Search Results

Multi-Target tracking is a central aspect of modeling the surrounding environment of autonomous vehicles. Automotive millimeter-wave radar is a necessary component in the autonomous driving system. One of the biggest advantages of radar is it measures the velocity directly. Another big advantage is that the radar is less influenced by environmental conditions. It can work day and night, in rainy or snowy conditions. In the expressway scenario, the forward-looking radar can generate multiple objects, to properly track the leading vehicle or neighbor-lane vehicle, a multi-target tracking algorithm is required. How to associate the track and the measurement or data association is an important question in a multi-target tracking system. This paper applies the nearest-neighbor method to solve the data association problem and uses an extended Kalman filter to update the state of the track.

Wheel Aerodynamics is an important part of vehicle aerodynamics. The wheels can notably influence the total aerodynamic drag, lift and ventilation drag of vehicles. In order to simulate the real on-road condition of driving cars, the moving ground and wheel rotation is of major importance in CFD. However, the wheel rotation condition is difficult to be represented exactly, so this is still a critical topic which needs to be worked on. In this paper, a study, which focuses on two types of cars: a fastback sedan and a notchback DrivAer, is conducted. Comparing three different wheel rotating simulation methods: steady Moving wall, MRF and unsteady Sliding Mesh, the effects of different methods for the numerical simulation of vehicle aerodynamics are revealed. Discrepancies of aerodynamic forces between the methods are discussed as well as the flow field, and the simulation results are also compared with published experimental data for validation.

Distributed drive electric vehicle (EV) is driven by four independent hub motors mounted directly in wheels and retains traditional hydraulic brake system. So it can quickly produce driving/braking motor torque and large stable hydraulic braking force. In this paper a new control allocation strategy for distributed drive electric vehicle is proposed to improve vehicle's lateral stability performance. It exploits the quick response of motor torque and controllable hydraulic pressure of the hydraulic brake system. The allocation strategy consists of two sections. The first section uses an optimal allocation controller to calculate the total longitudinal force of each wheel. In the controller, a dynamic efficiency matrix is designed via local linearization to improve lateral stability control performance, as it considers the influence of tire coupling characteristics over yaw moment control in extreme situations.

Life prediction is a major focus for a commercial fuel cell stack, especially applied in fuel cell electric vehicles (FCEV). This paper proposes a data driven fuel cell lifetime prediction model using particle swarm optimized back-propagation neural network (PSO-BPNN). For the prediction model PSO-BP, PSO algorithm is used to determine the optimal hyper parameters of BP neural network. In this paper, total voltage of fuel cell stack is employed to represent the health index of fuel cell. Then the proposed prediction model is validated by the aging data from PEMFC stack in FCEV at the actual road condition. The experimental results indicate that PSO-BP model can predict the voltage degradation of PEMFC stack at actual road condition precisely and has a higher prediction accuracy than BP model.

The research and development of data-driven highly automated driving system components such as trajectory prediction, motion planning, driving test scenario generation, and safety validation all require large amounts of naturalistic vehicle trajectory data. Therefore, a variety of data collection methods have emerged to meet the growing demand. Among these, camera-equipped drones are gaining more and more attention because of their obvious advantages. Specifically, compared to others, drones have a wider field of bird's eye view, which is less likely to be blocked, and they could collect more complete and natural vehicle trajectory data. Besides, they are not easily observed by traffic participants and ensure that the human driver behavior data collected is realistic and natural. In this paper, we present a complete vehicle trajectory data extraction framework based on aerial videos. It consists of three parts: 1) objects detection, 2) data association, and 3) data cleaning.

The active engine mount (AEM) is developed in automotive industry to improve overall NVH performance. The AEM is designed to reduce major-order signals of engine vibration over a broad frequency range, therefore it is of vital importance to extract major-order signals from vibration before the actuator of the AEM works. This work focuses on a method of real-time extraction of the major-order acceleration signals at the passive side of the AEM. Firstly, the transient engine speed is tracked and calculated, from which the FFT method with a constant sampling rate is used to identify the time-related frequencies as the fundamental frequencies. Then the major-order signals in frequency domain are computed according to the certain multiple relation of the fundamental frequencies. After that, the major-order signals can be reconstructed in time domain, which are proved accurate through offline simulation, compared with the given signals.

Recently, several datasets have become available for occupant monitoring algorithm development, including real and synthetic datasets. However, real data acquisition is expensive and labeling is complex, while virtual data may not accurately reflect actual human physiology. To address these issues and obtain high-fidelity data for training intelligent driving monitoring systems, we have constructed a hybrid dataset that combines real driving image data with corresponding virtual data generated from 3D driving scenarios. We have also taken into account individual anthropometric measures and driving postures. Our approach not only greatly enriches the dataset by using virtual data to augment the sample size, but it also saves the need for extensive annotation efforts. Besides, we can enhance the authenticity of the virtual data by applying ergonomics techniques based on RAMSIS, which is crucial in dataset construction.

Speech enhancement can extract clean speech from noise interference, enhancing its perceptual quality and intelligibility. This technology has significant applications in in-car intelligent voice interaction. However, the complex noise environment inside the vehicle, especially the human voice interference is very prominent, which brings great challenges to the vehicle speech interaction system. In this paper, we propose a speech enhancement method based on target speech features, which can better extract clean speech and improve the perceptual quality and intelligibility of enhanced speech in the environment of human noise interference. To this end, we propose a design method for the middle layer of the U-Net architecture based on Long Short-Term Memory (LSTM), which can automatically extract the target speech features that are highly distinguishable from the noise signal and human voice interference features in noisy speech, and realize the targeted extraction of clean speech.

Image corruptions due to noise, blur, contrast change, etc., could lead to a significant performance decline of Deep Neural Networks (DNN), which poses a potential threat to DNN-based autonomous vehicles. Previous works attempted to explain corruption from a Fourier perspective. By comparing the absolute Fourier spectrum difference between corrupted images and clean images in the RGB color space, they regard the noise from some corruptions (Gaussian noise, defocus blur, etc.) as concentrating on the high-frequency components while others (contrast, fog, etc.) concentrate on the low-frequency components. In this work, we present a new perspective that unifies corruptions as noise from high frequency and thus propose an image augmentation algorithm to achieve a more robust performance against common corruptions. First, we notice the 1/fα statistical rule of the natural image's spectrum and the channels-wise differential sensitivity on the YCbCr color space of the Human Visual System.

As a prerequisite for energy management of hybrid vehicles, the results of speed prediction can optimize the performance of vehicles and improve fuel efficiency. Energy management strategies are usually developed based on standard driving cycles, which are too generalized to show the variability of driving conditions in different time and locations. Therefore, this paper constructs a representative driving cycle based on driving data of the corresponding time and location, used as historical information for prediction. We propose a method to construct the driving cycle based on Markov chain theory before constructing the prediction model. In this paper, multiple prediction methods are compared with traditional parametric methods. The difference in prediction accuracy between multiple prediction methods under the single time scale and multiple time scale were compared, which further verified the advantages of the speed prediction method based on Markov chain theory.

Tire cavity noise of pure electric vehicles is particularly prominent due to the absence of engine noise, which are usually eliminated by adding Helmholtz resonators with arbitrary transversal section to the wheel rims. This paper provides theoretical basis for accurately predicting and effectively improving acoustic performance of wheel resonators. A hybrid finite element method is developed to extract the transversal wavenumbers and eigenvectors, and the mode-matching scheme is employed to determine the transmission loss of the Helmholtz resonator. Based on the accuracy validation of this method, the matching design of the wheel resonators and the optimization method of tire cavity noise are studied. The identification method of the tire cavity resonance frequency is developed through the acoustic modal test. A scientific transmission loss target curve and fitness function are defined according to the noise characteristics.

On the study of reducing the disturbance on driver’s attention induced by low frequency vehicle interior stationary noise, a subjective evaluation is firstly carried out by means of rank rating method which introduces Distraction Level (DL) as evaluation index. A visual-finger response test is developed to help evaluating members better recognize the Distraction Level during the evaluation. A non-linear back propagation artificial neural network (BPANN) is then modeled for the prediction of subjective Distraction Level, in which linear sound pressure RMS amplitudes of five Critical Band Rates (CBRs) from 20 to 500Hz are selected as inputs of the model. These inputs comprise an input vector of BPANN. Furthermore, active noise equalization (ANE) on DL is realized based on Filtered-x Least Mean Square (FxLMS) algorithm that controls the gain coefficients of inputs of trained BPANN.

The adhesion control is the basic technology of active safety for the four-wheel driven EV. In this paper, a novel adhesion control method based on fuzzy logic control is proposed. The control system can maximize the adhesion force without road condition information and vehicle speed signal. Also, the regulation torque to prevent wheel slip is smooth and the vehicle driving comfort is greatly improved. For implementation, only the rotating speed of the driving wheel and the motor driving torque signals are needed, while the derived information of the wheel acceleration and the skid status are used. The simulation and road test results have shown that the adhesion control method is effective for preventing slip and lock on the slippery road condition.

As turbulence modeling has become an indispensable approach to perform flow simulation in a wide range of industrial applications, how to enhance the prediction accuracy has gained increasing attention during the past years. Of all the turbulence models, RANS is the most common choice for many OEMs due to its short turn-around time and strong robustness. However, the default setting of RANS is usually benchmarked through classical and well-studied engineering examples, not always suitable for resolving complex flows in specific circumstances. Many previous researches have suggested a small tuning in turbulence model coefficients could achieve higher accuracy on a variety of flow scenarios. Instead of adjusting parameters by trial and error from experience, this paper introduced a new data-driven method of turbulence model recalibration using adjoint solver, based on Generalized k-ω (GEKO) model, one variant of RANS.

Based on the four-wheel-drive electric vehicle (4WD EV), a variable structure control (VSC) strategy is designed in this paper for the anti-lock braking control. With nonpeak friction coefficient as target, sign judgment method of switch function in this VSC strategy is improved and a new control algorithm is proposed. The improved VSC strategy is made robust to the parameters of the algorithm and verified by the computer simulation as well as the hard-in-loop test. The results show that the slip rate can be controlled to a point in the stable area near the optimal slip ratio and the control strategy can effectively realize the anti-lock braking control.

Intrusion Detection Systems (IDS), technically speaking, is to monitor the network, system, and operation status according to certain security policies, and try to find various attack attempts, attacks or attack results to ensure the confidentiality, integrity and availability of network system resources. Automotive intrusion detection systems can identify and alert by analyzing in-vehicle traffic and log when software applications or devices with malicious activity exist, or the in-vehicle network is tampered and injected. But unfortunately, automotive cybersecurity researchers hardly produce a comprehensive detection method due to the confidential nature of Controller Area Network (CAN) DBC format files, which is a standard long maintained by car manufacturers. In this paper, an enhanced intrusion detection method is proposed based on the double-decision-tree to classify different attack models for in-vehicle CAN network without the need to obtain complete DBC files.

Tire forces and moments play an important role in vehicle dynamics and safety. X-by-wire chassis components including active suspension, electronic powered steering, by-wire braking, etc can take the tire forces as inputs to improve vehicle’s dynamic performance. In order to measure the accurate dynamic wheel load, most of the researches focused on the kinematic parameters such as body longitudinal and lateral acceleration, load transfer and etc. In this paper, the authors focus on the suspension system, avoiding the dependence on accurate mass and aerodynamics model of the whole vehicle. The geometry of the suspension is equated by the spatial parallel mechanism model (RSSR model), which improves the calculation speed while ensuring the accuracy. A suspension force observer is created, which contains parameters including spring damper compression length, push rod force, knuckle accelerations, etc., combing the kinematic and dynamic characteristic of the vehicle.

Based on the emergency lane change cases extracted from the China naturalistic driving data, the driving steering behavior divides into three phases: collision avoidance, lateral movement and steering stabilization. Using the steering primitive fitting by Gaussian function, the distribution of the duration time, the relationship between steering wheel rate and deflection were analyzed in three phases. It is shown that the steering behavior essentially is composed of steering primitives during the emergency lane-change. However, the combination of the steering primitives is different according to the specific steering constraints in three phases. In the collision avoidance phase, a single steering primitive with high peak is used for the fast steering; in the lateral movement and stabilization phase, a combination of two or even more steering primitives is built to a more accurate steering.

Two control strategies, safety preferred control and master cylinder oscillation control, were designed for anti-lock braking on a novel integrated-electro-hydraulic braking system (I-EHB) which has only four solenoid valves in its innovative hydraulic control unit (HCU) instead of eight in a traditional one. The main idea of safety preferred control is to reduce the hydraulic pressure provided by the motor in the master cylinder whenever a wheel tends to be locking even if some of the other wheels may need more braking torque. In contrast, regarding master cylinder oscillation control, a sinusoidal signal is given to the motor making the hydraulic pressure in the master cylinder oscillate in certain frequency and amplitude. Hardware-in-the-loop simulations were conducted to verify the effectiveness of the two control strategies mentioned above and to evaluate them.

Vision based driving environment perception is current research hotspot in automatic driving field, which has made great progress due to the continuous breakthroughs in the research of deep neural network. As is well known, deep neural network has won tremendous successes in a wide variety of image recognition tasks, such as pedestrian detection and vehicle identification, which have accomplished the commercialization successfully in intelligent monitor system. Nevertheless, driving environment perception has a higher request for the generalization performance of deep neural network, which needs further studies on its design and training methods. In this paper, we presented a new boosted deep neural network in order to improve its generalization performance and meanwhile keep computational budget constant. Above all, the most representative methods to improve the generalization performance of deep neural network were introduced.