Dissertation > Excellent graduate degree dissertation topics show

Vehicle Longitudinal and Laterl Velocity Estimation Based on Unscented Kalman Filter

Author: ShiYanRu
Tutor: ZhangHongKun
School: Jilin University
Course: Vehicle Engineering
Keywords: Unscented Kalman Filter Vehicle model Dugoff tire model Speed ??estimation
CLC: U461.91
Type: Master's thesis
Year: 2011
Downloads: 186
Quote: 2
Read: Download Dissertation

Abstract

Automotive electronic stability control system ESP (Electronic Stability Program) as a new type of vehicle active safety control system not only integrates the ABS and TCS functions to maintain the stability of the vehicle, but also in more complex extreme working conditions. Currently, foreign ESP control system as a vehicle active safety technology has become the focus of the study and related products have been a lot of equipment and a variety of upscale car sedan. Automotive electronic stability control system in China, has just started, and therefore has important implications independent research and development of China's automobile industry, enhance the capacity and the development of related industries to carry out the study of ESP. The entire ESP control process to achieve this critical technology depends on the measurement and estimation of vehicle motion. State of motion of the car is the car traveling longitudinal speed and lateral speed, yaw angular velocity and sideslip angle reflect the state of the vehicle movement behavior, the incomplete status information for the realization and promotion of automotive active safety control system ESP bring great difficulties, therefore, accurate and real-time access to status information in vehicles is the premise and foundation of effective implementation of the ESP system control functions. ESP control systems require the vehicle travel state, measuring the degree of difficulty, depending on the physical properties thereof vary greatly. Yaw angular velocity can be measured directly through the vehicle sensors get sideslip angle difficult to determine due to the installation location, often using longitudinal velocity and lateral velocity estimated value estimate. Longitudinal velocity and lateral speed due to the direct measurement of the high cost, difficult production problems, generally obtained through the estimation algorithm. Therefore, how to obtain low-cost, accurate and real-time longitudinal velocity and lateral velocity estimates for other vehicle motion, and automotive ESP control system to practical important theoretical significance and engineering significance. In recent years, domestic and foreign research scholars vehicle motion estimation of vehicle dynamics model and state estimation method research ideas. Based on this new idea, combined with the 863 Special, international scientific and technological cooperation special as well as provincial and municipal science and technology support plan major projects for the project research, in close cooperation with China FAW Group Corporation Technology Center, the demand for the the FAW the Pentium cars of technical Pentium sedan stability control system critical state parameters: vehicle longitudinal velocity and lateral velocity estimation was carried out and a multi-condition, and more traffic simulation verification. The simulation results show that the automobile speed control algorithm developed not only to ensure that the real-time and accuracy and with high robustness. The research of this paper are as follows: 1. Kalman filtering theory and algorithms traditional Kalman filtering theory and algorithms and their advantages and disadvantages. On this basis, summed up the Unscented Kalman filtering theory (UKF) advantage, as well as specific processes for optimal estimation of the state variables of the nonlinear system, estimated to have laid the theoretical foundation for the subsequent speed. 2 road adhesion coefficient is estimated based on UKF filtering algorithm road adhesion coefficient is estimated vehicle speed estimated. The vehicle sensor information, combined Dugoff tire model and the four-wheel vehicle model, road adhesion coefficient is estimated using the UKF filtering. UKF filter-based state observer estimated results of simulation in Matlab / Simulink environment. The results show that such an estimation algorithm for real-time road adhesion coefficient is estimated to lay a reliable foundation for the estimated speed behind. 3. Estimated based on the UKF filter vehicle longitudinal speed and lateral speed road adhesion coefficient is estimated on the basis of, based on the seven degrees of freedom nonlinear vehicle model, combined with easy-the algorithm the realization of Dugoff tire model designed based on UKF filter car speed estimator. State board sensor measurements, road friction coefficient estimation algorithm and speed estimation algorithms are combined to form a complete set of vehicle status information to estimate the vehicle longitudinal and lateral speed estimation algorithm. Matlab / Simulink environment to achieve real-time estimates of the vehicle longitudinal speed and lateral speed. Simulation application automobile dynamic simulation and analysis software CarSim vehicle longitudinal and lateral speed estimation algorithm simulation complete different road conditions and different driving conditions. State estimates of this algorithm compared with the same the conditions of Carsim output results verify the feasibility of the algorithm; estimated value at the same time, the algorithm estimates using EKF filtering algorithm compared to verify the algorithm accurately and real-time. Proof car based on Unscented Kalman Filter longitudinal and lateral velocity to estimate the effectiveness of the algorithm. Based on the UKF filter theory, research on the road adhesion coefficient and the vehicle longitudinal and lateral speed estimation algorithm. Joint estimation algorithm of road adhesion coefficient and the speed of the simulation results show that the estimated design joint estimation algorithm of this paper is not only high precision and good real-time, ESP and other advanced automotive chassis electronic control system provides A new research ideas.

Related Dissertations

CLC: > Transportation > Road transport > Automotive Engineering > Car theory > Vehicle safety
© 2012 www.DissertationTopic.Net Mobile