An Innovative Data Fusion Method based on Mixed $\mathrm{H}_{\infty}$/Adaptive Kalman Filtering for Target Tracking of Intelligent Vehicles

For the target tracking of intelligent vehicles, data fusion methods based on filtering techniques are the focus of research today. In this paper, to increase the accuracy of data fusion, an innovative data fusion method based on mixed
$\mathrm{H}\infty$
/adaptive Kalman filtering (
$\mathrm{H}\infty$
/AKF) with attenuation factor is presented. The highlight of
$\mathrm{H}\infty$
/AKF is the improved data fusion robustness under different external excitations, uncertain internal parameters, measure and process noises. In addition, in order to handle noise variations caused by environmental disturbances, an attenuation factor is introduced to increase the weight of the currently measured. Finally, the
$\mathrm{H}\infty$
/AKF is compared with traditional AKF to verify the effectiveness of the presented method. Experiment results demonstrate that the
$\mathrm{H}\infty$
/AKF is more precise and reliable under noisy environments.