This paper analyzed the technical layout and key research and development direction of skateboard chassis technologies with patent analysis method from aspects of patent disclosure situation, patent technology composition, sorted out the development path of key technical points such as integrated body battery pack, wire controlled chassis and intelligent thermal management system, analyzed the technology development status and future development trend of skateboard chassis technologies from macro and micro perspective.

The difficulty in accurately predicting the future driving conditions of new energy vehicles leads to inaccurate estimation of the battery’s State Of Charge (SOC), to address this issue, this paper proposed a real-time online prediction of the future driving conditions of electric vehicles based on digital map real-time traffic information and machine learning algorithms, experimental verification was conducted on an Internet-distributed vehicle-in-the-loop simulation platform. The results show the proposed algorithm has high real-time performance and prediction accuracy.

In order to explore the practical effects of Cellular Vehicle-to-Everything (C-V2X) in specific application scenarios, the design and experimentation of in-vehicle traffic light application scenarios was conducted based on C-V2X. A real road environment was constructed for in-vehicle traffic light display, red light warning scenario, road test was conducted at different communication distance and vehicle speed conditions using two communication modes based on Long Term Evolution Vehicle-to-Everything (LTE-V2X) and the 4th Generation mobile communication technology (4G). The experimental results show that both scenarios were effectively triggered, with communication delays of less than 1 s, a packet loss rate of less than 2% and a message accuracy rate of no less than 99%, meeting the basic performance requirements of the scenarios. In addition, LTE-V communication has the advantages of long-distance, low latency and high reliability.

To improve the handling stability of distributed drive electric vehicles, this paper proposed an integrated control strategy for active front wheel steering and direct yaw torque based on model predictive control and phase plane method. The strategy consists of a vehicle stability judgment module based on the centroid sideslip angle phase plane, a stability controller based on improved model predictive control and a torque divider. Through the evaluation of vehicle stability on the phase plane, the vehicle stability formula was introduced into the objective function design of the model predictive controller, the active angle and additional yaw moment of the decision were adjusted. Considering vehicle stability and dynamics, a torque synthesis optimization allocation method based on phase plane was designed to allocate torque. Simulation under low adhesion and double lane shifting conditions shows that the control strategy improves the handling stability of the vehicle under medium and low adhesion conditions.

In order to improve the integration of free piston expander-linear generator, the motion of air in cylinder and the motion characteristics of piston were analyzed by using computational fluid dynamics method and dynamic grid technique, and the influence of cylinder structural parameters on the motion characteristics of piston was analyzed by orthogonal test. The results show that the air pressure and velocity change obviously when the air flows through the inlet and outlet. Cylinder orifice aperture has the greatest influence on piston velocity and displacement. Intake time, piston mass and cylinder diameter also have different influences on piston displacement and velocity.

A motor noise tracking method based on Empirical Mode Decomposition spectral Kurtosis Reconstruction Find Peaks (EMD-KR-FP) was proposed to track the noise of brushless DC motor. Firstly, the characteristic frequencies of radial electromagnetic force, torque ripple and resonance induced electromagnetic noise were obtained by theoretical calculation. The characteristic frequency set was defined, the time-domain signal of motor noise was decomposed by Empirical Mode Decomposition (EMD) method. Then, according to the spectral Kurtosis theory, the Intrinsic Mode Function (IMF) component was screened for signal reconstruction, the reconstructed signal was processed by Fourier transform. Peak location algorithm was used to extract the peak frequency of several peaks with the largest contribution on the spectrum and compare with the frequency of characteristic frequency set, so as to determine the cause of electromagnetic noise of the motor. Test results show that this method is effective and reduces the workload of traditional order analysis method.

In order to track the regenerative braking power of EV’s hybrid energy storage system accurately, avoid current impact of energy storage elements and bus voltage build-up caused by untimely distribution of braking energy, this paper proposed a dynamic power tracking control based on hybrid energy storage system. The supercapacitor was modeled based on the Thevenin model, the terminal voltage of the supercapacitor was predicted with Kalman filter. On this basis, a loss function was established for power tracking, the supercapacitor current was solved in real time under constraints. Finally, the effectiveness of the proposed dynamic power tracking control strategy was verified by contrast test. The test results show that the proposed dynamic power tracking control can quickly track the braking power and reduce the current impact of energy storage elements.

In order to realize the bidirectional flow of energy between electric vehicles and the power grid in Vehicle to Grid (V2G) mode, this paper analyzed the working principle using the non-isolated two-stage bidirectional AC/DC converter as the research object, a mathematical model was established. The feed forward decoupling double closed-loop, constant current charge and discharge control strategies were formulated. The Space Vector Pulse Width Modulation (SVPWM) technology was utilized to generate the driving signal of the power switch tube, the passive damping LCL filter was designed. Finally, the system simulation model was built in MATLAB/Simulink for system simulation and analysis. The results show that the proposed control strategy can realize the bidirectional power flow between the electric vehicle and the power grid.

For the problem that the autonomous navigation exploration algorithm is easy to fall into the local area, this paper proposed an exploration algorithm combining sampling and deep reinforcement learning. First, the Long-Short-Term Memory (LSTM) network was used locally to obtain the historical pose information of the unmanned vehicle to avoid repeated exploration of the explored area; secondly, the optimal action of the deep reinforcement learning strategy was used to output using deep reinforcement learning and the reward function was designed to encourage the unmanned vehicle to fully explore the unknown area; Finally, the horizontal movement factor of the unmanned vehicle was considered to generate a global exploration path conforming to its current attitude by solving the Asymmetric Travel Salesman Problem (ATSP). In the 2 000 s mine tunnel simulation environment, compared with the Technologies for Autonomous Robot Exploration (TARE) algorithm, the proposed algorithm increased the exploration area by 346.3 m² and reduced the total driving distance by 209.4 m; in the real scene test, the exploration algorithm completed the exploration of the underground garage with an area of 3 444.3 m² and returned to the starting point in 1 014 s and built the environment map.

To ensure the security and privacy of sensitive data in Internet of Vehicle (IoV) environments, this paper proposed a distributed differential privacy data protection scheme combining federated learning and reinforced learning mechanisms. In this scheme, a federated learning architecture was applied to keep data on vehicle nodes or edge devices for learning, enabling data privacy protection, reducing data transmission costs through distributed storage. The Laplace mechanism was employed to achieve differential privacy, the Layer-wise Relevance Propagation (LRP) was used to manage data perturbation, ensuring the privacy and efficiency of model parameter transmissions. Experimental results show that the proposed scheme can achieve approximately 80% global accuracy within 10 rounds of communication, with a maximum of 98%, can complete model aggregation within less communication rounds, achieving a good balance between privacy protection and global data accuracy, and accurately detecting the injection of false noise through the reinforced learning strategy, promoting the intelligence and security levels of IoV.