Search

An accurate state of charge (SOC) can provide effective judgment for the BMS, which is conducive for prolonging battery life and protecting the working state of the entire battery pack. In this study, the first-order RC battery model is used as the research object and two parameter identification methods based on the least square method (RLS) are analyzed and discussed in detail. The simulation results show that the model parameters identified under the Federal Urban Driving Schedule (HPPC) condition are not suitable for the Federal Urban Driving Schedule (FUDS) condition. The parameters of the model are not universal through the HPPC condition. A multitimescale prediction model is also proposed to estimate the SOC of the battery. That is, the extended Kalman filter (EKF) is adopted to update the model parameters and the adaptive unscented Kalman filter (AUKF) is used to predict the battery SOC. The experimental results at different temperatures show that the EKF-AUKF method is superior to other methods. The algorithm is simulated and verified under different initial SOC errors. In the whole FUDS operating condition, the RSME of the SOC is within 1%, and that of the voltage is within 0.01 V. It indicates that the proposed algorithm can obtain accurate estimation results and has strong robustness. Moreover, the simulation results after adding noise errors to the current and voltage values reveal that the algorithm can eliminate the sensor accuracy effect to a certain extent. [ABSTRACT FROM AUTHOR]

For hydrogen fuel cell vehicles, energy management strategies (EMS) are vital for balancing fuel cell and battery power, limiting fuel cell power, maintaining state of charge (SOC) fluctuation range and mitigating degradation. Reinforcement learning-based EMS, especially using deep Q-network (DQN) and deep deterministic policy gradient (DDPG), have demonstrated potential for enhancing the fuel economy of hybrid electric vehicles (HEV) and fuel cell electric vehicles (FCEV). This research proposes mutation protection DQN (MPD)-based EMS to improve hydrogen fuel economy and reduce fuel cell degradation under driving cycles with plenty of mutations. By quantifying mutation, exploring its relationship with driving conditions and integrating a mutation protection module with DQN, MPD-based EMS achieves approximately 11% and 6% better fuel economy compared to the other two learning-based EMS. Additionally, it also reduces fuel cell degradation by approximately 21% and 13%. • Quantifying the mutation of driving conditions through variance. • Utilizing information entropy to process mutations. • Introducing a dynamic mutation threshold value. • Designing mutation protection module and integrating it into the DQN. [ABSTRACT FROM AUTHOR]

The taper leaf spring of tandem suspension of commercial vehicle is different from the traditional taper leaf spring. Thus, the professional software MSC.ADAMS/CHASSIS leaf spring, which is specially applied in traditional leaf spring, does not reliably develop the computational model of the taper leaf spring of tandem suspension. The multi-body dynamic model of the taper leaf spring of tandem suspension, which is developed in this paper, is a secondary development model that is based on the leaf spring model built by MSC.ADAMS/CHASSIS leaf spring. End contact and friction in the modified model are redefined to exhibit the hysteretic characteristics of the taper leaf spring of tandem suspension. The test of the taper leaf spring of tandem suspension is conducted to validate the reliability of the modified model. The tests in this paper are divided into two groups. The first group started at an unloaded state at an excitation frequency of 1/30 Hz and amplitude of 70 mm to acquire quasi-static behavior. The second group is conducted at various frequencies (2, 3 and 4 Hz) and various amplitudes (±1, ±3 and ±10 mm) in a loaded state to acquire dynamic behavior. A formula to calculate dynamic spring rate for leaf spring is proposed, and details about the formula are presented. The simulations are conducted under the same conditions as the test. The hysteretic characteristics and the relative error of dynamic spring rate from the test are compared with the ones from the simulations for the validation of the reliability of the modified model. [ABSTRACT FROM AUTHOR]

For four-wheel independently driven (4WD) distributed electric vehicle (DEV), vehicle dynamics control systems such as direct yaw moment control (DYC) can be easily achieved. Accurate estimation of vehicle state variables and uncertain parameters can improve the robustness of vehicle dynamics control system. Various sensors are generally equipped to the acquisition of the vehicle dynamics. For both technical and economic reasons, some fundamental vehicle parameters, such as the sideslip angle and tire-road forces, can hardly be obtained through sensors directly. Therefore, this paper presented a state observer to estimate these variables based on Unscented Kalman Filter (UKF). To improve the accuracy of UKF, measurement noise covariance is also self-adaptive regulated. In addition, a nonlinear dynamics tire model is utilized to improve the accuracy of tire lateral force estimation. The simulation and experiment results show that the proposed observer can provide the precision values of the vehicle state. [ABSTRACT FROM AUTHOR]

Energy management is a fundamental task of a hybrid electric vehicle. However, dealing with multiple hybrid electric vehicles would be very time consuming, and developing a separate management strategy for each model is a huge workload to. Based on the above problems, this paper investigates the generalization capability of energy management strategies for hybrid electric vehicles. To improve the generalization of energy management strategies, a multi-agent reinforcement learning algorithm is proposed. To achieve this goal, the first analysis from the state values of reinforcement learning in the state selection, if all the typical features of the vehicle operation are added to the reinforcement learning algorithm, then it will make the model have a certain generalization ability. Then, with the help of the auxiliary agent, the reward value of reinforcement learning can be improved by using KL-divergence. The training and validation results show that the strategy can also achieve the training effect when tested on new models. In addition, a new driving cycle is selected for environmental testing, and the results show that the method also has strong generalization ability. • Diverse vehicle states are transferred into high-dimensional features. • Two DQN network are trained together with correlation function. • An Auxiliary agent is applied to guide the direction of agent training. • The proposed deep reinforcement learning framework realizes better performance and generalization. [ABSTRACT FROM AUTHOR]

Object tracking is an essential aspect of environmental perception technology for autonomous vehicles. The existing object tracking algorithms can only be applied well to simple scenes. When the scenes become complex, the algorithms have poor tracking performance and insufficient robustness, and the problems of tracking drift and object loss are prone to occur. Therefore, a robust object tracking algorithm for autonomous vehicles in complex scenes is proposed. Firstly, we study the Siam-FC network and related algorithms, and analyze the problems that need to be addressed in object tracking. Secondly, the construction of a double-template Siamese network model based on multi-feature fusion is described, as is the use of the improved MobileNet V2 as the feature extraction backbone network, and the attention mechanism and template online update mechanism are introduced. Finally, relevant experiments were carried out based on public datasets and actual driving videos, with the aim of fully testing the tracking performance of the proposed algorithm on different objects in a variety of complex scenes. The results showed that, compared with other algorithms, the proposed algorithm had high tracking accuracy and speed, demonstrated stronger robustness and anti-interference abilities, and could still accurately track the object in real time without the introduction of complex structures. This algorithm can be effectively applied in intelligent vehicle driving assistance, and it will help to promote the further development and improvement of computer vision technology in the field of environmental perception. [ABSTRACT FROM AUTHOR]

An accurate state-of-charge (SOC) can not only provide a safe and reliable guarantee for the entirety of equipment but also extend the service life of the battery pack. Given that the chemical reaction inside the lithium-ion battery is a highly nonlinear dynamic system, obtaining an accurate SOC for the battery management system is very challenging. This paper proposed a gated recurrent unit recurrent neural network model with activation function layers (GRU-ATL) to estimate battery SOC. The model used deep learning technology to establish the nonlinear relationship between current, voltage, and temperature measurement signals and battery SOC. Then the online SOC estimation was carried out on different testing sets using the trained model. The experiments in this paper showed that the GRU-ATL network model could realize online SOC estimation under different working conditions without relying on an accurate battery model. Compared with the gated recurrent unit recurrent neural (GRU) network model and long short-term memory (LSTM) network model, the GRU-ATL network model had more stable and accurate SOC prediction performance. When the measurement data contained noise, the experimental results showed that the SOC prediction accuracy of GRU-ATL model was 0.1–0.4% higher than the GRU model and 0.3–0.7% higher than the LSTM model. The mean absolute error (MAE) of SOC predicted by the GRU-ATL model was stable in the range of 0.7–1.4%, and root mean square error (RMSE) was stable between 1.2–1.9%. The model still had high prediction accuracy and robustness, which could meet the SOC estimation in complex vehicle working conditions. [ABSTRACT FROM AUTHOR]

Depth estimation of a single image presents a classic problem for computer vision, and is important for the 3D reconstruction of scenes, augmented reality, and object detection. At present, most researchers are beginning to focus on unsupervised monocular depth estimation. This paper proposes solutions to the current depth estimation problem. These solutions include a monocular depth estimation method based on uncertainty analysis, which solves the problem in which a neural network has strong expressive ability but cannot evaluate the reliability of an output result. In addition, this paper proposes a photometric loss function based on the Retinex algorithm, which solves the problem of pulling around pixels due to the presence of moving objects. We objectively compare our method to current mainstream monocular depth estimation methods and obtain satisfactory results. [ABSTRACT FROM AUTHOR]

Vehicle detection is an indispensable part of environmental perception technology for smart cars. Aiming at the issues that conventional vehicle detection can be easily restricted by environmental conditions and cannot have accuracy and real-time performance, this article proposes a front vehicle detection algorithm for smart car based on improved SSD model. Single shot multibox detector (SSD) is one of the current mainstream object detection frameworks based on deep learning. This work first briefly introduces the SSD network model and analyzes and summarizes its problems and shortcomings in vehicle detection. Then, targeted improvements are performed to the SSD network model, including major advancements to the basic structure of the SSD model, the use of weighted mask in network training, and enhancement to the loss function. Finally, vehicle detection experiments are carried out on the basis of the KITTI vision benchmark suite and self-made vehicle dataset to observe the algorithm performance in different complicated environments and weather conditions. The test results based on the KITTI dataset show that the mAP value reaches 92.18%, and the average processing time per frame is 15 ms. Compared with the existing deep learning-based detection methods, the proposed algorithm can obtain accuracy and real-time performance simultaneously. Meanwhile, the algorithm has excellent robustness and environmental adaptability for complicated traffic environments and anti-jamming capabilities for bad weather conditions. These factors are of great significance to ensure the accurate and efficient operation of smart cars in real traffic scenarios and are beneficial to vastly reduce the incidence of traffic accidents and fully protect people's lives and property. [ABSTRACT FROM AUTHOR]

This article summarises the development and experience of the Formula Student race car engine from 2018. According to the technical rules of Formula Student after the change in 2017, this engine adopts a new design concept, employs a 690-mL single-cylinder engine as the base, and applies 'response enhancement technology' with supercharging as the core to achieve a high-power output, a wide high-torque range and an excellent response capability. During the development, various studies on the dynamic performance of the vehicle and the engine were conducted, including vehicle dynamics analysis and track simulation, parameter matching of the supercharger and the engine, control strategy design, and the intake and exhaust system design. This research builds a supercharger air flow and efficiency test bench and an engine performance test bench. Test results show that the developed engine can output 122% of the original power and 120% of the original torque with a 20-mm diameter intake restrictor. Compared with previous generation race cars with a turbocharged four-cylinder engine, the new race car's 0–100 km/h acceleration time is shortened by 0.2 s, the torque response time under typical condition is shortened by 80%, and the lap time of the integrated circuit is reduced by 7%. [ABSTRACT FROM AUTHOR]

Pedestrian detection is an important aspect of the development of intelligent vehicles. To address problems in which traditional pedestrian detection is susceptible to environmental factors and are unable to meet the requirements of accuracy in real time, this study proposes a pedestrian detection algorithm for intelligent vehicles in complex scenarios. YOLOv3 is one of the deep learning-based object detection algorithms with good performance at present. In this article, the basic principle of YOLOv3 is elaborated and analyzed firstly to determine its limitations in pedestrian detection. Then, on the basis of the original YOLOv3 network model, many improvements are made, including modifying grid cell size, adopting improved k-means clustering algorithm, improving multi-scale bounding box prediction based on receptive field, and using Soft-NMS algorithm. Finally, based on INRIA person and PASCAL VOC 2012 datasets, pedestrian detection experiments are conducted to test the performance of the algorithm in various complex scenarios. The experimental results show that the mean Average Precision (mAP) value reaches 90.42%, and the average processing time of each frame is 9.6 ms. Compared with other detection algorithms, the proposed algorithm exhibits accuracy and real-time performance together, good robustness and anti-interference ability in complex scenarios, strong generalization ability, high network stability, and detection accuracy and detection speed have been markedly improved. Such improvements are significant in protecting the road safety of pedestrians and reducing traffic accidents, and are conducive to ensuring the steady development of the technological level of intelligent vehicle driving assistance. [ABSTRACT FROM AUTHOR]

Traffic sign detection and recognition are crucial in the development of intelligent vehicles. An improved traffic sign detection and recognition algorithm for intelligent vehicles is proposed to address problems such as how easily affected traditional traffic sign detection is by the environment, and poor real-time performance of deep learning-based methodologies for traffic sign recognition. Firstly, the HSV color space is used for spatial threshold segmentation, and traffic signs are effectively detected based on the shape features. Secondly, the model is considerably improved on the basis of the classical LeNet-5 convolutional neural network model by using Gabor kernel as the initial convolutional kernel, adding the batch normalization processing after the pooling layer and selecting Adam method as the optimizer algorithm. Finally, the traffic sign classification and recognition experiments are conducted based on the German Traffic Sign Recognition Benchmark. The favorable prediction and accurate recognition of traffic signs are achieved through the continuous training and testing of the network model. Experimental results show that the accurate recognition rate of traffic signs reaches 99.75%, and the average processing time per frame is 5.4 ms. Compared with other algorithms, the proposed algorithm has remarkable accuracy and real-time performance, strong generalization ability and high training efficiency. The accurate recognition rate and average processing time are markedly improved. This improvement is of considerable importance to reduce the accident rate and enhance the road traffic safety situation, providing a strong technical guarantee for the steady development of intelligent vehicle driving assistance. [ABSTRACT FROM AUTHOR]

To optimize the hydrogen consumption, lifespan of the fuel cell (FC), and durability of the lithium-ion battery in fuel cell electric vehicles (FCEVs), energy management strategies (EMSs) have been developed and implemented in the vehicle control units (VCUs) by engineers. Recently, with the rapid development of artificial intelligence, reinforcement learning (RL)-based EMS have shown great promise among various strategies. However, RL-based EMS tends to exhibit relatively poorer performance when faced with novel environments owing to the requirement for learning. This study proposes an enhanced EMS, Scalable Learning in Novel Environment (SLNE)-based EMS. This EMS performs well in unknown and dynamic environments by integrating a memory library (ML) composed of the Dirichlet process (DRP) clustering algorithm combined with the Chinese restaurant process (CRP) using the expectation-maximization (EM) algorithm for updates and the deep Q-network (DQN). Simulation and comparison are conducted under two novel driving cycles to evaluate the performance of the dynamic programming (DP)-based EMS, rule-based EMS, DQN-based EMS, and SLNE-based EMS. By comparing reward, power, and degradation, the results indicate that the proposed SLNE-based EMS demonstrates excellent convergence and adaptability when facing a novel and dynamic environment. Additionally, the proposed SLNE-based EMS achieves approximately 5% improvement in fuel economy, approximately 4.5% reduction in fuel cell degradation rate, and improved durability of the lithium-ion battery, compared with DQN-based EMS. [Display omitted] • The scalable learning in novel environment-based EMS is proposed for an FCEV. • The SLNE-based EMS performs well from the start when facing novel driving cycles. • The SLNE-based EMS demonstrates a higher reward ceiling and lower fluctuations. • Considering the fuel cell degradation, proposed EMS can better enhance the lifespan. • The fuel consumption of the SLNE-based EMS is 5% lower than that of the DQN-based EMS. [ABSTRACT FROM AUTHOR]

Lane detection is an important foundation in the development of intelligent vehicles. To address problems such as low detection accuracy of traditional methods and poor real-time performance of deep learning-based methodologies, a lane detection algorithm for intelligent vehicles in complex road conditions and dynamic environments was proposed. Firstly, converting the distorted image and using the superposition threshold algorithm for edge detection, an aerial view of the lane was obtained via region of interest extraction and inverse perspective transformation. Secondly, the random sample consensus algorithm was adopted to fit the curves of lane lines based on the third-order B-spline curve model, and fitting evaluation and curvature radius calculation were then carried out on the curve. Lastly, by using the road driving video under complex road conditions and the Tusimple dataset, simulation test experiments for lane detection algorithm were performed. The experimental results show that the average detection accuracy based on road driving video reached 98.49%, and the average processing time reached 21.5 ms. The average detection accuracy based on the Tusimple dataset reached 98.42%, and the average processing time reached 22.2 ms. Compared with traditional methods and deep learning-based methodologies, this lane detection algorithm had excellent accuracy and real-time performance, a high detection efficiency and a strong anti-interference ability. The accurate recognition rate and average processing time were significantly improved. The proposed algorithm is crucial in promoting the technological level of intelligent vehicle driving assistance and conducive to the further improvement of the driving safety of intelligent vehicles. [ABSTRACT FROM AUTHOR]

To address the problems of low efficiency and high fuel consumption during the cold start and warm-up processes of internal combustion engines, a series hybrid electric vehicle was selected as the research object and two optimal control strategies were designed. A bench test was performed to determine the following: (a) the influence of engine coolant temperature on effective thermal efficiency; and (b) the relationship between engine operating conditions and coolant temperature increase rate. On the basis of the test results, two sets of warm-up process optimization control strategies were designed using a dynamic programming method and a fuzzy control method based on equivalent consumption minimization strategy (ECMS). The test results show that the fuzzy control method for the coolant temperature can effectively shorten the time required to warm up the engine, and the energy consumption of warm-up process can be reduced by nearly 10% through the dynamic programming method. [ABSTRACT FROM AUTHOR]

As the core technology of hybrid electric vehicles (HEVs), energy management strategy directly affects the fuel consumption of vehicles. This research proposes a novel reinforcement learning (RL)-based algorithm for energy management strategy of HEVs. Hierarchical structure is used in deep Q-learning algorithm (DQL-H) to get the optimal solution of energy management. Through this new RL method, we not only solve the problem of sparse reward in training process, but also achieve the optimal power distribution. In addition, as a kind of hierarchical algorithm, DQL-H can change the way of exploration of the vehicle environment and make it more effective. The experimental results show that the proposed DQL-H method realizes better training efficiency and lower fuel consumption, compared to other RL-based ones. • DQL-H trains each level independently and is more efficient than counterparts. • Sparse rewards can be overcome during the training process. • Substantial rewards can accelerate the speed of convergence. • DQL-H changes the way of exploring the vehicle environment. [ABSTRACT FROM AUTHOR]

Reinforcement learning is a new research hotspot in the energy management strategy. At present, some problems in the application of reinforcement learning to energy management control still exist, including sparse reward, convergence speed, generalization ability, etc. This paper proposes a self-supervised reinforcement learning method based on a Deep Q-learning approach for fuel-saving optimization of a plug-in hybrid electric vehicle (PHEV). First, a detailed vehicle powertrain model of the Prius is built. Second, we use the self-supervised learning model to enrich the reward function. The reward function consists of two parts: internal and external rewards. Finally, to prevent the self-supervised model from falling into the "self-good" situation, a reinforcement learning calibration method is proposed. The vehicle exploration method is more effective because of the enrichment of the reward function. Furthermore, following the characteristics of self-supervised learning, we have also constructed a new driving cycle to verify the generalization ability. Results show that our proposed deep reinforcement learning method based on self-supervised and learning calibration realizes faster training convergence and lower fuel consumption than the traditional policy, and its fuel economy can reach approximately the global optimum under our new driving cycle. • The method used in this can enhance the vehicle's ability to explore the environment. • Internal rewards were constructed using self-supervised learning methods. • The algorithm has generalization ability in energy management control strategy. • The reward function is corrected with the reinforcement learning calibration. [ABSTRACT FROM AUTHOR]

This study designs an active equilibrium control strategy based on model predictive control (MPC) for series battery packs. To shorten equalisation time and reduce unnecessary energy consumption, bidirectional active equalisation is modelled and analysed, and the model predictive control algorithm is then applied to the established state space equation. The optimisation problem that minimises the equilibrium time is transformed to a linear programming form in each cycle. By solving the linear programming problem online, a group of control optimal solutions is found and the series equalisation problem is decoupled. The equalisation time is shortened by dynamically adjusting the equalisation current. Simulation results show that the MPC algorithm can avoid unnecessary energy transfer and shorten equalisation time. The bench experimental result shows that the equilibrium time is reduced by 31%, verifying the rationality of the MPC strategy. [ABSTRACT FROM AUTHOR]