Inter-vehicle Distance Aided Multi-vehicle Cooperative Localization using Split CIF

In autonomous driving, localization is very important to ensure the accuracy of all the other tasks, and multi- vehicle cooperative localization is a popular research direction in recent years for its improvement on state estimation. However, due to the complex environment, complete information about relative pose for cooperative localization could sometimes be hard to be obtained. Since the inter-vehicle distance is easily measured, the use of inter-vehicle distance for multi-vehicle cooperative localization has become a meaningful topic. In this paper, we propose a general decentralized framework for the inter-vehicle distance aided cooperative localization. In order to fuse the inter-vehicle distance with the state estimation of ego-vehicle, the split covariance intersection filter (Split CIF) is used as the data fusion method. Then, we verify the advantages of the proposed method through simulation-based experiments, demonstrate its applicable conditions and provide its practical application scenarios.