GTC 2020: A Study of Pedestrian Protection CAE Using GAN
After clicking “Watch Now” you will be prompted to login or join.
Click “Watch Now” to login or join the NVIDIA Developer Program.
A Study of Pedestrian Protection CAE Using GAN
Osamu Ito, Honda R&D | Jun Shiraishi, Honda R&D
According to the World Health Organization, there are over 270,000 pedestrians involved in traffic fatal accidents. That is 22% of all traffic fatalities verified in the world in 2013. To reduce pedestrian fatalities, third-party organizations in many countries have held New Car Assessment Program (NCAP) tests to evaluate pedestrian-protection performance. For that reason, an efficient method to design pedestrian-protection performance is required for automobile development for all over the world. Computer-aided engineering (CAE) is often used to verify pedestrian-protection performance. But as the necessity for pedestrian protection expands globally, expectations to improve efficiency have recently risen. We decided to reduce CAE verification time and improve the accuracy of CAE using deep learning. We also studied visualizing the basis for judgment in the trained models.