GroundSignal — The Sidewalk Told on Itself

Описание и применение

GroundSignal превращает данные движения телефона в приватный, проверяемый отчёт о поверхности тротуара, используя модель TabPFN и локальное хранение, без сбора локации.

Задача

How to turn motion data from a phone into useful, privacy‑preserving surface evidence without turning it into surveillance.

Результат

GroundSignal provides a privacy‑first, local‑first surface‑classification workflow that outputs probability‑aware reports without collecting location data.

Рабочий процесс

• Choose mobility mode and phone placement
• Record at least two known surface types
• Start an audit and place phone in pocket/bag/mount
• Move naturally while GroundSignal records motion locally
• Finish audit and inspect surface signals
• Confirm, correct, or discard uncertain predictions
• Export evidence as CSV or JSON

Доказательства

GroundSignal started with one question: can the motion already happening in your pocket become useful evidence without becoming surveillance?

Источник

The AI core is TabPFN, which is well suited to the small, personal calibration sets this problem creates—and gives me probability evidence instead of a mystery label.

Источник

GroundSignal is a local‑first pocket field instrument for the Hacktoberfest 2026 Week 1 Touch Grass challenge.

Источник

Raw motion stays on the device. Only bounded derived features and personal calibration labels cross the explicit analysis boundary; probabilities return for local human review.

Источник