Simulation API to test autonomous systems

Collecting emergency vehicle data is a huge challenge. We were able to achieve significant performance improvements in our emergency vehicle detection algorithms by using the Parallel Domain platform to generate synthetic datasets that matched our operational domain here in Japan.

Thanks to its high degree of realism, flexibility and scalability, the Parallel Domain platform enables rapid exploration of cutting-edge machine learning ideas. Its cost effectiveness enables accelerated paths to deployments at scale. This combination makes the platform really unique and a huge advantage for us to develop the future of robot autonomy.

Parallel Domain data boosted the machine learning model performance of our traffic light classification system. Their platform provided us with a diverse dataset with edge cases and accurate annotations that would not have been possible with our real world data operations.

Collecting emergency vehicle data is a huge challenge. We were able to achieve significant performance improvements in our emergency vehicle detection algorithms by using the Parallel Domain platform to generate synthetic datasets that matched our operational domain here in Japan.