Summary - Y.4487 (07/2024) - A functional architecture of roadside multi-sensor data fusion systems for autonomous vehicles

With the development of autonomous driving, perception methods relying solely on the vehicle's own sensors or on traditional roadside sensing systems that lack sufficient collaboration between devices are no longer sufficient to support higher-level autonomous driving applications. A higher requirement for roadside perception capabilities is therefore proposed. The roadside multi-sensor data fusion system (RMDFS) can provide new functionalities which will contribute to enhancing roadside perception capabilities by combining different types of roadside sensing devices such as cameras, lidars, millimetre wave radars, etc. according to their characteristics, and perform unified management and coordination so as to achieve accurate perception of road information, and support for autonomous driving applications.
Recommendation ITU-T Y.4487 defines a reference functional architecture of roadside multi-sensor data fusion systems. It clarifies the concept and components of the systems, and specifies the key functional entities of the systems and the reference points between the functional entities. Use cases based on roadside multi-sensor data fusion systems are also provided in the appendix.