1. Connected vehicle technology – everything you need to know in 2018

Connected vehicle technology – everything you need to know in 2018

Spring 2018 has been a busy and exciting season for me. I am happy for the chance to step out of the lab to share what we are working on at Ericsson with audiences at a number of events, including 5GAA, Mobile World Congress and the 2018 Geneva International Motor Show. When I visit such events, I get the opportunity to speak with people working for automotive companies, government regulators, city planners, telecom operators, app developers and service providers.

One thing all these people have in common is the hope that emerging technologies will soon make transportation safer, greener and more enjoyable. But there’s still a long ahead road before we get there. Right now, there is a lot of hard work going into deciding how next-generation connected vehicles, Intelligent Transport Systems (ITS), and communications networks will all work together.

To make our jobs easier and more profitable, we all need to speak the same language. I don’t mean choosing between Italian, English or Swedish (as I often do). Rather, I believe all players can benefit from a basic understanding of the technologies that are making smart intelligent transport and connected autonomous vehicles (CAVs) a reality. So, in this post, I will briefly explain some of the key pieces in the next-gen connected driving puzzle.

From onboard sensors to dynamic HD 3D maps

HD 3D mapping

As vehicles grow more autonomous, they must rapidly and accurately map their environment. One of the technologies that makes that possible is LIDAR, a sort of “light-radar” system which emits pulsating laser signals to survey a landscape and measure distance at the speed of light.

In combination with radar, inertial measurement unit (IMU) sensors and cameras mounted on the car LIDAR is already being used in advanced driver assistance systems (ADAS) to help navigate cars. However, there are a few drawbacks to LIDAR, the first of which is the price point, using high-quality LIDAR sensors for autonomous vehicles is still an expensive solution. Secondly, similar to cameras, LIDAR sensors are limited in range, and cannot see through or around obstacles.

Creating a live and accurate digital representation of the environment is one of the key enablers of autonomous driving. To extend the digital horizon of autonomous cars, the limitations of cameras and onboard sensors can be overcome by connecting autonomous cars to each other by use of cellular networks. When real time data from surrounding vehicles, road infrastructure, smartphones and sensors is combined, the result is an up to date high-definition, dynamic 3D map.

Distributed computing scales up the ecosystem

The model of sharing data loads across components and vehicles is known as distributed computing. To speed up data communication make computational platforms scalable, and optimize the way information flows across networks, distributed computing will be essential.

In the long term, it will even be possible to offload some of the in-vehicle processing to a more efficient and upgradable remote cloud platform. In order to progressively improve the performance of connected vehicles, carmakers will be able to monitor how their vehicles behave in a real traffic environment, apply machine-learning, and continuously implement software updates to make the necessary changes.

To further improve road safety, it is also possible for the network cloud to monitor the behavior of vehicles on the road, automatically detecting anomalies, as well as potential safety and security threats.  To avoid the disclosure of any potentially sensitive personal data, monitoring will be based on pseudo-identities and, when possible, anonymous aggregated data.

5G is needed for autonomous vehicles

Today, there are some great connected car services available over 4G LTE networks such as telematics, infotainment (passenger entertainment) and remote-control functions to start the car or unlock doors. But autonomous cars will need the near-real time data communications that only 5G mobile networks can provide.

Connected autonomous cars will generate an explosive amount of data. Data transmission volume is expected to grow exponentially in the coming years, making these huge data transfers more affordable will require 5G network optimizations.

The first 5G standards were agreed upon in December 2017 by the 3GPP, with the next set due in June 2018. This means we can start taking real steps to building 5G networks that will allow vehicles and road infrastructure to be part of a scalable IoT ecosystem.

Virtualized networks and network slicing get things rolling

For telecom operators, the process of updating their core network to 5G will be an evolution. One way that networks will evolve is using Network Functions Virtualization (NFV), which replaces physical network functions with Software Defined Networking (SDN).

NFV allows network functions to be distributed, focusing computational power where and when it is needed most. In relation to connected autonomous cars (CAVs), NFV will allow 5G connectivity for connected vehicles to be focused on the services and locations that need it most.

NFV will helps enable network slicing, which allows multiple logical networks to be created on top of a common shared physical infrastructure. This means a dedicated and secure “slice” of a 5G network can be created for a specific function. Network slicing will be essential to ensure self-driving vehicles get access to the critical data they need to operate safely.

Predictive mobility makes the best of available connectivity

Predictive Mobility

As I mentioned, it may still be some time before 5G data communication is available to the same extent that LTE is today. But we don’t have to wait for total 5G rollout to get autonomous cars on the road.

Using predictive mobility, CAVs services and applications will be able to adapt to predicted network performance. For example, big data transfers from the cloud such as over the air (OTA) security updates, mission-critical data, and even the kids’ videos can be anticipated or delayed depending on predicted network performance. This way, CAVs can still take long journeys, leveraging 5G whenever available and providing a uniform and smooth user experience.

An ITS ecosystem built on knowledge sharing

Did you get all that? Don’t worry, there won’t be a test (today). But hopefully you can use some of this knowledge to impress your friends and family, or at least some of your colleagues.
It is only through communication and cooperation that we will advance IoT and ITS. No matter what sector you work in, there’s a place for you in this ecosystem, where we are creating a common language to form new, cooperative working models.

Watch this video to see how all of these technologies will come together to help connected vehicles move safely through smart cities.

If you are attending the Geneva International Motor Show, please check out my panel on 5G for autonomous vehicles including current possibilities with 4G, as part of the Symposium on the Future Networked Car 2018. I will be on stage on March 8 at 12.00CET.

Find out more about how Ericsson is helping to capture the full potential of connected vehicles with our connected vehicle solutions here.

Written by Stefano Sorrentino

Stefano has been with Ericsson for ten years and currently coordinates automotive research. He is also Board Member and Chairman for the System Architecture and Solution Development group in the 5G Automotive Association (5GAA). Stefano has 6 years of 3GPP experience as a delegate, covering a range of topics including MIMO, Coordinated Multipoint Processing (CoMP), Sidelink/D2D and V2X. Stefano was also appointed to chair RAN1 ad-hoc sessions on topics related to Downlink CoMP for about one year during LTE Rel-11. He has attended numerous international conferences as author, panelist or keynote speaker, and has contributed to a several IEEE publications.

Commenting rules


You must accept cookies to be able to make a comment.