The present researches on vehicle-mounted photovoltaic systems mainly focus on increasing the installation area of photovoltaic panels by optimizing the folding mechanism to enhance the power generation capacity, while neglecting the issue of the synergistic optimization of power generation capacity and the additional drag energy consumption of the system. To enhance the net power of vehicle-mounted photovoltaic systems, by optimizing the aerodynamic performance of the system, this paper presents a new methodology to reduce the additional drag energy consumption imposed by the system on the vehicle, and thereby to increase the net power of the system. Firstly, a foldable vehicle-mounted photovoltaic system was designated as the object, and a high-transmittance fairing and a tail wing that conform to the aerodynamic principle were designed. Subsequently, three design variables, namely the front tilt angle of the fairing, the back angle, and the system height, were selected to optimize the shape of the fairing. Through the construction of an orthogonal test scheme and the analysis of polar deviation, the influence degree of the three design variables on the system aerodynamic drag was obtained as system height > front tilt angle > back angle. From the analysis of the main effect plot, the three variables are found exhibiting monotonic effects on the aerodynamic drag of the system, thus determining the structural parameters of the fairing shape as follows: a front tilt angle of 70°, a back angle of 0°, and a system height of 100mm. Subsequently, the tail attack angle of the vehicle-mounted photovoltaic system was optimized, a cubic spline interpolation approximation model was constructed based on the experimental data, and the tail attack angle with optimal lift-to-drag ratio was obtained as 33.96°. In addition, the vehicles equipped with the on-board photovoltaic system proposed in this paper were compared with those without the system. It is found that the air resistance coefficient decreases by 44.59%, the aerodynamic resistance decreases by 22.45% and the lift coefficient decreases by 226.15%, and that the direction of the lift undergoes a transformation from upward to downward. This transformation serves to mitigate the adverse impact of upward aerodynamic lift on the handling and safety performance of the entire vehicle. A comparison of the original vehicle model with the modified version reveals a significant decrease of air resistance coefficient of 17.35% and a modest aerodynamic resistance increase of 3.14 N, which means that the modification effectively mitigates the adverse effects of the on-board photovoltaic system on vehicle’s aerodynamic performance. Finally, a comparison and analysis of the net power of the on-board photovoltaic system before and after the optimization was conducted. It is found that the proposed optimization scheme can effectively increase the net power generation of the vehicle-mounted photovoltaic system during vehicle operation. When the vehicle speed is 40.0 m/s, the net power difference reaches 7723.62 W.

In order to realize the online joint estimation of three major states of ternary lithium-ion battery, namely SOC (State of charge), SOH (State of Health) and SOE (State of Energy), and to deal with the open-loop cumulative error caused by various noises in the actual use of electric vehicles, and, furthermore, to improve the stability of online estimation of lithium-ion battery, this paper proposed an online joint estimation method of the three major states of ternary lithium-ion battery in multiple time scales based on double adaptive extended Kalman filter (DAEKF). In the investigation, the state space equation of DAEKF algorithm is derived based on the second-order RC model, and the parameters are identified online by the recursive least square method with forgetting factor (FFRLS). The SOC and SOE of lithium-ion battery are estimated online in the micro time scale, and the SOH of lithium-ion battery is estimated online in the macro time scale. Thus, the online joint estimation of the three major states of lithium-ion battery can be realized. Finally, the proposed method was verified by experiments under different operating conditions of NVR18650B ternary lithium-ion battery. The experimental results show that the proposed method can rapidly converge the model parameters under the two verification conditions; that the estimation errors of SOC and SOE in the micro time scale are kept within 1%, and the estimation errors of SOH in the macro time scale are kept within 1.6%; and that, as compared with the EKF algorithm, the proposed method has a higher estimation accuracy and better estimation convergence and stability.

Traffic signs on the road contain a large amount of semantic information about traffic rules, and rapid and accurate access to this information helps to achieve higher levels of assisted driving functions, thus improving vehicle's safety performance. In view of the traffic signs are susceptible to external factors and the problems of high similarity between categories and small size, this paper made targeted improvements in data augmentation, feature extraction and feature enhancement based on YOLOv5s model. In the data augmentation part, color space transformation and geometric transformation matrix were used to simulate the possible color changes and shape changes of traffic signs in actual scenes, and the Mosaic algorithm and Copy-paste algorithm were used to improve the number of tiny traffic signs in the training set and the richness of the background. In the feature extraction part, a feature extraction module based on channel attention calibration was constructed to improve the model's ability to discriminate similar features. In the feature enhancement part, the number of prediction branches and downsampling multiplier were optimized by fusing shallow features and deep features with a dual-path enhancement structure, so as to increase the detection accuracy of tiny traffic signs. In addition, the K-means++ algorithm was used to cluster the prior bounding box templates and construct the loss function based on the CIoU metric, thus reducing the difficulty of the prior bounding box regression. Experiments on the TT100K and CCTSDB dataset test show that the mAP@0.5 of the proposed model is 88.8% and 83.5% respectively, and the speed of the model is 120.5 f/s and 114.7 f/s respectively. Compared with the existing traffic sign detection models, the proposed model reaches the advanced level in both accuracy and speed. Comparison experiments for data augmentation algorithms, prediction branches, and channel attention module positions further demonstrate the effectiveness of the proposed specific optimization methods.

With the development of autonomous driving technology, deep reinforcement learning has become an important means to realize the efficient driving policy learning. However, the implementation of autonomous driving is faced with the challenges brought by the complex and changeable traffic scenes, and the existing deep reinforcement learning methods have the problems of single scene adaptation ability and slow convergence speed. To address these issues and to improve the scene adaptability and policy learning efficiency of autonomous vehicles, this paper proposed a multi-task assisted driving policy learning method. Firstly, this method constructed the encoder-multi-task decoder module based on the deep residual network, squeezing high-dimensional driving scenes into low-dimensional representations, and adopted multi-task-assisted learning of semantic segmentation, depth estimation and speed prediction to improve the scene information richness of low-dimensional representations. Then, the low-dimensional representation was used as the state input to build a decision network based on reinforcement learning, and the multi-constraint reward function was designed to guide the learning of driving strategies. Finally, simulation experiments were conducted in CARLA. The experimental results show that, compared to classic methods such as DDPG and TD3, the proposed method improves the training process through multi-task assistance and learns better driving policies. It achieves higher task success rates and driving scores in several typical urban driving scenarios such as roundabouts and intersections, demonstrating excellent decision-making capabilities and scene adaptability.

Single-stage point cloud 3-dimensional object detection algorithms based on pillars have gained significant attention and widespread application in the industry due to their high operational efficiency. However, the loss of fine-grained information loss in 3-dimensional features of point clouds caused by pillar-based quantization results in weaker detection capabilities for small objects in sparse point clouds. Although some studies have proposed solutions to this problem, they often come at the cost of either greater detection time or compromised detection accuracy for large targets. For this reason, this paper proposed an enhanced pillar-based point cloud object detection algorithm with enhanced pillar feature encoding. Firstly, a pillar feature encoding network is constructed to combine local and global features of point clouds within pillar cells, enhancing the representation capability of pillar-quantized features. Then, a backbone network that combines 2-dimensional sparse convolutional blocks with a feature fusion network was designed to fuse multi-scale high-level abstract semantic features and low-level fine-grained spatial features, preventing excessive focus on small-size features and thus degrading the detection performance for large targets. Lastly, the model was trained and tested on the KITTI autonomous driving dataset, with experimental results visualized and ablation studies conducted. The results show that, the proposed algorithm, under the medium difficulty level of the KITTI dataset, has an average precision mean of 63.54% across multiple categories, an average orientation similarity mean of 70.72%, and an average detection frame rate of 31.5 f/s. Compared with the PointPillars, TANet, and PiFEnet, the average precision mean of the algorithm proposed in this paper has increased by 2.44, 2.05, and 2.38 percentage points respectively, and the average orientation similarity mean has increased by 4.69, 0.68, and 7.83 percentage points respectively, demonstrating potential for engineering applications in comparisons with similar algorithms.