영역별 오차 특성을 고려한 레이더 센서 융합

초록

In this paper, we propose the track-to-track fusion algorithm considering zone-based error analysis of two radar networks. Error analysis of each radar was performed by comparing radar values and ground truth values from Differential Global Positioning System (DGPS) in the uniformly divided area based on the relative distance to the target vehicle. After that two famous distributed fusion algorithm, Convex Combination, Covariance Intersection were applied and compared using real experimental data. Through radar network fusion, we can possibly track objects in blind areas near the ego vehicle, which could not be detected by rear radar alone.