The Transformational Power of Edge Analytics is Redefining Smart Cities Approaches

Subscribe To Download This Insight

By Dominique Bonte | 4Q 2020 | IN-5984

While roadside edge analytics provides real-time sensor data interpretation and emergency response locally, the edge cloud extends the reach to wider zones to capture, process, share, and use data for holistic emergency response modes. Combining the two is the foundation for the autonomous city of the future.

Registered users can unlock up to five pieces of premium content each month.

Log in or register to unlock this Insight.

 

The Rise of Roadside Edge Analytics in ITS and Smart Cities

NEWS


Artificial Intelligence (AI)-based analytics capabilities are increasingly embedded in roadside infrastructure like traffic cabinets, streetlights, and traffic and surveillance cameras, allowing local intelligence to be captured, interpreted, shared, and acted upon locally and in real time for low-latency use cases, such as pedestrian detection, traffic management, adaptive traffic lights, and advanced surveillance.

The prime example is smart traffic cameras monitoring traffic levels in real time to dynamically adapt signal phasing for optimizing traffic flow, maximizing throughput, and minimizing congestion. Advanced AI frameworks running on high-compute processors featuring hardware accelerators are capable of distinguishing between different types of traffic (passenger cars, trucks, emergency vehicles, two-wheel vehicles), a capability not available I legacy traffic sensors like magnetic loops.

A key benefit of roadside edge analytics is the ability to close the loop in real time between sensor data interpretation and emergency response; for example, in applying floodlighting following gunshot detection via audio sensors embedded into smart streetlight platforms. This represents a revolution in terms of automating and enhancing traffic management, urban safety, and security practices.  

The Emergence of the Edge Cloud: When Will It Happen and for Which Urban Use Cases?

IMPACT


But there is more to the edge analytics story. The prospect of low-latency 5G connectivity is now opening up the possibility to shift low-latency analytics to the edge of telco networks, in the form of the edge cloud, residing close to roadside systems and solutions, but still within the cellular networks of carriers.

The telco ecosystem, in partnership with cloud vendors (AWS-Verizon partnership is the flagship example), clearly considers the edge cloud as a key mechanism or paradigm to monetize expensive 5G networks, accelerating Return on Investment (ROI) and justifying wider deployments. Not only does the edge cloud bring low-latency analytics to the cloud (or conversely, brings the cloud to the edge), it also allows leveraging scalable, flexible, high-performance, heavy-lifting, and cloud-native capabilities developed and optimized by the companies like AWS and Microsoft over many years. Additionally, keeping connectivity, compute, and storage locally, avoiding long transmission paths over backbone networks, can achieve increased reliability, addressing a key concern about public networks.

Key edge cloud use cases that are centered around data crowdsourcing and sharing, collective and cooperative intelligence, emergency response, and the remote control and management of autonomous/automated systems include:

  • Automated Traffic Management Systems: Not just on a per junction basis, but holistically and collectively as in cooperative adaptive traffic lights and remote traffic management, such as remote enforcement of maximum speed based on the analysis of local traffic
  • Autonomous Assets: Examples include air traffic control for drones and remote operation of driverless delivery vehicles and robots
  • Automated Safety and Security Systems
  • Automated Emergency Response Solutions: An example is gas leak detection, alerting nearby pedestrians followed by the instantaneous and automated shutdown of gas distribution infrastructure through electronically controlled valves, avoiding explosion and casualties
  • Safety Hazards and Security Alerts: Capturing and communicating intelligence about road hazards like accidents and harsh weather events, as well as other security instances

However, edge cloud capabilities depend on the deployment of 5G networks and the installation of server equipment at (hyper) local “edge zones,” both of which will take several years to accomplish. Any significant adoption of edge cloud services for Intelligence Transportation Systems (ITS) and smart cities use cases will only materialize by 2025.

Roadside Edge versus Edge Cloud: A Battle for Controlling and Monetizing Urban Data?

RECOMMENDATIONS


While roadside edge systems mainly generate and act on data locally, the edge cloud extends the reach to wider zones in which data can be captured, processed, shared, and used for holistic emergency response modes. While Vehicle-to-Everything (V2X) solutions can extend the reach of localized analytics to around half a mile, it is no match for the ubiquitous nature of future 5G networks. In this respect, the edge cloud offers huge potential, not necessarily displacing first-line mission-critical roadside edge capabilities, but adding a wide range of additional use cases, especially as it relates to advanced real-time closed-loop responses and enabling remote control of automated systems.

Ultimately, the single most important benefit of the edge cloud, is its ability to aggregate, process, store, and distribute time-sensitive data in a flexible, scalable, and affordable way across wider geographical areas. This is where the real monetization opportunity will materialize in terms of commercializing premium real-time data in the form of data marketplaces, which will be critical for enabling next-generation smart city services and solutions. The defining value proposition of the edge cloud is offering collective/cooperative (near) real-time intelligence for advanced and comprehensive smart city solutions, the combination of which will constitute the first instances of the “autonomous city” of the future.

While a two-stage process, consisting of roadside edge analytics applied to raw sensor data for first-line alerting and response combined with higher-level remote analytics in the edge cloud for overall management and control is a realistic scenario, ultimately there is no alternative for cloud-like approaches, especially as it concerns data monetization. Roadside edge data are only valuable and usable in hyper-local contexts, while cloud edge data can be monetized more widely according to established practices and procedures. This will result in an ecosystem shift with many suppliers aiming to participate in the edge cloud data revolution.

ABI Research’s upcoming Edge Cloud Applications in Smart Cities and Intelligent Transportation report provides in-depth information about technologies, applications, and market drivers.

 

Services

Companies Mentioned