Abstract

The problem of intelligent monitoring of the vertical dynamics of wheeled inspection vehicles is addressed. With the independent MacPherson suspension system installed, the basic analysis focuses on the evaluation of the parameters of the so-called quarter car model. To identify a physically motivated continuous description, in practice, dedicated integral-horizontal filters are used. The obtained discrete model, which retains the original parameters, is effectively identified using the classic least squares procedure. Using the method of identification in the sense of the least sum of absolute values, the results of such an assessment become insensitive to sporadic outliers in the sampled data. However, the early signs of possible mechanical defects of the suspension can be seen using the forgetting mechanism. This helps to identify failures that can be recognized by changes in system parameters. Ultimately, the quality of the intelligent vehicle suspension monitoring developed is verified by means of numerical simulations. The problem of intelligent monitoring of the vertical dynamics of wheeled inspection vehicles is addressed. With the independent MacPherson suspension system installed, the basic analysis focuses on the evaluation of the parameters of the so-called quarter car model. To identify a physically motivated continuous description, in practice, dedicated integral-horizontal filters are used. The obtained discrete model, which retains the original parameters, is effectively identified using the classic least squares procedure. Using the method of identification in the sense of the least sum of absolute values, the results of such an assessment become insensitive to sporadic outliers in the sampled data. However, the early signs of possible mechanical defects of the suspension can be seen using the forgetting mechanism. This helps to identify failures that can be recognized by changes in system parameters. Ultimately, the quality of the intelligent vehicle suspension monitoring developed is verified by means of numerical simulations

Citations

Authors (2)