At Autonomous Vehicle Technology Expo 2019, French company Intempora was showcasing its new IVS Intempora Validation Suite, a set of software solutions for testing, benchmarking and validation of perception deep learning algorithms. It allows engineers to test algorithms against vast data sets in the cloud.
CEO Nicolas Du Lac said that “with RT Maps [Intempora’s sensor and data fusion software] you can acquire data from multiple sensors in the vehicle and record the raw data. You can also annotate situation, manually or through algorithms. IVS is a cloud-based solution to store up to petabytes of recorded data. This makes it easier to share data between different sites and different teams.”
IVS makes it easy to search for specific data, based on certain parameters such as road and vehicle type, speed, behavior and more. This data can then be downloaded for local processing or can be sent to the ‘test automation manager’ to trigger jobs in the cloud to be executed, for example running certain perception algorithms on specific scenarios.