| Summary: | 碩士 === 逢甲大學 === 資訊工程學系 === 103 === In this paper, we propose an infrastructure-less cooperative lane positioning (ICLP) framework using vehicle-to-vehicle (V2V) communications. The ICLP framework applies vehicular sensor networks to localize lane positions of vehicles on roads. ICLP allows vehicles equipped with image sensors to detect the current located lane without utilizing inaccurate GPS locations and costly roadside infrastructures. Through image sensors, ICLP can keep recognizing the lane position for a vehicle. With V2V communications, ICLP can request lane position information from other vehicles in the same lane as necessary. ICLP makes innovative applications of intelligent transportation systems relying on high positioning accuracy possible, such as lane-level cooperative collision avoidance, dynamic traffic control, and intelligent personal navigation. An analytical model is derived to estimate the expected positioning success ratio of ICLP. In addition, we verify the superiority and feasibility by implementing our framework in an Android-based prototype. Experimental results show that ICLP can achieve high positioning success ratios and provide accurate lane positions for vehicles.
|
|---|
