Connected intersections a key factor for smart mobility

If road mobility is to become truly ‘smart’, with optimal traffic flows and a high degree of safety, connected intersections would appear to be a must.

In the not-too-distant future, Intelligent Transportation Systems (ITS) will form a digital ecosystem whose task will be to ensure optimal vehicle flow management. Using advanced functionality in the fields of observation, analysis, memory and calculation, these systems will be able to process the information they receive in real time and, with their not only highly responsive but even predictive capabilities, will coordinate transport and traffic infrastructure for an entire city. They will also communicate with autonomous vehicles so as to warn them of any impending danger. In this configuration, connected road intersections will form key links in the smart mobility chain. The result should be not only much-improved local traffic flows and an end to traffic jams but also close-to-zero accidents.

Improving safety

In the near future, journeys across our Smart Cities will be managed by an intelligent system that will be able to take account of each individual vehicle. The system will comprise a comprehensive panoply of sensors placed at each road junction, including cameras, radar and even lidar systems that enable remote scanning using lasers which communicate with each other and with drivers. The sensors will keep track of all road users – whatever means of transport they are using – travelling in a defined zone. They will build up a set of data, including the type of vehicle, the speed at which it is travelling, and its position and trajectory, which help to create a complete 360° model of the environment. This information, analysed by algorithms, will then be relayed to all vehicles approaching an intersection or travelling in a high-risk zone, using short- or medium-range wireless communication channels specially designed for use in a car, known as Dedicated Short-Range Communications (DSRC) systems. The onboard DSRC will then sift the data it receives, check its relevance, and take avoidance measures or protective action if a critical situation appears to be brewing. Just as a chess player looks at all the pieces on the board and assesses a wide range of possible moves and combinations, the system’s algorithms will be geared up to calculate all the consequences for traffic flow if one vehicle turns left rather than right and another stops at a red light, while a bus is coming on a bit too fast in the opposite direction.

The predictive model also takes account of the most vulnerable road users – pedestrians and cyclists – who are usually victims of accidents but may also cause them. The anticipation capability provided by this technology, which can even take into account weather conditions, can achieve real benefits in terms of safety, producing very precise, sharp, real-time visualisations of traffic and of all the problems that might arise. “Active surveillance of intersections gives a driver, or self-driving vehicle, far more time to act in cases where the driver cannot actually see the problem,” underlines Jeremy McClain, Director, Systems & Technology North America, at German automotive components and systems manufacturer Continental AG.

And this is without taking account of moves to adapt the actual vehicles to these new traffic systems. Continental, a firm which is working at the cutting edge when it comes to future mobility and smart vehicle equipment, carried out some key tests last October on its test track at Brimley in the US state of Michigan, measuring the performance of the new V2X (‘Vehicle to Everything’) sensors developed by their R&D teams. V2X technology is designed to pass information from a vehicle to any entity that may interact with the vehicle, and vice-versa, incorporating a range of specific types of communication such as V2V (Vehicle to Vehicle), V2P (Vehicle to Pedestrian), and V2G (Vehicle to Grid). Using this onboard technology, vehicles can communicate with each other and also with an ITS or connected intersection. In the city of Colombus, Ohio, Continental is now preparing to trial its on-board technology so as to verify – in actual traffic conditions and linked up to a connected road junction – that the systems are able to communicate directly with each other and thus together fulfil their task of protecting all kinds of road users, from car drivers to cyclists and pedestrians.

Optimal traffic management

The other major purpose of connected intersections – whose results have actually taken a significant lead over parallel efforts in the safety field – is to streamline traffic flows and put an end to traffic jams. For a good number of years now, automated cameras have been filming at reputedly dangerous road junctions, enabling the authorities to proceed against drivers who infringe the rules of the road. This has in turn had a demonstrably positive effect on traffic flows. By modelling all the traffic in real time, the surveillance cameras – which are a lot smarter nowadays than they used to be – are able to take a series of decisions in order to prevent traffic jams from building up. This might simply mean increasing the amount of time a traffic light stays at red, actually telling cars to slow down, or suggesting that drivers take an alternative route. In this way, the connected intersections can improve the overall management of traffic flows.

There are however other innovative approaches in existence which go even further. In Canada, Dr Samah El-Tantawy, a Post-Doctoral Fellow in the Department of Civil Engineering at the University of Toronto, has developed a coordinated traffic signal control system drawing on game theory concepts and a Multi-Agent Reinforcement Learning Integrated Network (MARLIN). Marlin uses artificial intelligence and machine learning to optimise traffic light sequences. Instead of having the traffic lights programmed in a rigid, centralised fashion, Marlin enables them to communicate and coordinate among themselves. Sensors in the ground linked to a communication system relays data to them in real time. Marlin decides on a second-by-second basis what should happen, both for its own traffic sector and as part of the wider traffic environment, depending on information coming from other traffic lights.

During the first trial carried out on sixty sets of traffic lights in Toronto, Marlin reduced traffic jams by 40%, including during rush hours when the traffic is particularly dense, and succeeded in cutting journey times by 26%. This excellent result is only the start, given the current boom in innovations linked to intelligent traffic lights. Bill Valle, Director of Engineering & Capital Projects/City Engineer for the city of Chula Vista, near San Diego in California, has no doubt that “deploying systems for smart intersections that quickly respond to traffic patterns will improve mobility in our community and is a vital component of the long-term strategy to lay a foundation for smart cities infrastructure.”

Chula Vista, a city of 200,000 inhabitants located in southern California, has just decided to roll out on a large scale an ambitious system of ‘smart intersections’ designed to adapt to traffic patterns in real time. In the first quarter of 2018, SynchroGreen™ an adaptive traffic control system developed by Texas-based company Trafficware, which specialises in optimising car and pedestrian flows, will deliver data from 28 road junctions along Chula Vista’s three main arteries. SynchroGreen is able to take dynamic synchronisation decisions in real time. This is intended to be a win-win initiative at all levels. The Chula Vista municipal authorities have stated that the aim of this project is not only to achieve a substantial reduction in travelling time across the city, plus associated safety improvements, but also to make Chula Vista a more environmentally-friendly town by reducing greenhouse gas emissions through traffic optimisation.

All these innovations are clearly moving towards Smart Cities that are organised to take mobility into account. However, some degree of caution needs to be exercised. Disseminating private data on cars and their drivers left, right and centre will require specific regulation and a ‘secure by design’ framework if abuses are to be avoided. Moreover, appropriate safeguard systems will have to be put in place to defend against hacker attacks. Debate has been underway for some time regarding the vulnerability of connected objects and the need to protect the data they produce, and this will certainly be one of the main issues surrounding the rollout of these revolutionary ‘Smart Mobility’ tools.