Access to critical scenarios at scale
Deploying safe autonomous trucks requires a deep understanding of real-world driving. A key part in this process is the collection and analysis of traffic data to derive requirements, test cases, and statistical distributions for safety validation. DeepScenario’s unique approach allows Torc to accelerate this process. At the core of the solution are stationary cameras combined with DeepScenario’s AI computer vision software, providing Torc with unparalleled data collection capabilities at critical locations, such as intersections or highways. This enables identification of undiscovered critical edge cases that must be included in the validation process, and the derivation of a comprehensive statistical understanding of real-world driving. With this advantage, Torc can significantly increase the efficiency of scenario mining to continuously advance the safety of its autonomous trucks.
3D data extraction for scenario mining
DeepScenario‘s AI software processes video streams from any monocular camera, including traffic cameras and drones, regardless of the perspective view on the scene. At the core of their software is top-notch 3D computer vision algorithms that detect and track moving objects in a highly accurate and fully automated way. Our advanced 3D pipeline reconstructs all objects in metric space, enabling retrieval of crucial information such as positions, velocities, or distances for scenario mining.
DeepScenario’s foundational technology enables Torc to conduct large-scale traffic observations with commercially available cameras on its hub-to-hub routes. The extracted knowledge is seamlessly integrated into Torc’s virtual testing pipeline, allowing Torc to continuously advance its algorithms and enhance the safety of its self-driving trucks on the roads.