Study On Modeling For The Traffic Vehicle Motion Behavior In The Traffic Flow Simulation

With the rapid development of automobile industry, the traffic problem becomes moreand more prominent. The study on traffic problem attracts more and more attention. The trafficflow simulation technology using computer has been regarded as an effective and importantmeans for the research on the traffic problem because of its low cost and good repeatability.In addition, with the rapid development of automobile technology, a variety ofautomotive electronic technologies are gradually on the rise. A realistic traffic flow simulationenvironment is also needed when doing the research on the development of automotiveelectronics technology and driver behavior. The modeling on the motion behavior of vehiclesis a key to the traffic flow simulation. Currently the study on the traffic vehicles are mainlyinvolved in the modeling on vehicle behavior. But the essence of the study above is to researchon the decision-making behavior of the driver. There are few studies that is about how to builda vehicle model whose dynamics characteristic is taken into account. It makes the motioncharacteristic of the vehicle in the traffic flow simulation far from the real. To sum up, out ofnecessity for research on the traffic flow simulation, driver behavior and the development ofelectronics product for the automotive and considering the deficiency of current traffic flowsimulation system, the paper carries a study on the motion behavior of traffic vehicle in thetraffic flow simulation system. The traffic vehicle model which is consist of the vehicle modeland driver model should have the ability for self driving. The motion behavior of the trafficvehicle depends on its dynamic characteristics and driver’s operation. In this paper, we willfocus on the two aspects below: vehicle dynamic model and driver model. A seven-DOFvehicle dynamics model and a driver model integrated with the direction and speed controlfunctions are established. Some key problems about that are studied and explored. The mainresearch contents and points in the paper are listed below:1. Considering the requirement for real-time simulation and the degree of reality, thetraffic vehicle dynamics model established has seven degrees of freedom, on the one hand, itensures the accuracy of simulation; on the other hand the computing speed is greatly improved.Dynamic characteristics of our model that can reflect the main characteristics of a real vehicleis close to that in the Carsim.2. Based on the Preview Follower theory and PID control method, a driver modelintegrated with direction and speed control functions is established. Referring to the vehiclesteering control, a vehicle model with two degrees of freedom is regarded as the reference to take the effects of the vehicle parameters into account. A direction control method that previewunder the steady-state and correcting with the dynamic state is proposed; Referring to thespeed control, a new control method with PID algorithm is presented here.That is, thecontroller is divided into5control phases. The different control phase will be in use accordingto the current running state of vehicle and driving requirement. Simulation results show thatthe driver model has a good effect on the direction control and speed control.3. Some key parameters of traffic vehicle model are chosen and changed within theallowable range to make the behavior of every vehicle model in the simulation environmentdifferent from each other even when the running task is completely the same.The research in the paper enhances the realism of traffic simulation environment. It willbe the first step to improve the existing traffic flow simulation system and will be of great helpfor improving the traffic flow simulation system. In addition, the traffic vehicle modelestablished in the paper can be also applied to a wide range.