Edge Case Research (ECR) is set to collaborate with real-time 3D-reality capture developer Kaarta to improve road testing of self-driving vehicles.

Under the collaboration, ECR has initiated the process of mapping Pittsburgh’s streets using Kaarta 3D mapping and localisation technology.

The maps will be used to pilot ECR’s Hologram platform, a tool that transforms each recorded scenario into a million realistic potential scenarios to test.

Kaarta CEO Kevin Dowling said: “Kaarta’s mapping technology turns raw sensor data into intelligent 3D environment models in real time. These models provide high-dimensional fidelity of the real world for immediate testing.”

The platform is expected to improve testing of autonomous vehicles by informing the developers about safety and mission-critical perception issues.

Edge Case Research CEO Michael Wagner said: “Safe perception is critical for safe autonomous vehicles, and that’s why we’re building Hologram.

“Safe perception is critical for safe autonomous vehicles, and that’s why we’re building Hologram.”

“Hologram’s ability to create millions of unique test cases in a given street or environment means that developers of perception systems can train and validate their machine learning systems far more efficiently.”

Hologram procures recorded data during field testing to generate potential edge-case scenarios by adjusting all variables such as weather, lighting conditions, location of cars and pedestrians.

Subsequently, the platform combines these factors to offer countless realistic scenarios to evaluate the safety of autonomous perception systems.

The partnership is also expected to provide ECR with the 3D maps of the City of Pittsburgh to city planners and public stakeholders who require highly accurate models.

The City of Pittsburgh is the only urban location assigned by the US Department of Transportation to test automated vehicles.