Abstract
A multi-stage, power-efficient method for detecting when a wearable device is on the user’s head is disclosed. To reduce battery consumption, a cascaded verification process is used to control the activation of high-power components. An initial check of charging status and inertial measurement unit (IMU) data can identify movement and hinge-opening events. This is achieved through a differential analysis of orientation data from multiple IMUs to distinguish relative hinge motion from overall device movement. Once a hinge-open state is confirmed, mid-power neural processing units (NPUs) and eye-tracking cameras can be activated to perform a sparse, active stereo proximity check. This check can utilize infrared glint reflections to triangulate object distance, confirming the presence of a user’s face within a specific focal plane. The main application processor may then be woken or set to an active state after this tiered verification. This approach ensures an instant-on user experience while preventing false wake events caused by accidental handling.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Hurwitz, Jonathan D. and Mahomed, Yaaseen, "Low-Power On-Head Detection for Wearable Devices Using Sensor Gating", Technical Disclosure Commons, ()
https://www.tdcommons.org/dpubs_series/9971