Wu Chaozhong

Effects of Driver's Unsafe Acceleration Behaviors on Passengers' Comfort for Coach Buses

The driver’s unsafe acceleration behaviors from coach buses, such as hard acceleration and deceleration, are the significant factors that influence passengers comfort and may cause accidents. It is necessary to explain the relationship between passengers comfort and acceleration behaviors. The field experiment with 35 passengers and three professional bus drivers was conducted to determine whether bus passengers comfort is influenced by driving behaviors. Subjective comfort feelings of passengers were collected by the means of questionnaires. The questionnaires were distributed on board and acceleration and deceleration values of the bus were received from CAN bus. Therefore, the subject views of the passengers could be compared to object acceleration data. Finally, the effect of acceleration threshold for passenger comfort on the bus was found. The results show that the passergers start to feel uncomfortable when the acceleration, a is greater than or equal to 1.5 m/s 2 and the deceleration is less than or equal to -0.75m/s 2 . Therefore driver acceleration skills should be trained before driving buses.

Target tracking using Kalman Filter Embedded Trust Region

This paper proposes a novel algorithm, the Kalman Filter Embedded Trust Region (KFETR), for target tracking. Kalman filter and trust region are two successful methods for object tracking. The presented KFETR algorithm integrates the advantages of the two approaches. The new algorithm makes full use of the targets moving information and predicts the targets approximate position firstly. Because of these properties, the algorithm overcomes the problem that trust region converges to a local minimum which is not of interest caused by improper initial position. Promising experimental results on several image sequences demonstrate the robustness and effectiveness of KFETR.

Analysis on Vehicle Motion Parameters of Distracted Driving

Vehicle motion parameters are important indicators of detection distracted driving. However, vehicle motion parameters are rarely used by previous studies and attempts on detection distracted driving. In this study, the differences of vehicle motion parameters between distracted driving and normal driving are analyzed. In this study, we use the three-axis accelerometer and gyroscope of a device to record and analyze lateral acceleration of vehicle. On-road experiment was conducted to collect testing data. The results show that lateral acceleration change of the vehicle motion parameters have differences pattern between distracted driving and normal driving.It is found that vehicle motion parameters of distracted driving does not obey normal distribution. Suddenly maneuvers of lateral acceleration change under distracted driving respectively reach more than 0.25G (approximately 2.5m/s 2 ). This

Real-Time Automatic Evaluation Technology in Vehicle Road Test System Based on Neural Network

With the development of computer, sensors, artificial intelligence technology, network, advanced automotive technology, the vehicle road test by a automatic evaluation system from the traditional manual evaluation has become an irresistible trend. Currently, the research of intelligent test system for vehicle road test in China has just started. Automatic evaluation module is core component of the system. In this study, according to the theory and technology of artificial neural network and the characteristics and requirements of the vehicle road test system, we introduce the one-way transmission of multi-layer neural network structure to the automatic evaluation module, discuss the principle, structure, transfer function and specific implementation of the application. Finally the proposed model was applicated in the vehicle road test system.

A novel estimation algorithm for interpolating ship motion

Ship motions are usually employed to analyze water transportation system. Interpolation of ship motions can be used for estimating the lost ship movement information, which is important for analyzing water traffic. Although some methods are proposed to interpolate ship trajectory, these methods are needed to improve interpolation accuracy. In addition, the course and speed of ship are not interpolated in previous studies. This article proposes a novel estimation algorithm for interpolating ship motions. First, the bilateral filtering is used to smooth the navigation trajectories, then a novel interpolation approach forecasts the vessel positions during the interpolation time period by taking into account the dynamic information of Automatic Identification System (AIS), and then forms the interpolation track of the moving vessel. In order to validate the proposed method, the AIS data of Yangtze River is collected from the AIS stations and the proposed restoration methods are verified by MATLAB under simulation. The results show that the proposed method can effectively remove the inaccurate ship motion data and interpolate the lost ship motion data.

Analysis traffic safety for highway off-rampbased on visual reaction time on traffic signs

To ensure traffic safety of highway off-ramp area, it should leave enough reaction time for drivers taking corresponding actions after traffic signs. And the recognition time for traffic signs depends on information volume. In this study, traffic sign recognition process in highway off-ramp were analyzed and visual recognition time model was set up. Then, the information volume contained in traffic signs was calculated using the information theory, and the information was divided it into four grades. In order to analyze the recognition time for different information grades, a simulation driving experiment was implemented. 20 participants took part in this experiment and the time for four grades was obtained. An evaluation approached was proposed, which identifies the safety conditions of highway off-ramp into four levels according to the time ranges after recognition. The proposed traffic safety evaluation method may provide a new way to evaluate the safety for highway off-ramp for its intuitive, easy operation and reasonable, especially for setting traffic signs information volume.

An inexact bus departure frequency model for traffic pollution control

This study presents an interval bus departure frequency programming model for traffic system pollution control through odd-even restrictions on private vehicles policy under uncertainties. The restriction policy results in enormous cost of the public transportation system especially in peak hours. The developed model can effectively quantify the relationship among system cost, execution time of the restriction policy and traffic environment by considering the uncertainties of traffic system. The uncertainties presented as interval numbers in the public transportation system can be quantified with the interval programming. The developed model is applied to Wuhan city, China. The results indicate that the efficiency of the program is significantly influenced by automobile ownership, the execution time of the policy and the level of emission. Besides, the program can easily be extended to all bus routes for vehicle emissions management.

A design of brain-computer interface control platform for intelligent vehicle

It is always a hot spot of the brain-computer field to achieve various functions by utilizing electroencephalogram (EEG) to control external equipment. This paper describes a design of an intelligent vehicle control platform based on brain-computer interface (BCI), which obtains EEG through BCI, then simulation model is established by Matlab /Simulink to extract and classify EEG as control command, which will be sent to intelligent vehicle by wireless serial port communication, so as to realize the control of intelligent vehicle on roads of the sandbox. By imaging four status: left hand, right hand, right leg and resting, the platform can accomplish the real-time control of intelligent vehicle on road of the sandbox, including turning left, turning right, going forward and stopping. The findings of that experiment show that the control system has good feasibility and stability, which lays the foundation for the practical application of the control of intelligent vehicle via EEG signal. The accomplishment of this platform provides a new approach to expanding and improving human capability to control external equipment.

A Recognition Model for Lane Change Intention Based on Neural Network

Timely and accurate identification of driver’s intention is considered helpful to avoid collisions. Driving in an on-road traffic situation makes it much more difficult to recognize the driving intention because of dynamic and uncertain information. However, the intentions can be inferred by the driver’s behaviors as he prepares to execute an action. The current studies focus on classifying driving intentions into car-following and lane changing. Therefore, a lane change intention recognition method was proposed in this study. A total of 10 on-road experiments on freeways were conducted to acquire time headway, driver’s eye-motion data, steering angle and lane departure data. With these data, the typical features under free driving and lane change conditions were analyzed respectively. Back-Propagation Neural Networks method was applied in this study to achieve the driving intention classification function. The proposed model successfully recognized 95 intentions in 100 lane change behaviors. The recognition result of the model can fit the target output at a degree of 0.99. This model is useful for surrounding drivers to choose the safest route to avoid collisions in the vehicle-infrastructure technology.

A Multi-Sensory Approach for Comprehensive Detection and Warning of Vehicle Collision Risk

The collision warning system, based on multi-sensor, can perceive more information than a unilateral system, just like vehicle motion status, driver intention, and road traffic environment; thus, it can provide more accurate judgments of driving safety. In this paper, physical and logical framework of a collision warning system based on multi-sensor is designed in detail. Vehicle to vehicle communication (V2V), vehicle to infrastructure communication (V2I), and vehicular sensors are used to acquire the motion status of the subject vehicle, neighboring vehicles, and pedestrians. Then, the short-term future motion status is predicted based on the Kalman algorithm, and the vehicle collision risk can be detected in advance by a collision detection algorithm. Furthermore, the safe distance threshold as a collision risk indicator is calculated and compared with the predicted distance. Finally, the vehicle collision risk can be detected according to the comparison. Simulation results show that the method is effective for judging the collision risk and providing an accurate warning prompt.