Hardegger, Michael

Michael Hardegger

ETH Zürich
Michael Hardegger
Institut f. Elektronik
ETZ H 97
Gloriastrasse 35
8092 Zürich

Phone: +41 44 632 52 97
Fax: +41 44 632 12 10

Michael Hardegger studied electrical and biomedical engineering at ETH Zürich and ANU Canberra. After graduating in 2010 he worked for the ABB research center in Dättwil and started his PhD at the Wearable Computing Lab in December 2011. His main interests are rehabilitation engineering, pervasive computing and indoor pedestrian tracking with wearable systems (ActionSLAM). He is also passionate about ice hockey and supervised several student projects on ice skating, stick-integrated sensors and gameplay monitoring.

Project: CuPiD

CuPiD aims at alleviating Parkinson's Disease symptoms through sensor-assisted rehabilitation, and at supporting patients in daily life with context-aware feedback tools. We contributed to the project by developing a freezing-of-gait detection system that triggers rythmical auditory stimulation, helping patients to resume walking. We also aim at making such systems sensitive to particularily hazardous locations by integrating stand-alone, fully wearable person tracking.

Teaching: Digitaltechnik, Semester/Master Projects


Please contact me if you would like to test ActionSLAM, we can provide Matlab/Android code and datasets after requests.

This video shows ActionSLAM applied to a dataset simulating daily life. In green, the current ActionSLAM path and position estimate are depicted. In cyan the posterior path (of the full recording) position estimate is shown, and in magenta the ground truth position (where available; from video). Detected landmarks are plotted in yellow. The shape indicates the action type of a landmark.


LocAFusion is an algorithm for location-enhanced activity recognition. Rather than applying supervised learning techniques to model the mapping from location to activity, LocAFusion accumulates all activity observations during post-processing and build a semantic map of the environment. It then post-corrects the activity observations to correspond to the objects that the user most likely interacted with.

Presentation on "Enhancing Action Recognition through Simultaneous Semantic Mapping from Body-Worn Sensors" at Ubicomp/ISWC 2014 in Seattle, USA.


G8trainer is an Android game for gait rehabilitation and demonstration of the ZUPT-PDR algorithm. The user has to follow a pre-defined path while wearing a foot-mounted IMU and gets audio feedback if he is too far off the track.

The video shows how a person first defines a reference track and then follows this reference track. The smartphone app recognizes when the user is too far off the reference, and tells him whether he was too fast, slow, or too far on the right or left of the target track.


2016 2015 2014 2013 2012 2011

IEEE copyright disclaimer:
Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.


Wichtiger Hinweis:
Diese Website wird in älteren Versionen von Netscape ohne graphische Elemente dargestellt. Die Funktionalität der Website ist aber trotzdem gewährleistet. Wenn Sie diese Website regelmässig benutzen, empfehlen wir Ihnen, auf Ihrem Computer einen aktuellen Browser zu installieren. Weitere Informationen finden Sie auf
folgender Seite.

Important Note:
The content in this site is accessible to any browser or Internet device, however, some graphics will display correctly only in the newer versions of Netscape. To get the most out of our site we suggest you upgrade to a newer browser.
More information

© 2017 ETH Zurich | Imprint | 2 September 2015