Finding the unknown unknowns: intelligent data collection for autonomous driving development
Liang Yu - 2 years ago
In a safety-critical application such as autonomous driving, understanding what a model does not know is of paramount importance. Liang and her team are working on the Big Loop, an intelligent data aggregation system that supports data-driven development.
It helps identify a perception model’s weakness by detecting samples that pose a challenge to the model, and collects such data to improve its performance. Liang will dive deeper into the development of this system, specifically into a deep neural network-based software component called INSTINCT (IN SiTu INtelligent data CollecTor).
This piece of software analyses, in real time, all of the sensor data streamed throughout a vehicle to identify samples that can most efficiently improve the reliability of the automated driving systems.