Authors: Albert Budi Christian; Chih-Yu Lin; Lan-Da Van; Yu-Chee Tseng
Publication Date: February 4, 2022 (Conference held: 20–23 December 2021)
Conference: 2021 IEEE 14th International Symposium on Embedded Multicore/Many-core Systems-on-Chip (MCSoC) Singapore, Singapore
Abstract: This paper proposes a data fusion–driven lane-level precision data transmission system for V2X road applications, aiming to deliver targeted information only to relevant vehicles without interfering with others. The system integrates computer vision and sensor fusion techniques using three types of sensory inputs: roadside camera, GPS, and magnetometer. By extracting and combining common features across these inputs, the proposed approach accurately identifies vehicles located in specific lanes and selectively transmits messages to them. Experimental results demonstrate a recognition rate of 87.34%, with end-to-end message latency below 72 ms, showing the system’s effectiveness for low-latency, lane-specific V2X communication.
Leave a Reply