Blog Overview

Building a classroom for autonomous vehicles

15.5.18
With the 3D digital twin of the world, test drivings for autonomous vehicles will no longer be the same.
With the 3D digital twin of the world, test drivings for autonomous vehicles will no longer be the same.

Did you like this article? Please share it.

Tags

Downloads

With self-driving cars making the headlines, many are starting to question the societal aspects of the technology: How do we make cars safer and how can we teach them to interact with other road users on the streets? The answer is to train them in the 3D digital twin of the world.


Silicon Valley is not the only place where a lot of the innovations surrounding autonomous vehicles are taking place. Nomoko is a Swiss start-up that is actively contributing to this field – helping accelerate the development of the production-ready self-driving car with their digital twin of the world.

To make autonomous driving safe, it is not enough to strap some sensors and software on the wheels. Vehicles need to be able to see, interpret and predict the behavior of human drivers, cyclists and pedestrians – perhaps even communicate with them. Cars must understand when they’re in another vehicle’s blind spot and drive extra carefully. They must learn to interact with other road users and analyze constantly changing surroundings. In a way, it is not much different from passing your driver’s license exam – except there is no driving instructor that can train the software.

Test driving will no longer be the same

A digital twin of the world could help overcome this challenge by providing a classroom for testing autonomous vehicles. Test driving will no longer be conducted solely on actual roads. Instead, driving will be first simulated in the digital replica of our world and tested for different cities or under different weather conditions and behavioral scenarios. It will be inherently possible, under multiple conditions, to test for a child running onto the street – something that is not possible in the real world.

Nomoko is on its way to creating such a digital twin of the world – the photorealistic 3D models of entire cities, filled with contextual information, such as real-time traffic, weather, and agents in motion. For accessing this digital twin of the world, Nomoko is developing the platform for interacting with it and building solutions rooted in reality such as simulating self-driving cars. This will have an immense impact on how we build, test and deploy safe autonomous vehicles on the roads.

For the past three years, Nomoko has been working on developing their technology and has gathered insights from the industry on what it needs to move forward faster. The unifying factor is the need for a testing environment as close to the real conditions as possible, which will make it possible to test and train autonomous vehicle software, accelerating the time horizon towards the self-driving car.

Paradigm shift in mobility

This shift in thinking is already happening across the industry, yet there are still a lot of open-ended questions to be tackled. From the societal perspective, for example, it remains to be seen how the accelerated technological development will impact society’s acceptance of autonomous driving. From the urban development perspective, autonomous vehicles will fundamentally transform cities as we know them today – not only will our commutes change, but also our approach to mobility will be transformed in ways that we are currently unaware of.