Search Results

In the process of truck overload and over-limit detection, it is necessary to detect the characteristics of the vehicle's size, type, and wheel number. In addition, in some vehicle vision-based load recognition systems, the vehicle load can be calculated by detecting the vibration frequency of specific parts of the vehicle or the change in the length of the suspension during the vehicle's forward process. Therefore, it is essential to quickly and accurately identify vehicle features through the camera. This paper proposes a vehicle feature recognition method based on image semantic segmentation and Python, which can identify the length, height, number of wheels and vibration frequency at specific parts of the vehicle based on the vehicle driving video captured by the roadside camera.

It’s not easy to start the engine in winter, especially in frigid highlands, because the low temperature increases the fuel’s viscosity, decreasing the lubricating oil flow ability and the storage performance of battery. Current electrical heating method can improve the engine starting performance in low temperature condition, but this method adds an external power to the engine, leading to the engine cannot maintain an efficient energy utilization. A warming device using the solar energy is designed to conserve the energy during the daytime, and directly warm up the engine at the time when the engine turns off for a long time, especially during the night. A solar collector installed on the top of the vehicle is used to convert the solar energy to the thermal energy, which is then transferred to the heat accumulator that contain the phase-change medium which can increase the heat storage performance.

The mountain roads are curved and complicated, with undulating terrain and large distance between charging stations. Compared with traditional powered vehicles, in addition to safety issues, pure electric vehicles also need to deal with the driving range issue. At present, the relevant researches on automobile driving in mountainous areas mainly focus on the driving safety of traditional fuel oil vehicles when going uphill and downhill, while there are few researches on the driving planning of pure electric commercial vehicles on curved slope road. This paper presents a speed planning method for pure electric commercial vehicles based on vehicle-road collaboration technology. First, establish the vehicle dynamics model, analyze the vehicle dynamics characteristics when passing the downhill curve, calculate the safe speed range of the vehicle when passing the downhill curve, and establish the safe speed model of the downhill curve.

This research is based on the Controller Area Network (CAN) bus, and briefly analyzed its communication protocol with reference to the layered model of Open System Interconnect Reference Model (OSI). Subsequently, a data acquisition system was designed and developed including a Vehicle Communication Interface (VCI) and a laptop. After the overall architecture was built, the communication mechanism of the VCI was studied. Furthermore, the lap top app was built using the layered design followed by the implementation of a scheme for data collection and experimentation involving the test driving of a real car on road. Finally, the driving style was identified by means of fuzzy reasoning and solving ambiguity based on fuzzy theory; via training the acceleration sample and forecast using the excellent learning and generalization ability of Support Vector Machine (SVM) for high-dimensional, finite samples.

The development of electrification is widely considered to be the key to the transportation industry. In recent years, the number of electric trucks on the road is increasing year by year, so the safety of electric trucks is of great importance. At present, the batteries of electric trucks are mostly arranged on the two sides of the trucks. The protective devices are only the guardrails fixed on the vehicle body, so the protective effect is poor. In view of this situation, this paper designed a battery protection device for electric truck. When the truck is hit in a side collision, the transverse guardrail first generates plastic deformation and absorbs kinetic energy. At this time, the collision force is transmitted to the energy-absorbing box along the moving direction of the side collision vehicle, and the energy-absorbing box is contracted to produce a buffer effect.

In the Plug-in hybrid electric bus, the power train parameter and control strategy significantly affect the economy and dynamic performance. Thus, the simultaneous optimization of power train parameter and control strategy is designed for the trade-off between the dynamic and economic performance. Depending on the parallel electric auxiliary control strategy in a plug-in hybrid electric bus, a vehicle dynamic simulation model is built with the software AVL Cruise. Aiming at the minimization of equivalent gas consumption and acceleration time from 0 to 50kmph, the gear ratio, final drive ratio, gear shifting strategy and control strategy are chosen as optimal variable, which significantly impact power performance and fuel economy. The driving performance and the driving range with full battery are considered as constraints. Based on the software Isight, multi-objective optimization model is built by adopting non-dominated sorting genetic optimization algorithm (NSGA-II).

This paper presents a novel application of hydraulic electromagnetic energy-regenerative shock absorber (HESA) into commercial vehicle suspension system and vehicle road performance are simulated by the evaluating indexes (e.g. root-mean-square values of vertical acceleration of sprung mass, dynamic tire-ground contact force, suspension deflection and harvested power; maximum values of pitch angle and roll angle). Firstly, the configuration and working principle of HESA are introduced. Then, the damping characteristics of HESA and the seven-degrees-of-freedom vehicle dynamics were modeled respectively before deriving the dynamic characteristics of a vehicle equipped with HESA. The control current is fixed at 7A to match the similar damping effect of traditional damper on the basis of energy conversion method of nonlinear shock absorber.

The tank truck has a wide range of application. When the liquid in the tank is not fully loaded, the lateral movement of the liquid in the tank will shift the center of gravity of the tank truck and make the vehicle less safe. It is easy to roll over when the tank truck is turning. This study combines the vehicle dynamic characteristics and geographic information, which gives the driver safe speed and safe braking distance tips before turning, to reduce the traffic accidents caused by driver's misjudgment. The dynamic model of the tank truck is established, through collecting the real-time information of the vehicle, the vehicle load and braking torque are calculated by the relevant dynamic model. The system needs to measure the deviation of the center of gravity in the tank truck movement process, and the deviation of the center of gravity has a great influence on the safety speed.

Autonomous vehicles require the collaborative operation of multiple modules during their journey, and enhancing tracking performance is a key focus in the field of planning and control. To address this challenge, we propose a cooperative control strategy, which is designed based on the integration of model predictive control (MPC) and a dual proportional–integral–derivative approach, referred to as collaborative control of MPC and double PID (CMDP for short in this article).The CMDP controller accomplishes the execution of actions based on information from perception and planning modules. For lateral control, the MPC algorithm is employed, transforming the MPC’s optimal problem into a standard quadratic programming problem. Simultaneously, a fuzzy control is designed to achieve adaptive changes in the constraint values for steering angles.

With the development of highway transportation and automobile industry technology, highway truck overload phenomenon occurs frequently, which poses a danger to road safety and personnel life safety. So it is very important to identify the overload phenomenon. Traditionally, static detection is adopted for overload identification, which has low efficiency. Aiming at this phenomenon, a dynamic overload identification method is proposed. Firstly, the coupled road excitation model of vehicle speed and speed bump is established, and then the 4-DOF vehicle model of half car is established. At the same time, considering that the double input vibration of the front and rear wheels will be coupled when vehicle passes through the speed bump, the model is decoupled. Then, the vertical trajectory of the body in the front axle position is obtained by Carsim software simulation.

The phenomenon of liquid transfer in the liquid tank of the semi-trailer vehicle for transporting dangerous cargo (SVTDC) during braking is analyzed and the relevant mathematical model is established. The braking dynamic model of the SVTDC considering the liquid sloshing in the tank is established, and the model is verified based on the co-simulation method. Based on the typical conditions, the braking deceleration and axle load calculation functions of the model are simulated and analyzed, and the application prospect of the model in the development of driving automation control strategy is discussed.

The non-contact overload recognition method refers to the method of detecting the vibration state of the vehicle through visual recognition without touching the vehicle, and then calculating the vehicle load in combination with the vehicle dynamics model to determine whether the passing vehicle is overloaded. Due to the convenience of detection, low cost of infrastructure and informatization, this method has great advantages in the field of overload identification. However, the model used in this recognition method is the single mass vibration model at present, which will have a large error due to the interaction between the front and rear suspension, and the position of the center of mass needs to be acquired in the recognition process, which is difficult in the actual identification process. In this paper, a vehicle vibration model containing two modes of vibration is proposed, and uses Sobol algorithm to analyze the parameter sensitivity of the model.

The mountainous roads are rugged and complex, so that the driver can not make accurate judgments on dangerous road conditions. In addition, most heavy vehicles have characteristics of large weight and high center of gravity. The two factors above have caused most of the car accidents in mountain areas. A research shows that 90% of car accidents can be avoided if drivers can respond within 2-3 seconds before the accidents happen. This paper proposes a speed warning scheme for heavy-duty vehicle over the horizon in mountainous area, which can give the drivers enough time to respond to the danger. In the early warning aspect, this system combines the front road information, the vehicle characteristics and real-time information obtained from the vehicle, calculates and forecasts the danger that may happen over the horizon ahead of time, and prompts the driver to control the vehicle speed.

The heavy-duty vehicles have large transportation capacity. Gross mass and center of gravity position of the heavy-duty vehicles vary with the cargo mass and the driving condition, which affect driving safety and handling stability. Gross mass and center of gravity position of the vehicles are usually measured on fixed test platform, and the vehicles are stationary or pass the platform slowly in the measurement process. Most dynamic weighing system could not measure the center of gravity position of the vehicles. On-board mass and center of gravity of motor vehicles measurement system mainly based on the tire pressure information could measure gross mass and center of gravity position accurately in the driving process. The measurement errors of the sensors are effectively decreased by filtering collected sensor data. The relationship between the tire pressure and the tire load is built when the vehicle is stationary.

Dual-Clutch-Transmission (DCT) is one kind new automatic transmission which has double clutch structure. The most important unit of DCT is Transmission-Control-Module (TCM).In the development process of TCM, simulation is an important research tools. We have analyzed the DCT principle of work, established its mathematical model, created the charge and discharge oil models of typical wet dual clutch transmission, established the control logic to unify and separate double clutch in turn, and also designed out the shift control using fuzzy control using MATLAB/Simulink software. Utilizing engine model, driver model, the DCT model, the TCM model, the vehicle model, established the vehicle simulation model, and implemented simulation; Result indicated that, the established model can correctly reflect the torque and speed change when shifted gears and can correctly realize the automatic shift gears.

In this paper a new pressure control method of a modified accumulator-type Electro-hydraulic Braking System (EHB) is proposed. The system is composed of a hydraulic motor pump, an accumulator, an integrated master cylinder, a pedal feel simulator, valves and pipelines. Two pressurizing modes are switched between by-motor and by-accumulator to adapt different pressure boost demands. A differentiator filtering raw sensor signal and calculating pedal speed is designed. By using the pedal feel simulator, the relationship between wheel pressures and brake force is decoupled. The relationships among pedal displacement, pedal force and wheel pressure are calibrated by experiments. A model-based PI controller with predictor is designed to lower the influences caused by delay. Moreover, a self-tuning regulator is introduced to deal with the parameter’s time-varying caused by temperature, brake pads wearing and delay variation.

Accurate surrounding vehicle motion prediction is critical for enabling safe, high quality autonomous driving decision-making and motion planning. Aiming at the problem that the current deep learning-based trajectory prediction methods are not accurate and effective for extracting the interaction between vehicles and the road environment information, we design a target vehicle intention-aware dual attention network (IDAN), which establishes a multi-task learning framework combining intention network and trajectory prediction network, imposing dual constraints. The intention network generates an intention encoding representing the driver’s intention information. It inputs it into the attention module of the trajectory prediction network to assist the trajectory prediction network to achieve better prediction accuracy.

In heavy truck driveline system, the components often include clutch, transmission, transfer case, drive shaft, etc. A fluid torque converter could be equipped in front of the transmission in order to improve the starting performance. Meanwhile, a hydraulic retarder could be introduced for auxiliary braking so as to adapt the truck to the brake on long downgrade in mountainous regions. Thus, the driveline heat load would have a notable increase. Both the fluid torque converter and the hydraulic retarder would produce a large quantity of heat, and a special cooling system is needed for adjusting the transmission fluid temperature with which the gains are potentially very large [1]. The heat load for driveline is often calculated based on empirical formula. For the heavy truck, however, if the heat value is underestimated, driveline components would suffer from overheated damage.

Semi-active suspension has been widely applied in commercial vehicle suspension in order to get good riding comfortableness. Fuzzy logic control (FLC) has been widely applied in the field of kinetic control because control rule of FLC is easy to understand. But the gain of fuzzy rules and adjustment of membership functions usually depend on experts' experiences and repeated experiments, thus the fuzzy rules and membership functions has strong subjectivity, also are easily affected by environment of experiments, so the main problem of fuzzy logic controller design is selection and optimization of fuzzy rules and membership functions. Genetic Algorithms (GA) is the algorithm that searches the optimal solution through simulating natural evolutionary process and is one of the evolution algorithms which have most extensive impact.

For intelligent vehicles, a robust perception system relies on training datasets with a large variety of scenes. The architecture of federated learning allows for efficient collaborative model iteration while ensuring privacy and security by leveraging data from multiple parties. However, the local data from different participants is often not independent and identically distributed, significantly affecting the training effectiveness of autonomous driving perception models in the context of federated learning. Unlike the well-studied issues of label distribution discrepancies in previous work, we focus on the challenges posed by scene heterogeneity in the context of federated learning for intelligent vehicles and the inadequacy of a single scene for training multi-task perception models. In this paper, we propose a federated learning-based perception model training system.