Why is OPTIS a major player in the development of autonomous vehicles?

OPTIS wants to enhance the artificial intelligence of autonomous vehicles with data obtained in virtual reality

September 29th, 2017 | Innovation

Photo credit: medium.com©

Historically specialized in optics, OPTIS quickly turned to simulation with the creation of SPEOS, an offline and non-real-time simulation software integrated into CAD software such as Catia. SPEOS can generate photo-realistic scenes which very precisely simulate the interaction between light and matter (and sometimes require several days of calculation). Since 2007, the company also made real-time rendering. OPTIS incorporates its technology in driving simulators, for which the parameters can change without pre-calculation (hour of day, light sources, interior lighting ...). Now well-known as a virtual reality specialist, OPTIS applies its expertise in simulation to the new needs of the autonomous vehicle.


OPTIS has been working for a few years on the optical simulation of the intelligent car, capable of detecting pedestrians or other vehicles and adapting the illumination areas to avoid dazzling them without any driver's intervention. But current research focuses on the autonomous vehicle, and more precisely on its sensors. The company is working on the simulation of these sensors, such as cameras, LIDAR, ultrasound ... to better train their neural networks with data from virtual environments.
Let's take the example of a camera! OPTIS simulates the electronics of the camera as well as the digital conversion to reproduce the data generated in real situation as precisely as possible. In particular, the company is able to simulate the effects of noise and glare that may impact the sensors.

"We are trying to understand how to properly train a network of neurons with virtual data," said Nicolas Dalmasso, innovation director at OPTIS. The goal is to allow customers to use data from a virtual environment instead of real-world ones. "



SystemX partners with OPTIS to simulate the autonomous vehicle

VIDEO - The driverless vision becomes clearer with virtual prototyping 

Why is it so difficult to simulate an autonomous vehicle? 

Lidar Design challenge 


Photo credit: medium.com©