Smart Parking Leads the way in Infrastructure Management

Smart traffic management solutions powered by V2X communications and artificial intelligence have the potential to significantly reduce congestion and emissions.

However, successful implementation will require coordination between cities, automakers and others in the automotive supply chain. A recently published Juniper Research report found that by 2025, smart traffic management systems will save cities $277Bn. Adam Wears, research analyst and author of the report, predicted “massive growth” in the implementation of smart traffic management systems. “As automakers begin to rollout two-way communication between vehicles and roadside infrastructure as well as other vehicles, V2X can be used for everything from real-time signal actuation to warning about upcoming congestion and hazards to enabling the prioritization of emergency service vehicles at traffic signals,” Wears noted.

He said V2X connectivity will enable smart traffic management platforms to gather data directly from vehicles, rather than relying on traditional traffic actuation methods. “While automakers can easily develop and install V2X systems in their vehicles, these won’t be anywhere near as efficient as they can be without equal take-up of compatible systems by cities at intersections, freeways, and parking facilities,” Wears explained.

Sam Abuelsamid, principal research analyst leading Guidehouse Insights, noted an ideal V2X system would provide guidance to the vehicle the way as Audi has done in North America with traffic signal systems. Audi provides drivers with information on the timing of the lights, such as when a light is going to change, along with guidance like the optimal speed to maintain in order to hit a string of green lights. “That sort of thing can be very helpful,” he said. “As you tie in transportation modes like busses or emergency vehicles, and micro mobility, you can make your entire traffic flow much smoother.”

He pointed to pilot programs in Ann Arbor, Michigan, and Tampa, Florida, as evidence that smart traffic management systems are gaining some traction but said it will take far wider adoption before the benefits are seen. Technologies like AI and machine learning will also be essential components of smart traffic management solutions and can primarily be deployed in smart roadside infrastructure to enable these systems to process, analyze and make actuation decisions locally without needing to transmit traffic data to a centralized system. “In terms of smart intersections, we’re also seeing these technologies increasingly deployed in video cameras as a means of ‘recognizing’ approaching vehicles in poor visibility conditions that, ordinarily, impede video detection systems,” Wears said.

Machine learning

Abuelsamid said machine learning algorithms can also help the systems better understand patterns of traffic, such as being able to identify busses or trucks, which accelerate more slowly and are less responsive. “Understanding what types of users are on the road various times of the day can also help to improve the overall control strategy,” he said.

However, the most important smart intersection technologies are visual sensors such as video cameras and radar units that enable traffic signals to react in real-time to traffic conditions, rather than reacting based solely on pre-programmed timings. “Cellular technologies such as NB-IoT are well-suited to powering smart intersections and will remain the dominant technology for at least the next several years but 5G will be essential to ensuring the wider adoption of smart intersections around the world,” Wears said.

That adoption is likely to be a long time coming, Abuelsamid said, perhaps a decade or more, noting the high cost of investment into existing traffic management infrastructure. “Cities have limited resources, and there’s a lot of existing traffic management infrastructure that, for better or worse, works,” he said. “If something isn’t broken, a city is going to have a hard time justifying potentially billions of dollars updating its entire system. It’s a process that’s going to take a long time to get implemented on a wide scale and will be very gradual as it happens.”

The Juniper report also highlighted accelerating investment in smart parking, which is projected to reach $1Bn by 2025, up from $460M in 2021. Parking garages purpose built for CAVs are already in planning stages, as seen by the Trusted Autonomous Parking (Park-IT) project, a joint effort between engineering firm Horiba Mira and Coventry University. Smart parking includes technologies that help facilitate information between parking sensors, end users, management platforms and vendors, including smart displays, wireless communications and notifications that provide real-time availability. Abuelsamid said for urban centers, one of the big traffic challenges is people just trying to driving around looking for a space to park.

Understanding where there is parking available, enabling drivers to reserve a parking spot and pre-pay for it, or inform them before they arrive to go directly here to park, are the types of smart parking applications that could be “enormously beneficial” in getting those drivers off the road as quickly as possible. “This is also where a machine learning model can be very helpful in building up a predictive model of when a parking spot will be available,” he said. “Using those cloud-based V2X communications systems could provide an indication to the driver based on their destination for the probability of a street parking spot when they arrive.”

He also pointed to technologies like automated valet parking systems, where the driver pulls into a garage, and the car goes and parks itself with guidance from that facility. Automakers like Ford have already announced plans to start testing automated car parking technology from Bosch, while Chinese automaker Human Horizons claims to have developed the world’s first Level 4 autonomous parking technology on its premium sub-brand’s HiPhi X model.