The study of controllable suspension properties special in the characteristics of improving ride comfort and road handling is a challenging task for vehicle industry. Currently, since most suspension control requires the observation of unmeasurable state, how to accurately acquire the state of a suspension system attracts more attention. To solve this problem, a novel approach interacting multiple mode Kalman Filter (IMMKF) is proposed in this paper. Suspension system parameters are crucial for the performance of state observers. Uncertain suspension system parameters in various conditions, e.g. due to additional load, have significant effect on state estimation. Simultaneously, state transition among different models may be happened on the condition of varying system parameters. Hence, the interacting multiple mode Kalman Filter (IMMKF) observer is designed by employing the Kalman filter (KF) theory based on the interacting multiple mode (IMM) representation of the augmented suspension model in various working conditions. Considering the variation of suspension mass in the state transition process, a recursive least square (RLS) approach is used to estimate the sprung mass of suspension system. Finally, in simulations and experiments on a quarter vehicle test rig using International Standardization Organization (ISO) road excitation inputs, the KF algorithm and IMMKF algorithm are verified, and the results show that more accurate state estimation are obtained by combining the IMMKF and RLS algorithm. The research achievements provide a reasonable algorithm to apply to the controllable suspension system.
State Estimation Based on Interacting Multiple Mode Kalman Filter for Vehicle Suspension System
Sae Technical Papers
WCX™ 17: SAE World Congress Experience ; 2017
2017-03-28
Conference paper
English
State Estimation Based on Interacting Multiple Mode Kalman Filter for Vehicle Suspension System
British Library Conference Proceedings | 2017
|Extended Kalman Filter for Vehicle Suspension System
Trans Tech Publications | 2014
|Unscented Kalman filter for vehicle state estimation
Online Contents | 2011
|