In recent years, much has been written about Connected and Autonomous Vehicles (CAVs). Generally, these texts focus on the autonomous component, which is understandable given its futuristic and disruptive nature, not to mention all the ethical and legal issues it raises. However, the connected component is of equal or even greater importance.
Use cases for connected vehicles
A connected vehicle can communicate with other vehicles, road network infrastructure, and vulnerable road users such as pedestrians and cyclists. Thus, thanks to communications, a connected vehicle will be able to know when a pedestrian is behind an obstacle, out of reach of its sensors or the driver's view, what the status of a traffic light is that is out of its field of vision; or even extend the range of its sensors by receiving information about what the vehicle in front or behind it detects.
In addition, a connected vehicle can access services available through the network. An example of this is the work that the CCG/ZGDV produced as part of the PAC project, in which a service running on a remote server or an edge computer (both tested) was able to predict a possible collision between a connected vehicle and a pedestrian and warn the latter so that he could take the necessary steps to avoid an accident. Another example, which is being actively researched, is the use of the movement history of connected vehicles and artificial intelligence (AI) to predict the state of traffic shortly and use connected infrastructure to redirect traffic, avoiding traffic jams and other constraints.
I would be remiss if I didn't also mention "platooning," i.e., the "ad hoc" grouping of vehicles into virtual, dynamic platoons, in which the vehicles behind follow, via vehicular communications, the vehicle in front (autonomous or not) and communicate with each other instantly about changes in direction, speed or attitude. This allows for a substantial reduction in the space between vehicles since the reaction time of a human driver doesn't have to be taken into account, which in turn reduces the area of road use, among other advantages.
Of course, the examples above are just a tiny sample of the possibilities of connected vehicles. Ultimately, the technology could even be used for remote driving, either permanently or to take control of the vehicle if the autonomous driving system fails or the driver becomes incapacitated.
The obvious consequences of these applications are an increase in safety for all road users, as well as an increase in the efficiency of traffic management. The latter, in turn, will lead to shorter journey times and reduced traffic pollution.
Applied technologies
Various technologies, some more established than others, allow connected vehicles to operate en masse.
Of these, the most talked about today is undoubtedly 5G. High bandwidth and low latency make this technology particularly suitable for this use case, which requires communications closer to real-time.
Perhaps the most tested technology for this purpose is so-called vehicular Wi-Fi, supported by the IEEE 802.11p standards and the more recent 802.11bd (associated with Wi-Fi 6). This technology is often used for communication between vehicles and infrastructure, where the connection between the players is short-lived and highly variable.
Although still in its infancy, another technology whose applicability to vehicle communication issues has been the subject of scientific articles is communication using visible light (generally referred to as VLC). This technology uses light modulation to communicate without being perceptible to the human eye. The light source can be anything, such as a vehicle's headlights or a traffic light, which allows you to partly reuse already installed equipment or a light source installed specifically for this purpose (e.g., LED).
Overcoming difficulties
These technologies and the advantages they bring come at a cost.
They all involve the installation of specialized hardware and software in vehicles, increasing their production and maintenance costs. It's true that many manufacturers now include communication capabilities in their vehicles that allow, for example, the updating of internal software "Over the Air" (OTA) or preventive maintenance of vehicles. Still, these are often proprietary solutions that, in terms of hardware or software, are not necessarily compatible with the needs of vehicle communication. To this can be added the additional cost of communications, such as subscribing to a 5G service.
Another substantial cost will be infrastructure. As far as the telecoms infrastructure is concerned, it will have to be reinforced to support the additional load of connected vehicles. Moreover, additional elements may be required depending on the technology used. In the case of 5G (and the future 6G), installing "small cells" to extend network coverage is necessary. On the other hand, in the case of in-vehicle Wi-Fi, for vehicles to communicate, for example, with roadside equipment, the installation of access points is essential.
Of course, to make the most of connected vehicles, the current infrastructure (such as traffic lights) will have to be updated to transmit their status and changes over the network. In addition, installing new equipment, such as pedestrian radars at low-visibility crossing points, will be an important asset in reducing road accidents.
Finally, we need computing power to realize some of the possibilities that a system of connected vehicles and infrastructure allows us. For example, a collision prediction system will need equipment that constantly receives information about all the players on the roads and calculates possible future collisions. This equipment could be at a centralized point, but depending on the use case, the latencies involved in communications could be too high. In these cases, it may be convenient to do the processing closer to the vehicles and lanes to reduce latency, i.e., to do the processing using edge computing and fog computing. Of course, this implies more investment in infrastructure.
Another problem is the lack of standards in vehicular communication. The frequency ranges reserved for this purpose change from country to country and territory to territory. In addition, the format and other requirements of the messages exchanged between vehicles and other connected parties are not uniform. The European Telecommunications Standards Institute (ETSI) has created its standards in Europe, but non-member countries can also create their own, with a greater or lesser degree of compatibility.
Today, you can't talk about connectivity without referring to safety. Although connected vehicles generally use a private network and are not connected to the public Internet, the possibility that a malicious hacker could influence the vehicle's or its driver's behavior cannot be ruled out. We're not just talking about the obvious and potentially catastrophic possibility of a vehicle being entirely remotely controlled by a third party. Still, the subtle manipulation of information is passed on to the vehicle to provoke a specific behavior.
In addition, it is necessary to ensure that those who maintain the vehicle have the proven expertise to deal with the particularities of a connected vehicle. Therefore, it is necessary to limit the changes that vehicle owners can make to ensure that they do not endanger other road users.
Conclusions
Connected vehicles are rapidly proving essential in solving problems related to congestion, car accidents, and pollution. Several technologies are already available to support them. However, before they become a common sight on our roads, several hurdles must be overcome, such as the rising cost of vehicles, the installation of connected infrastructure on the roads, and the possible safety problems in their use. Given their importance in building a future in which mobility must be smarter, we must work towards this.
By: Joel Puga Researcher in the Urban and Mobile Computing (UMC) area of the CCG/ZGDV Institute
