| Popis: |
The article discusses the problem of identifying the parameters of the vehicle's directional movement, which are necessary for the operation of active safety systems (SAB). The inability to determine some of the parameters necessary for the functioning of the SAB by direct measurements with on-board sensors (due to the absence of corresponding sensors in production vehicles) makes it relevant to use indirect computational methods for identifying these parameters, which are based on mathematical structures called observers. The purpose of this work is to create a system for identifying vehicle motion parameters, which, using the measurements available on board the vehicle and the mathematical apparatus of the theory of observers and optimal filters, indirectly determines unmeasured parameters that are important for the operation of active safety systems. Based on the analysis of existing methods and tools, a diagram of the observer of the parameters of the vehicle's directional movement using the sigma-point Kalman filter is proposed. The observer identifies the lateral component of the vehicle speed vector, the coefficients of the lateral adhesion of the tires to the supporting surface and the wheel slip angles using the vehicle dynamics model and on-board inertial sensors that measure the linear acceleration and yaw rate of the vehicle. The observer's performance and adequacy was confirmed by comparing the parameters he identifies with direct measurements made during road tests of the vehicle. There was used a root-mean-square error of identification as a measure for assessing the accuracy with respect to direct measurements of the parameters of course movement. An additional assessment of the adequacy is made by comparing the identified grip characteristic (the dependence of the coefficient of adhesion on the slip angle) with the characteristic obtained by approximation using a mathematical model of the tire. The assessment showed a good quality of identification of course movement parameters provided by the developed observer, which gives grounds to consider it a useful tool for research and development of active safety systems. |
