Advanced driver-assistance systems (ADAS) are being developed for more and more complicated application scenarios, which often require more predictive strategies with better understanding of the driving environment. Taking traffic vehicles' maneuvers into account can greatly expand the beforehand time span for danger awareness. This article presents a maneuver-based strategy to vehicle collision threat assessment. First, a maneuver-based trajectory prediction model (MTPM) is built, in which near-future trajectories of ego vehicle and traffic vehicles are estimated with the combination of vehicle's maneuvers and kinematic models that correspond to every maneuver. The most probable maneuvers of ego vehicle and each traffic vehicles are modelled and inferred via Hidden Markov Models with mixture of Gaussians outputs (GMHMM). Based on the inferred maneuvers, trajectory sets consisting of vehicles' position and motion states are predicted by kinematic models. Subsequently, time to collision (TTC) is calculated in a strategy of employing collision detection at every predicted trajectory instance. For this purpose, safe areas via bounding boxes are applied on every vehicle, and Separating Axis Theorem (SAT) is applied for collision prediction so that TTC can be calculated efficiently and accurately. Finally, a threat level index based on reverse TTC is used to quantize the threat degree of every traffic vehicle potential collision to the ego vehicle. Experimental data collected in the field test are used in the model training, and the overall strategy is validated under PanoSim. An example of the application of the proposed strategy in Autonomous Emergency Braking (AEB) is also shown. Simulation results show that MTPM can accurately identify maneuvers such that the effective prediction on trajectories can be generated. TTC and threat index can be calculated timely. The proposed threat assessment strategy can not only assist collision avoidance systems to foresee dangerous situations but also eliminate false alarm to a certain extent.
Aiming at improving safety (anti-roll performance) with also consideration of ride comfort of vehicles during cornering and over road irregularities, magnetorheological (MR) fluid based semi-active anti-roll bar is investigated in this paper. The vehicle roll model with both roll stiffness and roll damping of the vehicle body influenced by the MR anti-roll bar is established to analyze the impact of the torsional stiffness and torsional damping. Combining with the Pareto front of the lateral load transfer ratio of the front axle, the optimal roll stiffness and roll damping of a vehicle are determined, and correspondingly the torsional stiffness and torsional damping of the anti-roll bar are determined. Then, the mathematical model and multibody dynamic model of the anti-roll bar are established, and the simulation of the MR semi-active anti-roll bar model is carried out via MATLAB/Simscape Multibody. CarSim vehicle model equipped with the MR anti-roll bar is built and a fuzzy controller is designed according to the roll angle and roll rate. Co-simulation based on CarSim and MATLAB/Simulink is conducted to analyze the impact of MR anti-roll bar on vehicle roll performance.
One of the techniques that accident reconstructionists and experts utilize to define the severity of an accident is based on the airbag deployment thresholds. As such, if during an event, the airbags did not deploy, it is concluded that the threshold could be considered as the upper bound for the forces and the accelerations that the vehicle experienced as a result of the impact. The National Highway Transportation Safety Administration (NHTSA) provides a database based on their investigations on motor vehicle accidents in which some of these investigations involved imaging the airbag control module (ACM) data. NHTSA made these data publicly available. The goal of this study was to analyze the event data recorder (EDR) data from these real-world incidents with a focus on the events in which vehicles' side airbags were deployed as a result of the impacts and determine the lower-bound side airbag deployment thresholds during real-world cases. In addition, this study is focused on the method of calculating the acceleration thresholds for airbag deployments and proposes an adjustment method that fits more adequately with the airbag module algorithm. Moreover, this article statistically studied the data to evaluate how the data may differ based on the crash type, vehicle body type, and manufacturers. The results showed that the lower limit for the maximum lateral acceleration of side airbag deployment is 3-5 g (g is the acceleration due to gravity). The corresponding lateral change in velocity (or delta-V) at maximum lateral acceleration was also calculated and analyzed. Ultimately, the study provides a range of lateral accelerations for side airbags deployment, which can be utilized to evaluate the severity of motor vehicle collisions. The analysis of the airbag deployment trigger points from the dataset is in general agreement with published data.
To address the difficulties in modeling the starting process of dual-clutch transmission (DCT) vehicles and poor adaptability of vehicles in complex driving conditions, this article proposes a new modeling and control strategy for the DCT starting system based on data-driven autoregressive moving average exogenous (ARMAX) modeling. Firstly, the DCT starting process is considered equivalent to the time series-related ARMAX model, and a data-driven ARMAX model could be obtained using input-output data relating to the starting process; also, the effectiveness of the data-driven ARMAX modeling technique is verified using the starting test of a real vehicle. Secondly, a data-driven adaptive model predictive control (A-MPC) strategy, which synthetically considers driving intention and clutch engagement status, is proposed. Finally, in order to verify the proposed control strategy, simulation analysis is conducted in different intentions; the results show that the proposed control strategy could realize the starting control effectively, and reflect driving intention. Compared with model predictive control only considering driving intention, the proposed control strategy could improve starting performance in different intentions; also, compared with the conventional control method, the A-MPC can improve the starting performance.
In this work, the multicarrier strategies for the three-phase five-level inverter are used on the rotor Field-Oriented Control (FOC) of the Induction Motor (IM). The H-bridge inverter gain uses the triangular carrier technique to produce two Pulse Width Modulation (PWM) command strategies. These two PWM-based strategies, the Phase Disposition Carrier-PWM (PDC-PWM) strategies and the Phase Shifted Carrier-PWM (PSC-PWM) strategies, are compared to find the appropriate command for the designed Electric Vehicle (EV) system. The system is improved by the Fuzzy Logic Control (FLC) to refine its surveillance and to detect any possible deflection in the system. The Automatic Hub (AH) is connected to the front axle of the EV. In case of any divergence, the FLC is programmed to detect the divergence according to the temperature of the semiconductors, the current, the speed, and the trajectory and then it changes the state of the AH to make the needed correction. An experimental assay is done in the Laboratory of Electronic Systems and Sustainable Energy (ESSE) to show the effect of the climatic and geographic conditions of Tunisia on the designed system. Furthermore, the three-phase multilevel inverter is coupled to the FOC-IM and then tested on the EV using the experimental assay and Matlab/Simulink. However, the realization of the designed system is not possible because it is expensive. Consequently, we now have an experimental assay of the FLC module and the simulation results that are used to verify the better command strategy between PSC-PWM and the PDC-PWM to improve the EV powertrain reliability.
Braking strategy is the core of the automatic emergency braking (AEB) system. How to make the braking strategy of the AEB system more suitable for real traffic conditions and more acceptable to drivers is a very important research. This article carries out the braking strategy optimization for the AEB system. From the vehicle-pedestrian accidents collected by the National Automobile Accident In-depth Investigation System (NAIS) of Songjiang Station, this article selects typical cases equipped with event data recorder (EDR) and then analyzes the EDR data. On this basis, the strategy for the optimization of the AEB system is formed. The optimized strategy is simulated in the test scenario specified in the European New Car Assessment Programme (Euro-NCAP) and compared with the real vehicle test results. The result illustrates that the optimized braking strategy is more effective to avoid or mitigate collisions.
This study explores the multivariable multi-input-multi-output technique based on nonlinear models to decouple actuator interaction and to reduce the calibration workload, as well as to increase control performances, under transient conditions, and also explores the robustness on model uncertainties and system parameter variations. The development of a nonlinear dynamic physical model of air in gasoline engine and its charging system provides the for the control technology. The model uses feedback linearization control to decouple the interaction between actuators and compensate the nonlinearity. A new set of inputs was defined through inversing the differential equation of the system. The relationship between the new inputs and the output is linear and decoupled. In addition, a linear control module is used to ensure transient and steady-state performance as well as closed-loop robustness. The control method has been confirmed on the bench test with a three-cylinder gasoline engine prototype for hybrid electric vehicles. Transient test results show that the controller is able to coordinate the actuator to achieve the targets.
ICVs are expected to make the transportation safer, cleaner, and more comfortable in the near future. However, the trend of connectivity has greatly increased the attack surfaces of vehicles, which makes in-vehicle networks more vulnerable to cyberattacks which then causes serious security and safety issues. In this article, we therefore systematically analyzed cyberattacks and corresponding countermeasures for in-vehicle networks of intelligent and connected vehicles (ICVs). Firstly, we analyzed the security risk of ICVs and proposed an in-vehicle network model from a hierarchical point of view. Then, we discussed possible cyberattacks at each layer of proposed network model. After that, we provided an overview of the state of the art of the potential countermeasures against cyberattacks, such as secure hardware architecture, encryption and authentication, network firewall, intrusion detection system, and secure Firmware over the air (FOTA), and then a reasonable defense mechanism is proposed. At last, some challenges and future works related to cyberattacks against ICVs were discussed. This article aims to review the cyberattacks for ICVs and suggest comprehensive security countermeasures.
To address the problem of a large inrush current in pure electric vehicles under high-power driving and braking conditions, which damages the cycle life of the battery, a hydraulic auxiliary drive system is used to design a four-wheel-drive sport utility vehicle (SUV) electric-hydraulic hybrid(EHH) power system. Energy management strategy (EMS) research on this EHH power system is conducted. An EMS based on optimal instantaneous efficiency is proposed. The simulation results under the China light-duty vehicle test cycle for passenger car (CLTC-P) show that control effect of the EMS can improve by 3.27% than the EMS based on rule and achieve 97.43% of control effect of the dynamic programming(DP) global optimal EMS in terms of economic performance. Then the model predictive control (MPC) method is used, and the instantaneous efficiency optimal EMS is optimized. The simulation results under the CLTC-P show that the instantaneous efficiency optimal EMS optimized by MPC has an economic performance improvement of 1.13% compared with the performance before optimization. It can improve by 4.36% than the EMS RB and achieve 98.54% of control effect of the dynamic programming(DP) global optimal EMS in terms of economic performance.
Present-day vehicles come with a variety of new features like the pre-crash warning, the vehicle-to-vehicle communication, semi-autonomous driving systems, telematics, drive by wire. They demand very high bandwidth from in-vehicle networks. Various ECUs present inside the automotive transmits useful information via automotive multiplexing. Transmission of data in real-time achieves optimum functionality. The high bandwidth and high-speed requirement can be achieved either by using multiple buses or by implementing higher bandwidth. But, by doing so, the cost of the network as well as the complexity of the wiring increases. Another option is to implement higher layer protocol which can reduce the amount of data transferred by using data reduction (DR) techniques, thus reducing the bandwidth usage. The implementation cost is minimal as the changes are required in the software only and not in hardware. This article presents a new data reduction algorithm termed as "Comprehensive Data Reduction (CDR)" algorithm. The article also demonstrates a comparison of the proposed algorithm with the boundary of fifteen compression algorithms and compression area selection algorithms. The results show that the proposed CDR algorithm provides better data reduction compared to the earlier proposed algorithms. The proposed algorithm has been developed for automotive environment, but it can also be utilized in any applications where extensive information transmission among various control units is carried out via a multiplexing bus.
Li-ion batteries have been widely applied in the areas of personal electronic devices, stationary energy storage system and electric vehicles due to their high energy/power density, low self-discharge rate and long cycle life etc. For the better designs of both the battery cells and their thermal management systems, various numerical approaches have been proposed to investigate the thermal performance of power batteries. Without the requirement of detailed physical and thermal parameters of batteries, this paper proposed a data-driven model using the adaptive neuro-fuzzy inference system (ANFIS) to predict the battery temperature with the inputs of ambient temperature, current and state of charge. Thermal response of a Li-ion battery module was experimentally evaluated under various conditions (i.e. ambient temperature of 0, 5, 10, 15 and 20 degrees C, and current rate of C/2, 1C and 2C) to acquire the necessary data sets for model development and validation. A Sugeno-type ANFIS model was tuned using the obtained data. The numbers of input membership functions (MFs) representing the three input parameters of this model are 1, 2, 3, respectively. The input and output MFs are Gaussian curve and linear types, respectively. The optimization method is a hybrid one which is a combination of the back-propagation and the least squares methods. Compared with the validating data, the ANFIS model was able to accurately predict the battery temperature under various operating conditions. With fewer sensors for data acquisition and less computation complexity, this method could be a possible tool for the online temperature prediction of power batteries in electric vehicle applications.
The paper presents the application of grey wolf algorithm for multidimensional engine optimization of converted parallel operated diesel plug-in hybrid electric vehicle to optimize specific fuel consumption (FC) and emissions. All emissions hydrocarbon (HC), carbon monoxide (CO), nitrogen oxide (NOx) and particulate matter (PM) are considered as optimization parameters. Offline engine maps of FC, HC, CO, NOx and PM are generated for 70 hp engine by data obtained from Oak Ridge National Laboratory for study. MATLAB program is used for simulation. A grey wolf coding is developed and tested extensively for various values of speed and torque. The optimization results obtained are verified by available engine maps. The optimization performance and its environmental impact are discussed in detail. It is observed that grey wolf optimizer (GWO) gives the global minimum value with slight deviation, although least computation time and simplicity makes this algorithm a potential candidate for real-time implementation.
EMC Component Validation Responsibilities encompass many realms. One of these realms is the effect of magnetic fields on silicon-based devices. This article describes a method for exposing these devices to magnetic fields with waveforms other than the traditional sinusoidal excitation. The method commonly used to explore the sensitivity of active silicon devices is exposure of the device to a representative sinusoidal field and observation of its reaction or lack thereof. The challenge is to characterize the representative field and be able to verify its effectiveness. Recent vehicle level testing of new designs has brought our attention to time-varying or transient magnetic field shapes that create deviations not previously detected with Military Standard 461 (MIL-STD-461) type sinusoidal magnetic field exposure. A facsimile of a number of existing and well-known vehicle transients are excited through a physical and simulated system to prepare representative and repeatable transient inputs for use with existing lab equipment.
Electric power steering (EPS) system is a kind of dynamic control system for vehicle steering, which can amplify the driver steering torque inputs to the vehicle to improve steering comfortable and performance, but the present EPS can't cater to the driving habits of different people. In this article, a personalized EPS controller is designed based on the driver behavior, which combines real-time driver behavior identification strategy with personalized assistance characteristic. Firstly, the driver behavior data acquisition system is designed and established, based on which, the input data of different kinds of drivers along with vehicle signals are collected under typical working conditions, then the identification of driver behavior online is realized using the BP neural network. Secondly, the personalized assistance characteristic curve is selected according to the identification results, and the close loop proportional-integral-derivative (PID) control strategies and pulse width modulation (PWM) method are adopted to control the target current of the motor of the EPS system. Finally, the co-simulation of Simulink and Carsim are carried out, and the results show that personalized electric power steering system based on driver behavior can adjust power-assisted characteristics timely, and it can match the driver usage modes automatically and meet the drivers' steering power demands better.
Although energy harvesting systems are extensively used in different fields, studies on the application of energy harvesters embedded in tires for vehicle control are rare and mostly focus on solving power supply problems of tire pressure sensors. Sensors are traditionally powered by an embedded battery, which must be replaced periodically because of its limited energy storage. Heightened interest in vehicle safety is expected to drive increased design and manufacture of in-tire sensors, which in turn, translates to rising demand for power generation in tires. These challenges emphasize the need to investigate the substitution of batteries and in-tire energy harvesting systems. Current in-tire energy harvesting methods involve piezoelectric, electromagnetic, and electrostatic power generation, whose energy sources include tire vibrations, deformations, and rotations. Piezoelectric harvesters are generally compact but operate for short durations. Electromagnetic generators exhibit higher power density but are less efficient than generators made from piezoelectric materials. The challenge presented by electrostatic generators is their requirement for certain launch conditions. The performance of different mechanisms depends on design, which is the key issue examined in research. The amount of harvested energy ranges from the microwatt to dozens of watts level, depending on individual structure and chosen principle. This article analyzes studies on in-tire energy harvesting systems to comprehensively compare existing principles and structures, as well as evaluate prospects for application. The modeling, simulation, and experimental validation in previous works are also summarized and categorized to provide guidance for future studies.
Real-time reconstruction of 3D environment attributed with semantic information is significant for a variety of applications, such as obstacle detection, traffic scene comprehension and autonomous navigation. The current approaches to achieve it are mainly using stereo vision, Structure from Motion (SfM) or mobile LiDAR sensors. Each of these approaches has its own limitation, stereo vision has high computational cost, SfM needs accurate calibration between a sequences of images, and the onboard LiDAR sensor can only provide sparse points without color information. This paper describes a novel method for traffic scene semantic segmentation by combining sparse LiDAR point cloud (e.g. from Velodyne scans), with monocular color image. The key novelty of the method is the semantic coupling of stereoscopic point cloud with color lattice from camera image labelled through a Convolutional Neural Network (CNN). The presented method comprises three main process: (I) perform semantic segmentation on color image from monocular camera by using CNN, (II) extract ideal surfaces and other structural information from point cloud, (HI) improve the image segmentation with the extracts and label the point cloud with the image segments. The whole process is done in a single frame, and the output of the system is labelled point cloud which can be used in construction of semantic object convex and alignment between frames. We demonstrate the effectiveness of our system on the KITTI dataset providing sufficient camera and LiDAR data, and present qualitative and quantitative results indicating the improvements in segmentation comparing to methods merely using either image or LiDAR data.
Heat generation characteristics of lithium ion batteries are vital for both the optimization of the battery cells and thermal management system design of battery packs. Compared with other factors, internal resistance has great influence on the thermal behavior of Li-ion batteries. Focus on a 3 Ah pouch type battery cell with the NCM/C material system, this article quantitatively evaluates the battery heat generation behavior using an Extended Volume-Accelerating Rate Calorimeter in combination with a battery cycler. Also, internal resistances of the battery cell are measured using both the hybrid pulse power characteristic (HPPC) and electro-chemical impedance spectroscopy (EIS) methods. Experimental results show that the overall internal resistance obtained by the EIS method is close to the ohmic resistance measured by the HPPC method. Heat generation power of the battery cell is small during discharge processes lower than 0.5 C-rate. The curve of heat generation power vs. time shows a U-shaped characteristic that displays some symmetry when the current rate is high. Compared with the EIS method, internal resistances measured by the HPPC method have a more close relationship with the heat generation behavior. Influenced by the internal resistance, the battery heat generation power is much higher during the beginning period of charge process and the end period of discharge process. in addition, battery heat generation in this article is mainly composed of irreversible ohmic heat, while the reversible heat and side reaction heat are relatively small in quantity.
Potential collisions with oncoming traffic while turning left belong to the most safety-critical situations accounting for similar to 25% of all intersection crossing path crashes. A Left Turn Assist (LTA) was developed to reduce the number of crashes. Crucial for the effectiveness of the system is the design of the human-machine interface (HMI), i.e. defining how the system uses the calculated crash probability in the communication with the driver. A driving simulator study was conducted evaluating a warning strategy for two use cases: firstly, the driver comes to a stop before turning (STOP), and secondly, the driver moves on without stopping (MOVE). Forty drivers drove through three STOP and two MOVE scenarios. For the STOP scenarios, the study compared the effectiveness of an audio-visual warning with an additional brake intervention and a baseline. For the MOVE scenarios, the study analyzed the effectiveness of the audio-visual warning against a baseline. The results showed that the brake intervention is highly effective resulting in significantly larger minimal distances between the two vehicles. For the MOVE scenarios, the warning strategy is only effective in one scenario, reflecting the higher complexity of MOVE scenarios with regard to the prediction of crash probabilities due to the variability of driving/turn trajectories. Concluding, the brake intervention in STOP scenarios could be further investigated as a promising HMI strategy. Future studies should figure out the appropriate strength of a brake intervention in a real vehicle. MOVE scenarios face big challenges due to the variance of driver behavior in turning situations.
Knowledge of intelligent vehicle absolute position is a vital premise for the implementation of decision programming, kinematic and dynamics control. In order to achieve high accuracy positioning and reduce running cost as much as possible under all operating conditions, this paper proposed an integrated positioning method based on GPS and Ultra Wide Band(UWB) for intelligent vehicle's navigation and position system. In this method, GPS and UWB are alternately active according to the confidence level of GPS signal. When the vehicle is traveling in a wide-open area and GPS signal is well received, the positioning results of Dead Reckoning system are corrected by the low frequency positioning output from GPS. During the correcting process, in order to realize the better fusion of measurement data, a simplified federal Kalman filter was designed by using indirect method. When the vehicle is in places where GPS signal can hardly be received such as tunnel, the positioning results based on UWB positioning technology can be adopted to substitute the lost GPS signal for vehicle integrated positioning. The algorithm used in the UWB positioning technology was two-phase positioning algorithm based on the signal arrival time, and Gaussian filtering method was also used in the pretreatment process of distance measuring values. Finally, a working test under the typical condition was conducted on the Matlab/Simulink-Carsim co-simulation platform. Simulation results demonstrate that even the vehicle is in the scenario without GPS and sensors are low cost, a better positioning accuracy can be still achieved with the integrated positioning method proposed in this paper.
In Advanced Driver Assistant System (ADAS), the automotive radar is used to detect targets or obstacles around the vehicle. The procedure of Constant False Alarm Rate (CFAR) plays an important role in adaptive targets detection in noise or clutter environment. But in practical applications, the noise or clutter power is absolutely unknown and varies over the change of range, time and angle. The well-known cell averaging (CA) CFAR detector has a good detection performance in homogeneous environment but suffers from masking effect in mufti-target environment. The ordered statistic (OS) CFAR is more robust in multi-target environment but needs a high computation power. Therefore, in this paper, a new two-dimension CFAR procedure based on a combination of Generalized Order Statistic (GOS) and CA CFAR named GOS-CA CFAR is proposed. Besides, the Linear Frequency Modulation Continuous Wave (LFMCW) radar simulation system is built to produce a series of rapid chirp signals. Then the echo signals are converted into a two-dimensional Range-Doppler matrix RDM), which contains information about the targets as well as background clutter and noise, through twice Fast Fourier Transform (FFT).The simulation experimental results show that compared to the two-dimensional OS-CA CFAR, the new 2-D GOS-CA CFAR can enhance the detection performance and robustness in the actual multi-target environment with lower computational complexity.
