![]() ![]() In for example, an artificial neural network is used to predict the roll angle, in order to be able to realize a vehicle roll control system. ![]() These have the potential to improve the accuracy of the estimation significantly while reducing the modeling effort. This also applies to models based on artificial intelligence. If not, the safety is compromised especially in connection with vehicle dynamics control systems. The determination of the states must be reliable and valid. All approaches share one superior requirement in common. The unscented Kalman filter outperforms the extended Kalman filter, especially due to linearization errors at larger sampling times. compare the performance of an extended and an unscented Kalman filter for vehicle dynamics state estimation. By integrating a second measurement update, the estimation accuracy is enhanced. In, a Kalman filter is utilized to predict the roll and pitch angle. The application of physical-based models within virtual sensors, for example by using Kalman filters, is a well-researched field. Machine learning methods and artificial intelligence in general are frequently used for this purpose. In addition to the classical theoretical modeling approach, it is also possible to generate these mathematical models through experimental modeling. Thus, a valid and reliable state estimate is available throughout. ![]() By using the hybrid method, unreliable estimations by the artificial intelligence-based model resulting from erroneous input signals are detected and handled. The implementation and validation is performed by a co-simulation between IPG CarMaker and MATLAB/Simulink. The state estimation is coupled with a central predictive vehicle dynamics control. The application example is the state estimation of the vehicle roll angle. Therefore, a hybrid method is presented that safeguards the reliability of artificial intelligence-based estimations. Due to the resulting black-box characteristics, virtual sensors based on artificial intelligence are not fully reliable, which can have fatal consequences in safety-critical applications. In addition to physical models resulting from theoretical modeling, artificial intelligence and machine learning approaches are increasingly used, which incorporate experimental modeling. ![]() The use of virtual sensors in vehicles represents a cost-effective alternative to the installation of physical hardware. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. Archives
December 2022
Categories |