Comparison of Parameters and States Estimators for the Longitudinal Dynamics of a Military Heavy Vehicle

Technological innovations in military vehicles are essential for enhancing efficiency, safety, and operational capability in complex scenarios. Advances such as navigation system automation and the introduction of autonomous vehicles have transformed military mobility. State estimators enable the precise monitoring of critical variables that are not directly accessible by sensors, providing real-time information to controllers and improving dynamic response under variable conditions. Their integration is crucial for the development of advanced control systems. This study aims to develop and compare parameter and states estimators for military heavy vehicles using three methodologies: particle filter, extended Kalman filter, and moving horizon state estimation. Computational simulations employ Pacejka’s magic formula to model tire behavior, and the vehicle modeling is based on a simplified quarter-car model, with an emphasis on longitudinal dynamics. In the end, the estimators are compared through simulated scenarios involving the friction coefficient between the tire and the surface. Their effectiveness, advantages and disadvantages are evaluated, highlighting their ability to provide accurate and robust real-time estimation of parameters and states of the longitudinal dynamics.