VERTICAL ROAD LOAD ESTIMATION FOR A VEHICLE FROM MEASURED ACCELERATION DATA USING KALMAN FILTER TECHNIQUE

摘要

Vertical loads acting on a vehicle under dynamic conditions due to terrain inputs are highly essential for obtaining optimum design of the vehicle structure; hence inverse problems such as input force estimation from the response (acceleration/velocity/displacement) of structures play a vital role in such applications where the input forces acting on the structures cannot be measured directly. The present work deals with the estimation of input vertical loads for a simplified N+1 degrees of freedom (DOF) in-plane half car model of a tracked vehicle on two different terrain: i) Sinusoidal and ii) Aberdeen Proving Ground (APG) using the Kalman filter technique. The vehicle model includes the hull and turret as the sprung mass, the wheel station assemblies as the unsprung masses and the torsion bar suspension elements and dashpots connecting the sprung and unsprung masses. The DOF considered here are the bounce motions of the N road wheels on one side of the vehicle and that of the sprung mass. In Kalman filter technique, the system is modelled in terms of state space equations and system parameters such as mass, stiffness and damping coefficient are necessary for estimating the input force. The continuous state space form is discretized. Studies have been carried out on a twelve-wheeled tracked vehicle (military tank). The vehicle was made to run over the two tracks at a constant speed of 15 kmph and acceleration responses were measured at various locations to estimate the vertical excitation load. Both the vehicle model and methodology used in the inverse problem are verified using experimental trials and the results are comparable. Comparison between estimated and measured road loads gives a reasonable estimate for Sinusoidal and APG tracks. The error was found to be 13.12% for the Sinusoidal track and 30.74 % for the APG track.