A smart car is a vehicle using AI to connect to the network, and eventually to drive itself. A smart road is an instrumented road network, built as part of the same programme as the smart city.
Policy support and technology have given the transport sector an opening it has not had before, and it is moving toward digitisation, networking and intelligence at speed. City administrators and residents both feel the difference when traffic management gets smarter and the service gets warmer.
The question is how AI can be applied to transport, and beyond it to city governance. The argument from the research side is that smart cars plus smart roads — an instrumented urban road network — is the effective way to solve the traffic problem.
What AI has already done for urban traffic
AI and the technologies around it are being fused into transport, and they give city administrators new ways to handle the conditions that make large cities difficult: parking that is scarce and disorderly, management that is hard, congestion that will not clear.
Parking is the clearest case. Beijing began reforming on-street parking in 2019, and by 2021 the reform had been running three years. It established the principle that a car parks in a bay, pays for the bay, and is fined if it does not, and it produced measurable results: by January 2022 the city had marked out more than 90,000 bays on over 1,000 streets, all under electronic charging, with a compliance rate above 90% and 244 million parking sessions served over the three years.
Behind the electronic charging is a technology change. High-position video — cameras mounted above the kerb rather than at ground level — captures parking information and handles the charge automatically, which is what allows one standard, one technology and one platform across the whole road network. Coverage now extends across all 16 districts and the Beijing Economic-Technological Development Area.
What the data shows
AI enables a full or multimodal perception of traffic. Accumulated data and data-driven methods build a macroscopic picture at district or city level. Microscopically, a single junction can look chaotic, and an accident can look unpredictable. Macroscopically — over a larger span of time and space, a district’s whole day of traffic history, for instance — there are patterns, and AI can find them by modelling, mining and analysing the data, which is how the relationship between the aggregate behaviour and the local detail gets established.
That matters because it runs in both directions. A macroscopic picture can guide the treatment of the microscopic: the traffic around a school, a hospital, a shopping district, a scenic area or a residential neighbourhood can be managed as a system rather than junction by junction.
Which is the shape of the opportunity. With smart transport infrastructure underneath, AI can work bottom-up and top-down at the same time, building a district-level and then city-level transport system, and it can do it while the commercial deployment of connected vehicles scales up. The two have to be held together — the aggregate and the detail, each informing the other — because the alternative is a city that manages its junctions well and its traffic badly. The precondition is interoperability: people, vehicles and roads speaking the same protocol, under the same standards, on infrastructure that was built for it.
Republished for information. The original text has not been independently verified and does not necessarily reflect our views.
Want to talk through what you just read?
We work end to end, from product design to launch. Tell us about your scenario and we can start there.
Shenzhen XZS Technology Co., Ltd. · Tel 0755-82494862 · 13760424861