Cold Start Problem of Vehicle Model Recognition under Cross-Scenario Based on Transfer Learning

As a major function of smart transportation in smart cities,vehicle model recognition plays an important role in intelligent transportation.Due to the difference among different vehicle models recognition datasets,the accuracy of network model training in one scene will be greatly reduced in another one.However,if you don’t have a lot of vehicle model datasets for the current scene,you cannot properly train a model.To address this problem,we study the problem of cold start of vehicle model recognition under cross-scenario.Under the condition of small amount of datasets,combined with the method of transfer learning,load the DAN(Deep Adaptation Networks)and JAN(Joint Adaptation Networks)domain adaptation modules into the convolutional neural network AlexNet and ResNet,and get four models:AlexNet-JAN,AlexNet-DAN,ResNet-JAN,and ResNet-DAN which can achieve a higher accuracy at the beginning.Through experiments,transfer the vehicle model recognition from the network image dataset(source domain)to the surveillance-nature dataset(target domain),both Top-1 and Top-5 accuracy have been improved by at least 20%.

MULTI-AXLE VEHICLE STABILITY BASED ON WHOLE VEHICLE MODEL

From the analysis of experiment data of the multi-axle vehicle chassis searching process, it is less accurate to predict multi-axle vehicle dynamic characteristic with simplified two-axle vehicle model. So it is important to find out a more effective modeling method in the study of multi-vehicle stability. In the development of heat transfer fluid（HTF） six-axle vehicle, a whole vehicle multi-body dynamic model is built through collaborate flowchart using Teamcenter Engineering, UG NX3 and MSC.Adams. The modeling method of connected hydragas spring suspension is validated by running test results. Based on this whole vehicle model, a kinematical analysis of suspension is implemented to achieve optimized suspension geometry parameters according to the stable requirement. Then, different handling simulations are carried out with regard to various tire characteristics, driving con- figurations, and equipments. According to the evaluation of whole vehicle handling characteristic, some design rules are summarized to improve the stability of multi-axle vehicle.

Active suspension with optimal control based on a full vehicle model

The 7-DOF model of a full vehicle with an active suspension is developed in this paper.The model is written into the state equation style.Actuator forces are treated as inputs in the state equations.Based on the basic optimal control theory,the optimal gains for the control system are figured out.So an optimal controller is developed and implemented using Matlab/Simulink,where the Riccati equation with coupling terms is deduced using the Hamilton equation.The all state feedback is chosen for the controller.The gains for all vehicle variables are traded off so that majority of indexes were up to optimal.The active suspension with optimal control is simulated in frequency domain and time domain separately,and compared with a passive suspension.Throughout all the simulation results,the optimal controller developed in this paper works well in the majority of instances.In all,the comfort and ride performance of the vehicle are improved under the active suspension with optimal control.

Development and Validation of a Scaled Electric Combat Vehicle Tire Model

Pneumatic tire modeling and validation have been the topic of several research papers, however, most of these papers only deal with pneumatic passenger and truck tires. In recent years, wheeled-scaled vehicles have gained lots of attention as a feasible testing platform, nonetheless up to the authors’ knowledge there has been no research regarding the use of scaled tires and their effect on the overall vehicle performance characteristics. This paper presents a novel scaled electric combat vehicle tire model and validation technique. The pro-line lockdown tire size 3.00 × 7.35 is modeled using the Finite Element Analysis (FEA) technique and several materials including layered membrane, beam elements, and Mooney-Rivlin for rubber. The tire-rim assembly is then described, and the rigid body analysis is presented. The tire is then validated using an in-house custom-made static tire testing machine. The tire test rig is made specifically to test the pro-line tire model and is designed and manufactured in the laboratory. The tire is validated using vertical stiffness and footprint tests in the static domain at different operating conditions including several vertical loads. Then the tire is used to perform rolling resistance and steering analysis including the rolling resistance coefficient and the cornering stiffness. The analysis is performed at different operating conditions including longitudinal speeds of 5, 10, and 15 km/h. This tire model will be further used to determine the tractive and braking performance of the tire. Furthermore, the tire test rig will also be modified to perform cornering stiffness tests.

Application of Modelica Based Multi- Domain Modeling and Simulation for Gravity-1

In the R&D phase of Gravity-1(YL-1), a multi-domain modeling and simulation technology based on Modelica language was introduced, which was a recent attempt in the practice of modeling and simulation method for launch vehicles in China. It realizes a complex coupling model within a unified model for different domains, so that technologists can work on one model. It ensured the success of YL-1 first launch mission, supports rapid iteration, full validation, and tight design collaboration.

Approaches for a 3D assessment of pavement evenness data based on 3D vehicle models

Pavements are 3D in their shape. They can be captured in three dimensions by modern road mapping equipment which allows for the assessment of pavement evenness in a more holistic way as opposed to current practice which divides into longitudinal and transversal evenness. It makes sense to use 3D vehicle models to simulate the effects of 3D surface data on certain functional criteria like pavement loading, cargo loading and driving comfort. In order to evaluate the three criteria mentioned two vehicle models have been created： a passenger car used to assess driving comfort and a truck-semitrailer submodel used to assess pavement and cargo loading. The vehicle models and their application to 3D surface data are presented. The results are well in line with existing single-track （planar） models. Their advantage over existing 1D/2D models is demonstrated by the example of driving comfort evaluation. Existing ＂geometric＂ limit values for the assessment of longi- tudinal evenness in terms of the power spectral density could be used to establish corre- sponding limit values for the dynamic response, i.e. driving comfort, pavement loading and cargo loading. The limit values are well in line with existing limit values based on planar vehicle models. They can be used as guidelines for the proposal of future limit values. The investigations show that the use of 3D vehicle models is an appropriate and meaningful way of assessing 3D evenness data gathered by modern road mapping systems.

Practical Model for Real Structure and Performance of Heavy Vehicles

In this paper, a new practical model for real heavy vehicle structure is developed to investigate dynamic responses under steering/acceleration or braking maneuvers. The generalized six DoFs （degrees-of-freedom） nonlinear vehicle model M1 including longitudinal, lateral, yaw, vertical, roll and pitch dynamics is validated using the measured data reported in different studies. This model takes the CG （center of gravity） of sprung mass, unsprung mass and total vehicle mass into account. Based on this model, the effects of the inertia parameters on the vehicle dynamic responses are investigated for more comprehensive assessments of the model structure. Another nonlinear vehicle model 342 derived from M1 which assumes that the vehicle has a single CG as reported in literature is also developed. The dynamic responses of the vehicle model Mj are compared with those of the model M2 to demonstrate the performance potential of the proposed nonlinear model. The results of dynamic responses with the nonlinear vehicle model MI suggest that the model could offer considerable potential in realizing enhanced ride and handling performance, as well as improved roll and pitch properties in a flexible manner.

Smartphone-based bridge frequency identification using vehicle contact-point response

Bridge frequency(BF)identification using the vehicle scanning method has attracted considerable attention during the last two decades.However,most previous studies have adopted unrealistic vehicle models,thus finding limited practical applications.This study proposes a smartphone-based BF identification method that uses the contact-point acceleration response of a four degree-of-freedom vehicle model.The said response can be inferred from the vehicle body response measured by a smartphone.For realizing practical applications,this method is incorporated into a self-developed smartphone app to obtain data smoothly and identify BFs in a timely manner.Numerical and experimental investigations are performed to verify the effectiveness of the proposed method.In particular,the robustness of this method is investigated numerically against various factors,including the vehicle speed,bridge span,road roughness,and bridge type.Furthermore,laboratory calibration tests are performed to investigate the accuracy of the smartphone gyroscope in measuring the angular velocity,where anomalous data are detected and eliminated.Laboratory experiment results for a simply supported bridge indicate that the proposed method can be used to identify the first two BFs with acceptable accuracy.

Localization Of Introduced Vehicle Models

1. General Since early 1980s’, Chinese auto plants have successively introducedheavy truck models, light vehicle models, car models and their manufacturing technologies from abroad (see Table Ⅰand Ⅱ). Here is a summary of LCR of different vehicle models introduced. (1) Cars Presently, the economic scale for car production of Shanghai-VW, FAW-VW,

LCR of Imported Vehicle Models

Santana of Shanghai VW: Since its formal operation in Sept. of 1985, except the new model of Santana 2000, Shanghai VW has produced 500 thousands units of Santana passenger cars totally with the

Vehicle kinematics modeling and design of vehicle trajectory generator system

A trajectory generator based on vehicle kinematics model was presented and an integrated navigation simulation system was designed.Considering that the tight relation between vehicle motion and topography,a new trajectory generator for vehicle was proposed for more actual simulation.Firstly,a vehicle kinematics model was built based on conversion of attitude vector in different coordinate systems.Then,the principle of common trajectory generators was analyzed.Besides,combining the vehicle kinematics model with the principle of dead reckoning,a new vehicle trajectory generator was presented,which can provide process parameters of carrier anytime and achieve simulation of typical actions of running vehicle.Moreover,IMU(inertial measurement unit) elements were simulated,including accelerometer and gyroscope.After setting up the simulation conditions,the integrated navigation simulation system was verified by final performance test.The result proves the validity and flexibility of this design.

BRAIN INJURY BIOMECHANICS IN REAL WORLD VEHICLE ACCIDENT USING MATHEMATICAL MODELS

This paper aims at investigating brain injury mechanisms and predicting head injuries in real world accidents. For this purpose, a 3D human head finite element model （HBM-head） was developed based on head-brain anatomy. The HBM head model was validated with two experimental tests. Then the head finite element（FE） model and a multi-body system （MBS） model were used to carry out reconstructions of real world vehicle-pedestrian accidents and brain injuries. The MBS models were used for calculating the head impact conditions in vehicle impacts. The HBM-head model was used for calculating the injury related physical parameters, such as intracranial pressure, stress, and strain. The calculated intracranial pressure and strain distribution were correlated with the injury outcomes observed from accidents. It is shown that this model can predict the intracranial biomechanical response and calculate the injury related physical parameters. The head FE model has good biofidelity and will be a valuable tool for the study of injury mechanisms and the tolerance level of the brain.

Vehicle Detection Based on Visual Saliency and Deep Sparse Convolution Hierarchical Model

Traditional vehicle detection algorithms use traverse search based vehicle candidate generation and hand crafted based classifier training for vehicle candidate verification.These types of methods generally have high processing times and low vehicle detection performance.To address this issue,a visual saliency and deep sparse convolution hierarchical model based vehicle detection algorithm is proposed.A visual saliency calculation is firstly used to generate a small vehicle candidate area.The vehicle candidate sub images are then loaded into a sparse deep convolution hierarchical model with an SVM-based classifier to perform the final detection.The experimental results demonstrate that the proposed method is with 94.81% correct rate and 0.78% false detection rate on the existing datasets and the real road pictures captured by our group,which outperforms the existing state-of-the-art algorithms.More importantly,high discriminative multi-scale features are generated by deep sparse convolution network which has broad application prospects in target recognition in the field of intelligent vehicle.

Dynamics Modeling and Simulation of Autonomous Underwater Vehicles with Appendages

To provide a simulation system platform for designing and debugging a small autonomous underwater vehicle＇s （AUV） motion controller, a six-degree of freedom （6-DOF） dynamic model for AUV controlled by thruster and fins with appendages is examined. Based on the dynamic model, a simulation system for the AUV＇s motion is established. The different kinds of typical motions are simulated to analyze the motion performance and the maneuverability of the AUV. In order to evaluate the influences of appendages on the motion performance of the AUV, simulations of the AUV with and without appendages are performed and compared. The results demonstrate the AUV has good maneuverability with and without appendages.

MODELING AND IMPLEMENTATION OF SPEED GOVERNOR FOR THE HYBRID ELECTRIC VEHICLE ENGINE

A speed control analysis for an in-line gasoline fueled internal combustion （IC） engine is presented for the purpose of alleviation of high frequency oscillations in engine revolutions. A dynamic cylinder-by-cylinder model is proposed, base on slider-crank mechanism, which is extended to develop a digital governor providing a high fidelity estimation of rotary speed oscillation for hybrid vehicle engines. A modified PID controller that P and I gain is placed in feedback path is also described for hybrid electric vehicle （HEV） engine speed regulation, By comparison between measured and estimated signals, it is demonstrated that a good agreement has been achieved and the governor behaves an excellent damping speed ripple.

Speed prediction models for car and sports utility vehicleat locations along four-lane median divided horizontal curves

Sites with varying geometric features were analyzed to develop the 85 th percentile speed prediction models for car and sports utility vehicle(SUV) at 50 m prior to the point of curvature(PC), PC, midpoint of a curve(MC), point of tangent(PT) and 50 m beyond PT on four-lane median divided rural highways. The car and SUV speed data were combined in the analysis as they were found to be normally distributed and not significantly different. Independent parameters representing geometric features and speed at the preceding section were logically selected in stepwise regression analyses to develop the models. Speeds at various locations were found to be dependent on some combinations of curve length, curvature and speed in the immediately preceding section of the highway. Curve length had a significant effect on the speed at locations 50 m prior to PC, PC and MC. The effect of curvature on speed was observed only at MC. The curve geometry did not have a significant effect on speed from PT onwards. The speed at 50 m prior to PC and curvature is the most significant parameter that affects the speed at PC and MC, respectively. Before entering a horizontal curve, drivers possibly perceive the curve based on its length. Longer curve encourages drivers to maintain higher speed in the preceding tangent section. Further, drivers start experiencing the effect of curvature only after entering the curve and adjust speed accordingly. Practitioners can use these findings in designing consistent horizontal curve for vehicle speed harmony.

Mechanics analysis of vehicle bumping at approach slabs with superelevation

In order to analyze the pavement stress caused by vehicle bumping at an approach slab, a simplified four-wheeled bi- axle vehicle-moving model is proposed. The effect of damping and vibration reduction in the process of vehicle-moving is not considered. Based on the position change of vehicle wheels at the approach slab, the vehicle dynamic vibration equations are summarized. Meanwhile, the undetermined coefficients of the vibration equations are obtained using the boundary and initial conditions of the vehicle. The analytical motion solutions of rear and front wheels at different stages are concluded. Consequently, a four-wheeled vehicle model is developed and vibration equations are provided, which can be used to analyze the impact of complicated stress on pavement. The results show that the excessive stress and stress concentration will occur at the approach slab, and it needs to be strengthened.

Adaptive Backstepping Control Design for Semi-Active Suspension of Half-Vehicle With Magnetorheological Damper

This paper investigates the problem of controlling half-vehicle semi-active suspension system involving a magnetorheological(MR)damper.This features a hysteretic behavior that is presently captured through the nonlinear Bouc-Wen model.The control objective is to regulate well the heave and the pitch motions of the chassis despite the road irregularities.The difficulty of the control problem lies in the nonlinearity of the system model,the uncertainty of some of its parameters,and the inaccessibility to measurements of the hysteresis internal state variables.Using Lyapunov control design tools,we design two observers to get online estimates of the hysteresis internal states and a stabilizing adaptive state-feedback regulator.The whole adaptive controller is formally shown to meet the desired control objectives.This theoretical result is confirmed by several simulations demonstrating the supremacy of the latter compared to the skyhook control and passive suspension.

A New Application of Multi-Body System Dynamics in Vehicle-Road Interaction Simulation

In vehicle-road interaction simulation, multibody system (MBS) dynamics as well as the corresponding software ADAMS has been employed to model the nonlinear vehicle in more detail. The simulation method has been validated by the test data, and been compared to the former simple models. This method can be used for estimating the effects of dynamic tire forces and other vehicle features on road damage so that the “road-friendliness” can be assessed in vehicle design process.