Search Results

An autonomous vehicle is a comprehensive intelligent system that includes environment sensing, vehicle localization, path planning and decision-making control, of which environment sensing technology is a prerequisite for realizing autonomous driving. In the early days, vehicles sensed the surrounding environment through sensors such as cameras, radar, and lidar. With the development of 5G technology and the Vehicle-to-everything (V2X), other information from the roadside can also be received by vehicles. Such as traffic jam ahead, construction road occupation, school area, current traffic density, crowd density, etc. Such information can help the autonomous driving system understand the current driving environment more clearly. Vehicles are no longer limited to areas that can be sensed by sensors. Vehicles with different autonomous driving levels have different adaptability to the environment.

Overloading of trucks can easily cause damage to roads, bridges and other transportation facilities, and accelerate the fatigue loss of the vehicles themselves, and accidents are prone to occur under overload conditions. In recent years, various countries have formulated a series of management methods and governance measures for truck overloading. However, the detection method for overload behavior is not efficient and accurate enough. At present, the method of dynamic load identification is not perfect. No matter whether it is the dynamic weight measurement method of reconstructing the road surface or the non-contact dynamic weight measurement method, little attention is paid to the difference of different vehicles. Especially for different vehicles, there should be different load limits, and the current devices are not smart enough.

The vehicle overload seriously jeopardizes traffic safety and affects traffic efficiency. At present, the static weighing station and weigh-in-motion station are both relatively fixed, so the detection efficiency is not high and the traffic efficiency is affected; the on-board dynamic weighing equipment is difficult to be popularized because of the problem of being deliberately damaged or not accepted by the purchaser. This paper proposes an efficient, accurate, non-contact vehicle overload identification method which can keep the road unimpeded. The method can detect the vehicle overload by the relative distance (as the characteristic distance) between the dynamic vehicle's marking line and the road surface. First, the dynamics model of the vehicle suspension is set up. Then, the dynamic characteristic distance of the traffic vehicle is detected from the image acquired by the calibrated camera based on computer vision and image recognition technology.

The RON(Research Octane Number) is the most important indicator of motor petrol, and the petrol refining process is one of the important links in petrol production. However, RON is often lost during petrol refining and RON Loss means the value of RON lost during petrol refining. The prediction of the RON loss of petrol during the refining process is helpful to the improvement of petrol refining process and the processing of petrol. The traditional RON prediction method relied on physical and chemical properties, and did not fully consider the high nonlinearity and strong coupling relationship of the petrol refining process. There is a lack of data-driven RON loss models. This paper studies the construction of the RON loss model in the petrol refining process.

Intelligent sweeper vehicle is gradually applied to human life, in which the accuracy of garbage identification and classification can improve cleaning efficiency and save labor cost. Although Deep Learning has made significant progress in computer vision and the application of semantic network segmentation can improve waste identification rate and classification accuracy. Due to the loss of some spatial information during the convolution process, coupled with the lack of specific datasets for garbage identification, the training of the network and the improvement of recognition and classification accuracy are affected. Based on the Unet algorithm, in this paper we adjust the number of input and output channels in the convolutional layer to improve the speed during the feature extraction part. In addition, manually generated datasets are used to greatly improve the robustness of the model.

A feasible method was developed to reconstruct the three-dimensional flame surface of SI combustion based on 2D images. A double-window constant volume vessel was designed to simultaneously obtain the side and bottom images of the flame. The flame front was reconstructed based on 2D images with a slicing model, in which the flame characteristics were derived by slicing flame contour modeling and flame-piston collision area analysis. The flame irregularity and anisotropy were also analyzed. Two different principles were used to build the slicing model, the ellipse hypothesis modeling and deep learning modeling, in which the ellipse hypothesis modeling was applied to reconstruct the flame in the optical SI engine. And the reconstruction results were analyzed and discussed. The reconstruction results show that part of the wrinkled and folded structure of the flame front in SI engines can be revealed based on the bottom view image.

The acquisition of vehicle dimensions in a vehicle’s moving process has a wide application in road monitoring, transportation, vehicle model recognition and non-contact overload recognition. At present, the detection of the vehicle dimensions mostly adopts the methods of human visual inspection and tool detection, which has a low detection efficiency and difficult to replicate on a large scale. Based on the image background subtraction method, this paper proposes a vehicle dimensions detection method, which can realize real-time detection of road vehicle dimensions. This method uses an adaptive Gaussian Mixture Model (GMM) to establish a background model based on the video stream. Initially, the moving target image is obtained by the background subtraction method, and then the edge detection under the Canny operator and Hough transform circle detection are performed on the image to obtain the pixel dimension of the vehicle's outline.

As motor assembly of Battery Electric Vehicle (BEV) replaces engine system of Internal Combustion Engine (ICE) vehicle, interior structure-borne noise induced by road random excitation becomes more prominent under middle and high speed. The research is focused on central driving type BEV. In order to improve interior noise in middle and low frequency range, dynamic load of BEV body must be identified. Consequently the structural noise induced by road excitation is conducted. The limitations of common identification method for dynamic body load are analyzed. The applied several identification methods are proposed for deterministic dynamic load such as engine or motor. Random dynamic load generated by road excitation is different from deterministic dynamic load. The deterministic load identification method cannot be applied to the random load directly. An identification method of dynamic body load for BEV is presented based on power spectrum decomposition.

Recent research has shown that butanol, instead of ethanol, has the potential of introducing a more suitable blend in diesel engines. This is because butanol has properties similar to current transportation fuels in comparison to ethanol. However, the main downside is the high cost of the butanol production process. Acetone-butanol-ethanol (ABE) is an intermediate product of the fermentation process of butanol production. By eliminating the separation and purification processes, using ABE directly in diesel blends has the potential of greatly decreasing the overall cost for fuel production. This could lead to a vast commercial use of ABE-diesel blends on the market. Much research has been done in the past five years concerning spray and combustion processes of both neat ABE and ABE-diesel mixtures. Additionally, different compositions of ABE mixtures had been characterized with a similar experimental approach.

For the purpose of lessening fuel consumption, engine noise, shift jerk and clutch friction work in the shift process of Automatic Mechanical Transmission (AMT), a fuzzy-bang bang dual mode control strategy for engine rotate speed is put forward in this paper, which takes the advantages of time optimal control and fuzzy control. The combined control strategy is applied to the shift process control of AMT test minibus named SC6350 and proved to be successful by the experimental results.

In this paper, we focus on the active vehicular suspension controller design. A quarter-vehicle suspension system is employed in the system analysis and synthesis. Due to the difficulty and cost in the measuring of all the states, we only choose two variables to construct the feedback loop, that is, the control law is a static-output-feedback (SOF) control. However, the sensor reduction would induce challenges in the controller design. One of the main challenges is the NP-hard problem in the corresponding SOF controller design. In order to deal with this challenge, we propose a two-stage design method in which a state-feedback controller is firstly designed and then the state-feedback controller is used to decouple the nonlinear conditions. To better compensate for the varying vehicle load, a robust load-dependent control strategy is adopted. The proposed design methodology is applied to a suspension control example.

Lane detection is an important component in automatic pilot system and advanced driving assistance system (ADAS). The stability and precision of lane detection will directly determine precision of control and lane plan of vehicles. Traditional mechanical vision lane detection approaches in complicated environment have the deficiencies of low precision and feature semantic description disabilities. But the lane detection depending on deep learning, e.g. SCNN network, LaneNet network, ENet-SAD network have imbalance problems of splitting precision and storage usage. This paper proposes an approach of high-efficiency deep learning Segnet-LSTM semantic segmentation network. This network structure is composed with encoding network and corresponding decoding networks. First, convolution and maximum pooling. The proposal extracts texture details of five images and stores searching position of maximum pooling. Meanwhile, it will implement interpolate processing to the lost points.

In this paper, it was researched the degradation characteristics of catalytic performance of three kinds of DPFs (C1, C2 and C3, with precious metal concentrations being 15, 25 and 35 g/ft3 respectively) after diesel truck aging. It is found out that the crystallinity of three kinds of DPF samples (Used) in full vehicle aging was higher than that of fresh samples (Fresh) and aged samples (Aged) in the laboratory. Compared with Fresh samples, the concentration of Pt atom in precious metal on the surface of Aged and Used samples tends to decrease in most cases. Activities to CO and C3H8 of Aged and Used samples of three kinds of DPFs had all been degraded, and activity degradation showed a substantial correlation with concentration reduction rate of precious metal on the carrier surface. NO2 productivity of Used samples all rose. Crystallinity of DPF samples after full vehicle aging in Inlet, Middle and Outlet areas successively increased.

Before the new car rolls from the line, due to assembly errors, inaccurate four-wheel positioning, etc., the vehicle will run off on a flat road, which will affect the driving comfort and safety. At present, most automobile manufacturers choose to perform the deviation test in the process of vehicle rolls from the line. Compared with other detection methods, the online deviation test system is developed with high precision of laser ranging, fast response and good reliability, which can realize fast and high-precision detection of vehicle deviation. In this paper, test system software is developed based on the LABVIEW, a variety of communication methods to build the communication system, using information check and queue task processing to control, to meet the test needs. Firstly, the calculation model of the deviation of the test system is established.

The issue of car-to-pedestrian impact safety has received more and more attention. For leg protection, a legform impactor with 2 degrees-of-freedom (DOF) proposed by EEVC is required in current regulations for injury assessment, and the Japan Automobile Manufacturers Association Inc. (JAMA) and Japan Automobile Research Institute (JARI) have developed a more biofidelic pedestrian legform since 2000. However, studies show that those existing legforms may not be able to cover some car-to-pedestrian impact situations. This paper documents the development of a new pedestrian legform with 4 DOFs at the knee-joint. It can better represent the kinematics characteristics of human knee-joint, especially under loading conditions in omni-direction impacts. The design challenge is to solve the packaging problem, including design of the knee-joint mechanisms and layout of all the sensors in a limited space of the legform.

Driving assistance system is regarded as an effective method to improve driving safety and comfort and is widely used in automobiles. However, due to the different driving styles of different drivers, their acceptance and comfort of driving assistance systems are also different, which greatly affects the driving experience. The key to solving the problem is to let the system understand the driving style and achieve humanization or personalization. This paper focuses on clustering and identification of different driving styles. In this paper, based on the driver's real vehicle experiment, a driving data acquisition platform was built, meanwhile driving conditions were set and drivers were recruited to collect driving information. In order to facilitate the identification of driving style, the correlation analysis of driving features is conducted and the principal component analysis method is used to reduce the dimension of driving features.

Adaptive cruise control is one of the key technologies in advanced driver assistance systems. However, improving the performance of autonomous driving systems requires addressing various challenges, such as maintaining the dynamic stability of the vehicle during the cruise process, accurately controlling the distance between the ego vehicle and the preceding vehicle, resisting the effects of nonlinear changes in longitudinal speed on system performance. To overcome these challenges, an adaptive cruise control strategy based on the Takagi-Sugeno fuzzy model with a focus on ensuring vehicle lateral stability is proposed. Firstly, a collaborative control model of adaptive cruise and lateral stability is established with desired acceleration and additional yaw moment as control inputs. Then, considering the effect of the nonlinear change of the longitudinal speed on the performance of the vehicle system.

In order to solve the problems of insufficient accuracy for theoretical models and data-driven models for objective drivability evaluation requiring a large amount of data, an objective drivability evaluation method based on a hierarchical mixture model is proposed. First, a novel method of constructing a drivability evaluation system is developed, which combined by work breakdown structure (WBS) and analytic hierarchy process (AHP). Then, downshift condition is taken as a case study, and the subdivision condition is identified based on the hybrid mixture model. What's more, the drivability evaluation indexes of downshift condition are analyzed to establish the evaluation system of drivability.

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.

Driver individualities is crucial for the development of the Advanced Driver Assistant System (ADAS). Due to the mechanism that specific driving operation action of individual driver under typical conditions is convergent and differentiated, a novel driver individualities recognition method is constructed in this paper using random forest model. A driver behavior data acquisition system was built using dSPACE real-time simulation platform. Based on that, the driving data of the tested drivers were collected in real time. Then, we extracted main driving data by principal component analysis method. The fuzzy clustering analysis was carried out on the main driving data, and the fuzzy matrix was constructed according to the intrinsic attribute of the driving data. The drivers’ driving data were divided into multiple clusters.