Poster: Fast Field-of-View Expansion for Collaborative Object Detection

Citations

WEB OF SCIENCE

1
Citations

SCOPUS

4

초록

As interest in autonomous driving and advanced driver-assistance systems (ADAS) has grown, various sensing technologies have been developed to accurately determine the position and situation of surrounding vehicles and objects. In particular, light detection and ranging (LiDAR) sensors have attracted attention and are widely used in autonomous driving and ADAS because of their accuracy and reliability. However, when LiDAR sensors are used on a single vehicle, they can encounter blind spots caused by obstacles, which limits the detection of the environment. To overcome this issue, a method that can register and identify objects using LiDAR data from multiple vehicles in real-time is needed. Conventional artificial intelligence and iterative closest point (ICP) approaches need faster processing speed for practical use. Therefore, this work proposes an object-based single-point ICP (SP-ICP) which enables faster processing while maintaining accuracy using only a single point centered on each of the objects. © 2024 Copyright is held by the owner/author(s). Publication rights licensed to ACM.

키워드

collaborative sensing; field-of-view expansion; object detection
제목
Poster: Fast Field-of-View Expansion for Collaborative Object Detection
저자
Ryu, Junhyeong; Paek, Jeongyeup
DOI
10.1145/3643832.3661420
발행일
2024-06
유형
Proceedings Paper
저널명
MOBISYS 2024 - Proceedings of the 2024 22nd Annual International Conference on Mobile Systems, Applications and Services
페이지
684 ~ 685