Advancements in IT, including communications, sensing, data processing and control, are transforming transportation systems from traditional technology-driven systems to more powerful data-driven intelligent transportation systems.
Vehicles, potentially, will support a host of new applications that integrate fully autonomous vehicles, IoT and the environment, making the Internet of Vehicles, or IoV, the standard. The latest developments in communications and networking will enable vehicles to utilize resources such as cloud storage and computing with IoV.
Scientists from WIMI Hologram Academy of WIMI Hologram Cloud have worked to improve the performance of smart vehicles by deploying storage and compute resources at the edge of wireless networks.
Smart vehicles are equipped with multiple cameras and sensors, including radar, LiDAR sensors, sonar and global navigation satellite systems. In the future, more than 200 sensors are expected to be within a single vehicle, and the total sensor bandwidth will reach up to 40Gb/s.
The big data generated by the network will put a strain on communications, storage and computing infrastructure. Cloud computing solutions, such as cloud-based software updates or training deep learning models, have been implemented to help smart vehicles, but they are not enough. Cost and power consumption remain constraints for in-vehicle computing, while long latency and massive data transfer are bottlenecks for cloud processing.
Mobile edge computing is an emerging technology that has potential to combine telecommunications with cloud computing to deliver cloud services directly from the edge of the network and support latency-critical mobile applications.
And with the rising interest in AI, edge AI is enabled by edge caching and computing platforms, which train and deploy machine learning models on edge servers and mobile devices.
Edge Information System, which includes edge computing and edge AI, will play an important role in the future of intelligent vehicle networking. For smart vehicles, EIS will assist with data acquisition for situational and environmental awareness, data processing for navigation and path planning.
With on-board sensors, EIS will improve perception capabilities. It improves the sensing accuracy of cameras and stereo vision through deep learning techniques and enables complex multi-sensor fusion by shifting computationally intensive sub-tasks to edge servers.
EIS will be important for smart vehicle HD mapping. Edge caching helps with HD map dissemination and map data aggregation. Edge computing helps with map construction and map change detection. Edge servers also coordinate vehicles for crowd-sourced mapping through the region. With this, more efficient HD mapping is achieved by storing and processing data locally and constructing maps where they are needed.
EIS will also assist with simultaneous localization and mapping, or SLAM. The computational demands of SLAM for autonomous driving are intensive. For example, one hour of drive time generates 1Tb of data, while interpretation of 1Tb of collected data using high computational power will require two days to obtain usable navigation data.
Although cloud-based SLAM algorithms were proposed to reduce the computational burden on vehicles, their propagation latency cannot meet the real-time execution requirements. Edge computing platforms will solve that challenge by handling some of the computationally intensive subroutines. Multi-vehicle SLAM with inter-vehicle cooperation also helps to improve the performance of SLAM.
The upcoming smart IoV will require support from the automotive, transportation, wireless communications and networking domains, among others, but processing data at the edge of the network will save significant communication bandwidth and meet the low latency requirements for vehicle critical tasks.
Edited by
Erik Linask
