Ride Recall - NUI-driven in-cabin occupant monitoring for the connected car

Every year, children and pets die in hot cars. Forgotten bags and medications cause daily stress. Ride Recall turns a parked vehicle into an intelligent guardian - sensing what's left behind, and who.

Existing driver monitoring systems focus entirely on safety during the drive. The moment the engine stops, cabin awareness disappears - creating a window where the most dangerous forgetting happens.

A research-grade prototype assembled from off-the-shelf hardware: pressure sensor mats, Raspberry Pi, and a wide-angle IR camera. Three iterations before the study - each refining detection accuracy and alert latency.

Twelve semi-structured interviews with daily drivers. Two mental models emerged: routine neglect and situational neglect. Both groups strongly preferred a phone notification over in-car audio - and tiered severity over a single undifferentiated alert.

Designed for one-handed, glance-first interaction. Dark backgrounds match the automotive context. Alert hierarchy uses colour and scale - legible in bright sunlight, unambiguous in a rush.

A within-subjects study with real participants: place an item in a parked car, exit, receive a notification, evaluate clarity and urgency. Conditions varied detection type, severity tier, and dismissal interaction - results published at ACM AutoUI 2025.

A five-person team across UX, hardware, and machine learning. End-to-end ownership of the user experience: from initial research through notification design, prototype testing, and co-authorship of the published paper.