Advanced Driver Assistance Systems

Automotive teams use camera data, cloud ML infrastructure, and driver-support software to develop and improve advanced driver assistance systems.

Business impact

• avoidable collisions — Abellio London reported a 29% reduction in avoidable collisions.
• injuries from avoidable collisions — Abellio London reported a 60% reduction in injuries from avoidable collisions.
• development speed — SUBARU says the cloud setup let it prepare a development environment the same day, reducing setup delays.

Data requirements

• Vehicle-camera imagery (image) — Used to label objects and train perception models for driving support.
• Annotation data (tabular) — SUBARU adds annotation data to image data for learning and training records.
• Code repositories (text) — Used to synchronize and manage model code across environments.

AI methods and techniques

• Object detection — SUBARU uses it to identify objects such as cars and people in stereo-camera images.
• Semantic segmentation — SUBARU uses it to label pixels and recognize drivable areas.
• Image annotation — Tesla-employed labelers identify objects in vehicle-camera imagery for driving-system development.
• Collision warning logic — Mobileye provides warnings for forward collision, pedestrians, cyclists, headway, lane departure, and speed limits.

Industries

Real-world evidence

4 documented case studies on record.

Companies using this: Tesla, Abellio London, SUBARU Corporation, Chery Auto.