Search Results

Robust data association is a core problem of visual odometry, where image-to-image correspondences provide constraints for camera pose and map estimation. Current state-of-the-art visual semantic odometry uses local map points semantics, building semantic residuals associated with all classes to realize medium-term tracking of points. Considering the problem of inefficient semantic data associations and redundant semantic observation likelihood model in the visual semantic odometry, we propose a visual odometry, Local Semantic Odometry (LVSO), which is integrated with medium-term semantic constraints based on local nearest neighbor distance model.

For the improvement of the vehicle yaw stability, this paper studies the control problem of the active front steering (AFS) system with actuator input delay. A novel sliding mode predictive control method to handle actuator input delay is proposed for the AFS system. Firstly, considering the nonlinearities of the vehicle system, a linear parameter varying vehicle system model with two-level structure is proposed to capture the vehicle dynamic behaviors. Secondly, to deal with the issues of actuator input delay and system constraints, a novel sliding mode predictive control method is put forward. In the process of controller design, a sliding mode control algorithm is employed for the improvement of the robustness of the control system, and then a model predictive control algorithm is employed to deal with system constraints.

The trajectory planning and the accurate path tracking are the two key technologies to realize the intelligent driving. The research of the steering wheel angle plays an important role in the path tracking. The purpose of this study is to optimize the steering wheel angle input during the automated lane changing. A dynamic programming approach to trajectory planning is proposed in this study, which is expected to not only achieve a quick reaction to the changing driving environment, but also optimize the balance between vehicle performance and driving efficiency. First of all, the lane changing trajectory is planned based on the positive and negative trapezoidal lateral acceleration method. In addition, the multi-objective optimization function is built which includes such indexes: lateral acceleration, lateral acceleration rate, yaw rate, lane changing time and lane changing distance.

The influence of engine cooling fan on the working state of engine cooling system under different driving forms and control strategy is studied, and a simulation model of engine thermal management system of a commercial vehicle is established. The model takes into account the measured performance parameters of the cooling system components, the gear shift logic of the transmission, the effect of vehicle speed on the airflow rate of the radiator, and proposes a modeling method for different cooling fan driving forms. The performance parameters such as engine outlet coolant temperature and corresponding cooling fan speed under different vehicle speeds and engine loads are calculated and analyzed by using the established model. The road measurement test of the engine thermal management system under the same working condition was carried out to read the relevant data from the engine ECU and confirm the reliability of the data.

Electric vehicle battery thermal management based on liquid cooling is the mainstream form of cooling for new energy vehicles. According to energy consumption, the system is divided into active cooling system and passive cooling system. The cooling of battery modules in these two cooling systems is carried out by liquid-cooled plate, which is connected in series in the cooling system. Therefore, the design of the liquid-cooled plate has a great impact on the effect of battery heat dissipation. In this paper, considering the advantages of existing liquid-cooled plates, the author proposed a series-parallel hybrid dc channel liquid-cooled plate structure, taking square lithium iron phosphate battery pack as the research object. Finally, the effects of different inlet flows and temperatures of the liquid-cooled plate on the thermal performance of the liquid-cooled plate were investigated by using single factor analysis.

In the automotive industry, testing the sound absorption coefficient of acoustic materials through reverberation chambers has been widely used. The advantage of this method is that sound waves are incident on the surface of acoustic materials randomly, which is more in line with actual engineering. At present, most of the reverberation chamber design and construction refers to the international standard ISO 354-2003. However, although the design indicators of the reverberation chamber have already met the requirements of the standard ISO 354-2003, there are still some differences between the test results of different reverberation chambers on the same group of samples to be tested, and sometimes the differences are so big they affect the engineering applications. In this paper, the sound absorption coefficients of the same group of samples in different reverberation chambers are tested, and there are some differences in the sound absorption coefficients.

Battery Run-down under the Electric Vehicle Operation (BREVO) model is a model that links the driver’s travel pattern to physics-based battery degradation and powertrain energy consumption models. The model simulates the impacts of charging behavior, charging rate, driving patterns, and multiple energy management modules on battery capacity degradation. This study implements reinforcement learning (RL) to the simplified BREVO model to optimize drivers’ decisions on charging such as charging rate, charging time, and charging capacity needed. This is done by a reward function that considers both the driver’s daily travel demands and the minimization of battery degradation over a year. It shows that using appropriate charger type (No Charge, Level 1, Level 2, direct-current Fast Charge [DCFC], extreme Fast Charging [xFC]) with an appropriate charging time can reduce battery degradation and total charging cost at the end of the year while satisfying driver’s daily travel demand.

To enhance the transient vibration performance of the vehicle at key on and key off, a method for optimizing mount parameters of a powertrain mounting system (PMS) is proposed. Uncertainties of mount parameters widely exist in a PMS, so a method for optimizing mount parameters of a PMS, which treats the mount parameters of a PMS as uncertain, is also proposed in this paper. Firstly, a 13 degrees of freedom (DOFs) model including car body with 3 DOFs, a PMS with 6 DOFs and unsprung mass with 4 DOFs is established, and the acceleration of the active side of mounts is calculated. An experiment is carried out to measure the accelerations located at active and passive sides of each mount and the accelerations of seat track. A comparison is made between the measured and estimated accelerations, and the proposed model is validated. Two optimization methods for the PMS are proposed based on the developed 13 DOFs model.

The distributed parameter method is used to establish the dynamic simulation model of the electric vehicle thermal management system and various parts, and the finite difference method is used to solve the model. A thermal management system model with same structure is established by AMESIM, and the accuracy of the dynamic simulation model is verified by comparing the deviation of the calculation result between this dynamic simulation model and AMESIM. Based on the established model, the influence of expansion valve opening on the temperature of battery pack and the influence on the heating comfort of the cabin were studied. A control strategy for the rapid cooling of the battery pack was proposed. The results show that the model established by the distributed parameter method provides quite well agreement with commercial equivalent software and can well reflect the flow state of the refrigerant in different zones of the same component.

In order to study the influence of engine silicone oil fan clutch on the performances of engine cooling system under different control strategies, a model of engine cooling system for light truck is established. The working characteristics of the silicone oil clutch and the measured performance parameters of the cooling system components are taken into account in our proposed model. Modeling methods for different silicone oil fan control strategies are also given. Using the established model, the performance parameters under different vehicle speeds, such as coolant temperature of engine outlet and power consumption of cooling fan, are calculated and analyzed. The in-suite measurement of the engine cooling system is carried out to get the temperatures of engine coolant inlet and outlet from engine ECU. The model is validated by the comparison between the calculation and the measured results.

Electronic expansion valve as a throttle element is widely used in heat pump systems and flow characteristics are its most important parameter. The flow characteristics of the electronic expansion valve (EXV) with a valve port diameter of 3mm are studied, when the refrigerant R134a is used as the working fluid. The main factors affecting the flow characteristics are researched by adopting the orthogonal experiment method and single factor control method, for example, inlet pressure, inlet temperature, outlet pressure and valve opening. The results show that the expansion valve opening degree has the greatest influence on mass flow rate. In view of the complicated phase change of the refrigerant passing through electronic expansion valve, it is difficult to model the flow characteristics accurately.

The characteristics of suspension elastic elements (i.e., spring, damper and anti-roll bar) are directly related to the handling and ride comfort performances, how to tune the characteristics of suspensions' elastic elements is always a big issue in developing the chassis of a vehicle. In this paper, a multi-body dynamics model of a passenger car within MSC.ADAMS® is integrated with iSight FD®, an optimization tool, to carry out a multi-objective optimization for improving the behavior of vehicle handling and ride comfort. The characteristics of suspension elastic elements (i.e., spring, damper and anti-roll bar) are considered as design variables. For handling, the objectives are defined by the measurements from multi-body dynamics simulation of typical double lane change according to ISO3888 standard. For ride comfort, the frequency-weighted RMS (Root Mean Square) value of vertical acceleration of the front seat rail according to ISO2631 standard is set as the objective.

Current gasoline-gas vapor recovery system is incomplete, for it cannot adjust the vapor-liquid ratio automatically due to the change of working temperature. To solve this problem, this paper intends to design a new system and optimize its parameters. In this research, variables control method is used for tests while linear regression is used for data processing. This new system moves proportion valve away and adds a DSP control module, a frequency conversion device, and a temperature sensor. With this research, it is clearly reviewed that the vapor-liquid ratio should remains 1.0 from 0 °C to 20 °C as its working temperature, be changed into 1.1 from 20 °C to 25 °C, be changed into 1.2 from 25 °C to 30 °C, and be changed into 1.3 when the working temperature is above 30 °C.

The mount bracket is an important part of the mount system, and its dynamic characteristics will affect dynamic characteristics of the mount system, which means it will affect NVH(Noise, Vibration, Harshness) of the vehicle. Based on the large error between the test result and the finite element analysis(FEA) result, the dynamic finite element model of the mount bracket can be modified from the material parameters and the equivalent boundary of the bolt joint. In this paper, a method to identify the parameters of the mount bracket model by combining modal test, FEA, and the mathematical optimization model was presented. Firstly, based on HyperStudy platform, the optimization objective was minimizing the natural frequency error between FEA and free mode test, and the material parameters of the bracket to be identified were used as design variables to build the optimization function. The global response surface method was used for iteration to complete the identification.

Uniaxial fatigue tests of rubber dumbbell specimens under different mean and amplitude of strain are carried out. An Extreme Learning Machine (ELM) model optimized by Dragonfly Algorithm (DA) is proposed to predict the fatigue life of rubber based on measured rubber fatigue life data. Mean and amplitude of strain and measured rubber fatigue life are taken as input variables and output variables respectively in DA-ELM model. For comparison, genetic algorithm (GA) and particle swarm optimization (PSO) are used to optimize ELM parameters, and GA-ELM and PSO-ELM models are established. The comparison results show that DA-ELM model performs better in predicting the fatigue life of rubber with least dispersion. The coefficients of determination for the training set and test set are 99.47% and 99.12%, respectively. In addition, a life prediction model equivalent strain amplitude as damage parameter is introduced to further highlight the superiority of DA-ELM model.

The good mobility and large carrying capacity promote the popularity of intercity coach in mass transit, especially in the long distance passenger transport nowadays. However, accidents related to coach and bus usually involve large casualties. Higher risk of fatalities is exhibited in rollover than the other coach accident types. In order to protect the occupants when a rollover accident occurs, coach structure must have sufficient strength to resist the impact loads. This paper presents a rollover test of an intercity coach body section using both numerical simulation and experimental testing to investigate its rollover crashworthiness in accordance with ECE R66. A full scale coach body section is manufactured and a tilting bench is designed and fabricated. Displacement transducers and accelerometer are equipped to record the time history of superstructure deformation and impact acceleration. And the FE model was developed accordingly.

Because autonomous vehicles (AVs) equipped with active front steering have the features of time varying, uncertainties, high rate of fault, and high burden on the in-vehicle networks, this article studies the adaptive robust control problem for improving lateral stability in steer-by-wire (SBW) vehicles in the presence of abrupt nonlinear faults. First, an upper-level robust H∞ controller is designed to obtain the desired front-wheel steering angle for driving both the yaw rate and the sideslip angle to reach their correct values. Takagi-Sugeno (T-S) fuzzy modeling method, which has shown the extraordinary ability in coping with the issue of nonlinear, is applied to deal with the challenge of the changing longitudinal velocity. The output of the upper controller can be calculated by a parallel distributed compensation (PDC) scheme.

The thermal management system, which is used to improve driving safety and thermal comfort, is one of the most important systems in electric vehicles. In recent years, researchers have coupled the heat pump system and the battery cooling system to effectively improve the heating COP (Coefficient of Performance). Therefore an accurate dynamic model of thermal management system plays a key role in investigating system performance and optimal control strategies. In this paper, an electric vehicle thermal management system based on four-way valve heat pump system is designed. The moving boundary method is improved by considering the unsteady flow of the external fluid, and then a 13-order dynamic model of the thermal management system is established. Firstly, the control equations of evaporator, condenser and chiller are derived according to the principle of conservation, and then a dynamic model of thermal management system is established in Simulink.

Considering the interaction between fan blades and the surrounding air when a cooling fan rotates, the Fluid-Structure Interaction (FSI) model of the fan is established, and flow rate, static pressure, efficiency versus speed of the fan are calculated and analyzed. The aerodynamic performance of the fan is carried out, and the measured performance parameters are compared with calculated to validate the developed model. Using the established model, the performance of fans with different rotating speeds, diameters and blade installation angles is calculated. The effects of fan speed, diameter and blade installation angle on blade deformation and aerodynamic performance are studied.

A dynamic performance calculating model for a powertrain-subframe mounting system (PSMS) is presented. Calculation methods for determining the dynamic displacements of a powertrain center of gravity (CG), the dynamic displacements of a subframe CG and the dynamic reaction forces of each mount in a PSMS under ground and motor shake excitation are developed in this paper. An optimization procedure based on the genetic algorithm and SQP is developed for reducing resonance peaks of the reaction forces at mounts. A generic PSMS with three powertrain-subframe mounts and four subframe-body mounts is used to validate the optimization method. The optimization results demonstrate that the results using the optimization procedure can effectively reduce the reaction forces at mounts.