Nowadays, cars have been becoming more and more essential as means of transport in our daily life. With the unstoppable development of science and technology, self-driving cars were developed as a smart transportation solution, which can solve a lot of problems such as car accidents, congestion, parking lots, and so on. Pedestrian detection is a part of object detection, plays an important foundation in advanced driving assistance systems, especially autonomous vehicles. Pedestrian detection using LiDAR signals is has attracted great attention from researchers because of its ability to provide accurate 3D information about the environment and operate in low-light conditions, which improves disadvantages when using a Camera. Pedestrian detection for autonomous vehicles not only needs to ensure real-time perception but also to ensure high accuracy because it is directly related to safety and eventually human life. Hence, this article will summarize and analyze the currently LiDAR-based detection methods namely grid-based and point-based approaches to apply the detection of pedestrians for autonomous vehicles. Then, we analyze and compare the results between the highly regarded algorithms using this approach and their applicability to autonomous vehicles, target finally is pedestrian. The results show us that there are advantages and disadvantages of the two grid-based and point-based approaches. The point-based approach stores information better, but requires higher computational resources than the grid-based method. And the combination of both methods yields an overall improvement in detection accuracy. Finally, we look at a few challenges that need to be overcome for greater detection accuracy.
An Overview of Pedestrian Detection Based on LiDAR for Advanced Driving Assistance System
Lect.Notes Mechanical Engineering
Regional Conference in Mechanical Manufacturing Engineering ; 2021 ; Hanoi, Vietnam December 10, 2021 - December 12, 2021
The AUN/SEED-Net Joint Regional Conference in Transportation, Energy, and Mechanical Manufacturing Engineering ; Kapitel : 28 ; 352-363
2022-06-01
12 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Advanced driving assistance system and advanced driving assistance method
Europäisches Patentamt | 2021
|Pedestrian Detection System using YOLOv5 for Advanced Driver Assistance System (ADAS)
DOAJ | 2023
|