In order to improve the shortcomings of poor smoothness and potential collision in traditional Rapidly-exploring Random Tree (RRT) algorithm for global path planning, the paper proposed a dual-optimization RRT algorithm. Based on the traditional RRT algorithm, an adaptive target bias strategy was introduced to shorten the sampling time, and an angle-constrained sampling strategy was introduced to adapt to the vehicle’s maximum steering angle. After the initial path was obtained, a binary optimization function (reducing path curvature and avoiding obstacles) was established and used as a basis for gradient descent secondary optimization, generating a path that can be driven by vehicles with good smoothness and low collision probability, which was then simulated and verified. The results show that compared with RRT algorithm, RRT-Connect algorithm and RRT* algorithm, the optimized RRT algorithm reduces average curvature by 38.1%, 36.4% and 24.7%, respectively; while reducing curvature variance by 38.4%, 38.4% and 27.2%, respectively.

Based on the historical driving data of trucks in a province, this paper proposed a prediction method of dangerous driving behavior based on Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) network and self-attention mechanism. For the characteristics of large amount of truck driving data, high dimension, difficult feature extraction and strong time sequence, this method first used XGBoost to filter the features, then used CNN to extract spatial features and LSTM to further capture the temporal information of driving behaviors. Finally, dangerous driving behaviors were predicted by self-attention mechanism. Experimental results show that the method presented in this paper performs better than other long time series prediction methods on highway freight driving data in a province, with recognition accuracy reaching 85.05%, the weighted average recall rate reaches 83%, and the F1-score reaches 84%.

In order to further improve the fuel economy of autonomous vehicle driving at intersections, this paper analyzed quantitatively the performance index functions and constraints including vehicle safety, economy, comfort, etc., based on the Model Predictive Control (MPC) theory, and an ecological driving controller of autonomous vehicle at intersections was designed. The simulation results show that the control strategy proposed in the paper can ensure good safety and comfort, reduce fuel consumption by 15.83% and 34.98% respectively compared with Linear Quadratic Regulator (LQR) under conditions with/without vehicle ahead affecting vehicle driving.

For the fact that single control algorithm cannot simultaneously meet the requirements of autonomous vehicles for path tracking accuracy and controller solving speed, this paper proposed a hybrid control strategy based on Linear Quadratic Regulator (LQR) and Model Predictive Control (MPC). The strategy used an LQR in the low-speed condition and an MPC algorithm in the high-speed condition, on the basis of which a switching mechanism of the control algorithm based on a Finite State Machine (FSM) was designed and the control parameters were optimized by Genetic Algorithm (GA). The hybrid control strategy was simulated and verified based on CarSim and MATLAB/Simulink simulation platforms, and the real vehicle test was further completed. The experimental results show that the designed hybrid control strategy can reduce the computation time on the basis of improving the tracking accuracy, and the average lateral error and average heading error are reduced by 26.3% and 39.6%, respectively, and the average computation time is reduced by 10.9% compared with the single control algorithm.

In order to improve the prediction accuracy of State Of Charge (SOC) of power batteries, this paper proposed a method to improve the traditional ampere-hour intergral method to estimate the SOC of the lithium-ion battery based on a single particle Li diffusion model. The solver programs for estimating SOC with traditional and improved ampere-hour integration method and communication interface programs for battery data acquisition were written in software LabVIEW, which achieved the on-line monitoring of the battery SOC under different environmental temperatures and currents by the two methods. The results show that, under the above discharging conditions, the maximum estimation errors of the improved method and the traditional ampere-hour integral method are 1.11% and 1.89%, respectively. When the discharge current changes dramatically, the battery SOC estimated by the improved method fluctuates less than the traditional ampere-hour integral method.

To improve the collision avoidance performance of the Autonomous Emergency Braking (AEB) system in dangerous braking scenarios, this paper proposed an AEB control strategy considering the braking intention of the front vehicle and its test and evaluation method. A joint platform of PreScan, Simulink, and driving simulator was established to collect drivers’ braking operation data, classify the braking intention of the front car based on K-means clustering method, and use sliding time window to extract intention recognition model to train the dataset. The front car uses a double-layer hidden Markov model to identify the driver’s braking intention, while the main car calculates the critical safety distance threshold based on different braking intentions and performs collision avoidance control. A PreScan+Simulink virtual simulation and test environment was established, and AEB strategy comprehensive evaluation method was proposed based on Analytic Hierarchy Process (AHP). Four typical AEB control models were compared, which verified that the proposed method can timely trigger braking to avoid collisions in different braking scenarios, while reducing driving discomfort caused by premature braking.

In order to improve the NVH performance of the timing housing of a gasoline engine specific for a range extender, several support bosses were added locally to the timing housing, which significantly improves the dynamic stiffness of the timing housing, and at the same time effectively reduces the vibration response and deformation per unit of excitation, so as to reduce the noise of the range extender assembly 1 m from the engine front-end side in the full-load acceleration condition by 4.0 dB(A), the linearity of the noise SPL and the acceleration of the vibration of the timing housing were significantly improved. The validation results show that improving the dynamic stiffness can effectively improve the noise radiation from the structural vibration of thin-walled parts.

Based on the test platform of volume constant combustion bomb, the influence of geometrical parameters of the premixed single-orifice active pre-chamber on the ignition characteristics of the pre-chamber were investigated by the combustion visualization method. During the pressure constant combustion process in the early stage of flame development, the time from the start of ignition until the flame area reaches half of the visible window area of the combustion vessel is defined as the initial flame development time, which is used as a reference index to measure the ignition effect of the active pre-chamber under different geometrical parameters: under the test conditions of different orifice diameters (2.0~4.0 mm), different inner diameters of the pre-chamber channel (3.0~5.5 mm), and different open angles of the lower end (0°~75°), the maximal differences of initial flame development time are 9.3 ms, 6.8 ms, and 2.9 ms, respectively. Therefore, the effect of the three geometrical parameters on the ignition effect of the single-orifice active pre-chamber are, in descending order, the orifice diameter, the inner diameter of the pre-chamber channel and the opening angle of the lower end of the pre-chamber.

To address combustion and emission issues in the lean combustion of gasoline direct injection (GDI) engine, ethanol-doped reformate gas was used to enhance the engine’s performance. The CONVERGE three-dimensional simulation technology was employed to explain the engine’s performance at a microscopic level. The results show that introducing ethanol-doped reformate gas can improve the distribution of equivalence ratio in the cylinder, promote the generation of H radicals and OH radicals, facilitating effective combustion. With an increase in the doping ratio of reforming gas, the distribution range and concentration of HO₂ and H₂O₂ become wider and higher, and the stratification phenomenon in the combustion chamber becomes more pronounced. As the doping ratio of reforming gas increases to 20%, the peak values of in-cylinder soot generation mass and number density are reduced by 90% and 56.25%, respectively, effectively improving the soot emissions of GDI engines.

To address the limitation that the traditional Automatic Emergency Braking(AEB) strategy based on Time To Collision (TTC) does not take into account the speed of the vehicle, this paper proposed a dynamic collision time threshold model that took into account the vehicle speed and an AEB control strategy based on dynamic collision time threshold was designed. In order to ensure the comfort and safety of the braking process, a two-stage braking strategy was determined and the deceleration rate was designed, using PI control to complete the vehicle deceleration control. The TTC thresholds for different vehicle speeds were determined through simulation and a dynamic crash time threshold model was established. Hardware-In-the-Loop (HIL) simulation results show that the AEB strategy designed in this paper improves the collision avoidance success rate by 47.6% compared with the traditional AEB strategy under the premise of ensuring comfort, thus has better comprehensive performance.