarchical structure is adaptively activated to prevent large deviation from the reference path and maintain the stability of vehicle. For the tire force allocation, an optimization algorithm is proposed, which has flexible equality constraints to coordinate the longitudinal and lateral motions according to the driving conditions. Simulations based on Carsim-Simulink co-simulation platform show that the proposed method is effective and has excellent performance in both normal and extreme driving conditions.

Modern diesel engines are commonly equipped with NOx sensors for precise and efficient selective catalytic reduction (SCR) reductant injection control, but due to the limitations of operating characteristics and environmental conditions, the output signal of NOx sensors may show abnormalities such as drift, deviation, or even distortion. Therefore, the evaluation and adjustment of the NOx sensor signal is essential to achieve intelligent control of the engine SCR system. In this study, an engine NOx calculation model is designed using deep neural network (DNN) to predict the upstream NOx emission concentration of engine SCR and provide reference values for sensor fault diagnosis, evaluate the sensor fault status by calculating the relative integration error index (IE), and reconstruct the sensor signal using the model prediction signal after the fault is confirmed. In addition, this paper uses the firefly algorithm (FA) to find the appropriate diagnostic threshold (IEthre) and integration period (T) to meet the requirements of diagnosis response speed. Finally, based on the data collected from World Harmonized Transient Cycle (WHTC) bench test of a commercial high-pressure common rail diesel engine, the IEthre= 12.76% and the T= 7.57 s. The root mean square error (RSME) of the NOx concentration signal predicted by the engine NOx calculation model was 37.35 ppm, and the normalized cumulative NOx emission was 98% of the actual signal. The performance of the designed algorithm was tested using the synthesized bias fault and drift fault signals. The sensor signal RSME was reduced from 108.46 to 29.66 ppm for the bias fault, and from 106.92 to 31.30 ppm for the drift fault.