Claytex is adopting a holistic approach to the simulation of autonomous vehicles, explains Theodor Ensbury, one of the company’s simulation engineers.
Reality is but a construct of millions of interdependent phenomena. As humans, we use five main senses – sight, sound, smell, touch and taste – to observe the world around us and relay that information to our brains. We are often fooled though. A pungent smell can overpower other odours, while something soft can dull something sharp. Autonomous vehicles exist in the same reality, with their sensors equally fallible.
The RAND corporation estimates that some five billion miles need to be covered by an autonomous vehicle to have 95% confidence in its safety versus a human driver. It is clear, therefore, that simulation will play a central role in bringing autonomous vehicles to market. A holistic approach to recreating autonomous vehicles, and the conditions they experience, is imperative in order to truly recreate reality in a digital twin. Three areas need to be addressed for this to be achieved: the sensor, the environment and the vehicle.
Sensor models must behave in precisely the same way as their real-life counterparts. Optical effects such as lens distortion and field of view, for instance, must be correctly recreated from the real camera, with the deliverable signal formatted exactly how the control algorithm expects. LiDAR sensor models need to spin at the same rate as in real life, and be able to measure the distance and intensity of the reflection to each point in the environment. Weather effects, such as the reduction of effective range with higher humidity, or the beam ‘scatter’ effect from raindrops on LiDAR sensor performance, need to be incorporated. Radar sensors also need to recreate the reflection from the environment. Noise also needs to be accounted for. In all cases, the data output from the sensor model must be the same as from the real unit.
To access the complete article, please visit: Recreating reality
If you have any questions or for more information, please do not hesitate to contact us.
Telephone: +44 1926 885900 — Email: email@example.com