Although traffic is a challenge in many cities across the globe, it is a challenge that can be managed and solved. Governments and city officials should work with the private sector to deploy intelligent solutions that not only improve the present state of roads and highways, but also prepare for the future.
Some of the solutions to traffic challenges lie in using artificial intelligence (AI), data collection and analytics, and sophisticated hardware that automates and streamlines processes to achieve high levels of accuracy, safety, and reliability.
Coupled with existing systems comprised of network cameras and monitoring solutions, the innovation and application of AI heralds a new era in traffic management.
Cities are getting bigger and smarter
Globally, our cities are growing and becoming more populous. The migration from rural to urban areas, combined with overall population growth, will see another 2.5 billion people living in cities by 2050, with 90% of this increase taking place in Africa and Asia, as per a United Nations report. This means more people travelling on more roads, and without smart interventions, this could mean more traffic.
In some areas, with this migration comes the emergence of smart cities – urban areas that use ICT, AI, and data-related processes to maximise operations and investment.
In Abu Dhabi and Dubai – two cities rated 28th and 29th respectively on the IMD Smart City Index 2021 – traffic congestion is not considered a problem among residents. Such a result is, in part, thanks to the adoption of AI processes to deal with traffic and road management.
Managing traffic and the challenges thereof
Traffic management centres (TMCs) are the central nervous systems used for monitoring and managing traffic in an area. Imagine sprawling rooms with hundreds of surveillance cameras watching intersections, highways, and other road infrastructure for congestion, accidents, and the like to allow for quick-time responses from officials.
However, traditional TMC setups only allow for a certain level of monitoring and oversight.
Intelligent solutions, comprising both hardware and software, can fill these gaps.
Quality hardware is the first step
The strength and effectiveness of a TMC is indicated by its ability to collect data, and the application of AI requires up-to-date and robust hardware for it to achieve its full capability.
Centres use an extensive network of video management systems (VMSs) to monitor infrastructure and capture incidents in real time. This includes a range of fixed-point, panoramic, and thermal cameras – hardware that’s built to perform regardless of factors such as environment, object layout, and time of day.
The application of AI
To address traffic challenges, hardware is coupled with advancements in machine learning (ML), which focuses on gathering and analysing data to identify and replicate patterns in behaviour. By design, roads are ideal for this process as users conform to its rules, producing data which can be used to track deviations and create optimised models.
Deep learning (DL) takes this a step further. A sub-field of ML, DL uses raw data to automatically determine features that distinguish different sets of data from one another. The processing of data in this way does not need human intervention.
What does this look like in the context of traffic management? It automates several processes and decreases the necessity for human intervention or oversight. Data can be used to identify weak or exposed points in infrastructure or throughout the overall scope of surveillance. Roads are made safer through quicker response times from emergency services and officials.
Edge computing and edge-based analytics
Cameras capture a lot of data. A network of cameras tied to a VMS captures even more data, and with this comes the necessity of cloud or server-based data storage and processing. Edge computing offers an alternative, more effective solution to constantly transfer data from devices to the cloud.
With edge computing, captured data is analysed closer to its point of origin rather than having to go back-and-forth between servers and central data points.[1] The localisation of this process results in less dependency on other infrastructure and less time is lost due to factors such as network latency.
Urban populations may continue to grow, but traffic problems do not need to accompany this growth. Leaders can harness the power of AI combined with smart surveillance hardware as we build the cities of the future. Cities and urban areas stand to benefit from the application of AI in this manner as we head into a “smart” future together.