The Volvo Group and NVIDIA have signed an agreement to collaborate on the development of the decision making system of autonomous commercial vehicles and machines.
The system will be designed to safely handle fully autonomous driving on public roads and highways, the companies say.
The system will use NVIDIA’s end-to-end artificial intelligence platform for training, simulation and in-vehicle computing. It will be built on NVIDIA’s full software stack for sensor processing, perception, map localization and path planning, which will facilitate a number of possible autonomous driving applications including public transport, freight transport and refuse and recycling collection.
“Automation creates real-life benefits for both our customers and the society in terms of safety, energy efficiency and as a consequence productivity,” says Martin Lundstedt, president and CEO of the Volvo Group.
“We continue to gradually introduce automated applications in the entire spectrum of automation, from driver support systems to fully autonomous vehicles and machines. This partnership with NVIDIA is an important next step on that journey.”
The Volvo Group says that its partnership with NVIDIA will focus on developing a “flexible, scalable Autonomous Driving System.” The plan is for the system to first be used in commercial pilots, and later in commercial offerings from the Volvo Group.
The partnership between the companies is a long-term partnership, expected to last several years. Work will begin immediately, as personnel from the two companies will be co-located in Gothenburg, Sweden and Santa Clara, California.
“Automation is a key technology area for the Volvo Group. With this partnership we will further increase our speed of development and strengthen our long-term capabilities and assets within automation, to the benefit of our customers in different segments and markets,” says Lars Stenqvist, Volvo Group chief technology officer.
The companies say that the partnership covers “end-to-end computing fundamental to autonomous vehicles,” which includes “accelerated computing technology in the datacenter for training deep neural networks; large-scale simulation for hardware-in-the-loop testing and validation of autonomous vehicle systems; and finally deployment of the NVIDIA DRIVE platform in the vehicle running the full software stack for 360-degree sensor processing, mapping and path planning.”