Improved actionSLAM for long-term indoor tracking with wearable motion sensors

Abstract

We present an indoor tracking system based on two wearable inertial measurement units for tracking in home and workplace environments. It applies simultaneous localization and mapping with user actions as landmarks, themselves recognized by the wearable sensors. The approach is thus fully wearable and no pre-deployment effort is required. We identify weaknesses of past approaches and address them by introducing heading drift compensation, stance detection adaptation, and ellipse landmarks. Furthermore, we present an environment-independent parameter set that allows for robust tracking in daily-life scenarios. We assess the method on a dataset with five participants in different home and office environments, totaling 8.7h of daily routines and 2500m of travelled distance. This dataset is publicly released. The main outcome is that our algorithm converges 87% of the time to an accurate approximation of the ground truth map (0.52m mean landmark positioning error) in scenarios where previous approaches fail.

Wi-Fi Fingerprint-Based Indoor Positioning: Recent Advances and Comparisons

“…Therefore, some compensation on rotation angle is needed to correct the drift, which is introduced by relative change of device position from the user body [106]. Another approach is to use wearable devices, which can be attached to human body, and capture more accurate motion information for localization [102], [107], [108].…”

Fusion of visual odometry and inertial navigation system on a smartphone

“…Scientific writings also describe other methods for the step detection, e.g. the ZUPT (Zero Velocity Update) method [36], and methods based on frequency analysis of the accelerometer signal. In the described algorithm for the step detection the information (i.e.…”

Automatic Update of Indoor Location Fingerprints with Pedestrian Dead Reckoning

“…Several studies constructed radio maps with no supervision by using simultaneous localization and mapping (SLAM) techniques [Ferris et al 2007;Robertson et al 2011]. Hardegger et al [2013] performed SLAM based on the fact that certain daily activities are performed at particular places (e.g., sleeping in a bedroom). Rai et al [2012] also tried to automatically construct radio maps with the PDR technique in the similar way to our approach.…”