[ad_1]
The implementation of innovative services for the road transport of the future requires removing several technological obstacles. Thus, the autonomy of vehicles will only be achieved at the cost of developing a connected, reliable, resilient infrastructure capable of transmitting information with high throughput and low latency.
The technological improvements necessary to see developments on our roads driverless cars are numerous and Europe subsidizes numerous projects, 148 in total, to move towards this objective.
Among them, InDiD is one of the 13 French projects selected by the European Commission as part of the latest call for projects of the Connecting Europe Facility (MIE). It is part of the continuation of the projects of Cooperative Intelligent Systems and Transportationles C-TIS.
Security, infrastructure, connectivity… The challenges are therefore numerous and the InDiD project will make its contribution to this immense edifice.
Toufik Ahmed, Deputy Director ofENSEIRB-MATMECA – Bordeaux INPand researcher at BRIexplained to Techniques de l’Ingénieur the specificities of the InDiD project, which brings together 24 actors at the institutional and research level.
Engineering Techniques: Can you present to us the InDid project, which has been extended by six months and which will end at the beginning of 2024?
Toufik Ahmed : The leader of this project is the Ministry of Transition Ecological and Solidarity, which coordinates the 24 players (industrialists, communities, highway companies, research laboratories, institutional partners), of which Bordeaux INP is a part. The InDID project focuses on the development of the digital road infrastructure of tomorrow in addition to physical infrastructure, to promote innovative transport services and more particularly Cooperative Intelligent Transport Systems (C-TIS).
What issues are you working on?
The issues addressed concern improving connectivity on digital road infrastructure and network infrastructure (core of the network). Namely how to guarantee an efficient exchange of information between vehicles, but also between the vehicle and the digital infrastructure.
In this context, our work focuses on two aspects, on the one hand technological hybridization and on the other hand data processing in the digital infrastructure.
On the first part, the emphasis is placed on the contribution of 5G to promote innovative services on C-TIS vehicles. For example, services such as autonomous vehicle support or platooning, which consists of a convoy of vehicles guided by a single person. Platooning is mainly aimed at the transport of goods.
The ambition of the InDiD project as a whole is to continue to develop C-TIS services through the project’s pilot sites and to experiment with new services.
The second part aims to understand how to bring services closer to the consumer and the service provider. On this point, we investigated the deployment of data processing infrastructures at the edge of the network, called Edge Computing. Strategically located in key locations on roads and highways, this infrastructure aims to run services as close as possible to the user, thereby significantly reducing latency and improving performance.
Let’s come back to technological hybridization. What experiments have you implemented?
Several technologies exist beyond 4G/5G cellular networks, including ITS-G5, the European C-ITS technology based on wifi, which allows an ad hoc connection between several vehicles. We have found that this technology alone cannot provide sufficiently high throughput. We therefore worked on improving this technology, which made it possible to obtain better results in terms of reliability, throughput and error rates. However, despite these advances, services such as platooning or remote driving cannot be fully provided by these technologies when considered in isolation, whether it is ITS-G5 or even 5G.
The guiding idea of our work was therefore to hybridize 5G and ITS-G5, in order to benefit from the advantages of each technology and obtain optimal results in terms of throughput and latency time, while allowing several vehicles to access the radio interface simultaneously.
What results have you obtained?
Hybridization made it possible to obtain an improvement of around 30% in the packet reception rate, and an improvement in throughput of 20%, and a reduction in the channel occupancy rate of around 10 %.
These improvements are sufficient today to support the first phase of deployment of C-TIS, which mainly concerns information services: road congestion, GLOSA type services (optimal speed for crossing a traffic light). These services concern the first phase of deployment, which is recorded and deployed today.
Tell us about phases 2 and 3, which should follow this first phase?
The second phase brings together improvements relating to the perception of the vehicle’s environment and the exchange of information between vehicles. Finally, phase three ensures automated support for autonomous vehicles, which includes collaborative perception, and the capacity for vehicle to enter efficiently and therefore safely on roads frequented by other vehicles.
The latencies required for this type of phase 3 service, or for platooning for example, are of the order of milliseconds, which is today inaccessible, even with the hybridization between 5G and ITS-5G . We will have to wait for improvements in the radio interface to have more throughput and less latency.
When it comes to data processing at the edge of the digital infrastructure, what results have you achieved, and what are the remaining obstacles?
Currently, everything is centralized. If we look at the road infrastructure in France, there is a national data center, in Paris, which is used for all services. Which obviously induces a significant latency time. The idea here is therefore to deploy mini-data centers, called Edges, to carry out calculations and decision-making locally, and not on the national node.
The objective is therefore to deploy nodes at the local level, while allowing the service to follow the user once the latter moves away spatially from a node, this is what we call service continuity. . This is how we manage to significantly reduce latency, and therefore improve service performance.
How did you experience this?
We experimented with this at an unmarked toll barrier during the experimentation period by the APRR, at which we deployed our ITS-G5 antennas and the Edges. We then simulated a connected autonomous vehicle to experiment with the calculation speed and latency time between it and the local data center: of course, the quality of service obtained is much better than using a much more distant national data center of the vehicle: in this case the vehicle had to choose the most relevant toll lane to make traffic flow as smoothly as possible.
Many projects are currently being carried out at European level on the various issues inherent to the development of autonomous and connected vehicles. How can we pool all the progress made on these different projects?
You should know that industry and research stakeholders are working on several projects simultaneously, at national and European level. This makes it possible to pool knowledge, but also to work on a very important aspect essential to the deployment of the first phase that we mentioned, which is the interoperability of services from one country to another. Today, the deployment of digital road infrastructure throughout the territory will ultimately ensure better coverage of phase 1 services, before moving on to phases 2 then 3.
Comments collected by Pierre Thouverez
[ad_2]
Source link