DoorINet: Door Heading Prediction through Inertial Deep Learning
This addresses the issue of accurate heading estimation for moving objects like pedestrians and appliances in indoor settings, representing an incremental improvement over prior methods.
The authors tackled the problem of poor heading angle estimation in indoor environments due to magnetometer interference by proposing DoorINet, a deep-learning framework that uses only accelerometer and gyroscope data from door-mounted sensors, and demonstrated it outperforms existing methods on a dataset of 391 minutes.
Inertial sensors are widely used in a variety of applications. A common task is orientation estimation. To tackle such a task, attitude and heading reference system algorithms are applied. Relying on the gyroscope readings, the accelerometer measurements are used to update the attitude angles, and magnetometer measurements are utilized to update the heading angle. In indoor environments, magnetometers suffer from interference that degrades their performance resulting in poor heading angle estimation. Therefore, applications that estimate the heading angle of moving objects, such as walking pedestrians, closets, and refrigerators, are prone to error. To circumvent such situations, we propose DoorINet, an end-to-end deep-learning framework to calculate the heading angle from door-mounted, low-cost inertial sensors without using magnetometers. To evaluate our approach, we record a unique dataset containing 391 minutes of accelerometer and gyroscope measurements and corresponding ground-truth heading angle. We show that our proposed approach outperforms commonly used, model based approaches and data-driven methods.