To gain reliable and accurate perception is essential for autonomous driving in complex traffic environments, which often requires the cooperative perceptions based on on-board and roadside sensors. However, fusion of roadside sensors is rarely studied. Moreover, road-side perception is not only essential for autonomous driving, but also helpful for traffic management. In this chapter, we first give a detailed introduction on cooperative perceptions, along with core tasks and data sources of multi-sensor fusion for perception in complex traffic environments. We then propose a more reasonable division rule, which divides multi-sensor fusion methodologies into three categories including decision-level fusion, decision-feature-level fusion, and feature-level fusion. Furthermore, this chapter will give some recommendations on the forward research directions to address the problems that have been encountered in the field of multi-sensor fusion.
Multi-sensor Fusion for Perception in Complex Traffic Environments
Communication, Computation and Perception Technologies for Internet of Vehicles ; Chapter : 8 ; 147-161
2023-11-01
15 pages
Article/Chapter (Book)
Electronic Resource
English
Multi-sensor fusion , Autonomous driving , Complex traffic environments , On-board perception , Roadside perception Engineering , Computational Intelligence , Wireless and Mobile Communication , Cyber-physical systems, IoT , Professional Computing , Mechanical Engineering , Security Science and Technology
Traffic Sign Perception in Road Environments
British Library Conference Proceedings | 1994
|Data Fusion in Multi Sensor Platforms for Widearea Perception
British Library Conference Proceedings | 2006
|Traffic data fusion method and system based on multi-source perception
European Patent Office | 2025
|