Motion Occupancy 2 /18 Motion Occupancy 3 /18 Walkway Sensing 4 - PowerPoint PPT Presentation

WalkSense: Cla lassify fying Home Occupancy States Usin ing Walk lkway Sensing Elahe Soltanaghaei, Kamin Whitehouse Department of computer science, University of Virginia 1 /18

Motion ≠ Occupancy 2 /18

Motion ≠ Occupancy 3 /18

Walkway Sensing 4 /18

Overview Active Sleep Active Sleep Away Away Bedroom Living room Bathroom Bedroom 5 /18

Outline • WalkSense: Occupancy detection algorithm • Automatic training feature • Evaluation 6 /18

WalkSense Candidate Transition Events Sleep Walkway Active-zone events Time 7 /18

Training WalkSense How to train the models? By exploiting the WalkSense in offline mode 8 /18

Training WalkSense • Candidate Sleep/Away Intervals: Inactivity longer than K Candidate Away Intervals Outside Walkway Active-zone events Time Sleep Walkway < K Candidate Sleep Intervals Automatic Labeling  No user involvement Incremental Training  Improve detection model over time 9 /18

Experimental Setup • 350 days worth of data bedroom Living room bedroom Living room Living room Kitchen Kitchen bedroom bedroom Kitchen bathroom House B House A House C, D Study room Kitchen bedroom Living room Living room Kitchen bedroom House E Aruba 10 /18 Active Sensing Zone Outside Walkway Zone Sleep Walkway Zone

Experimental Setup • Baseline: HMM-base • Evaluation Metrics • Energy penalty: away or sleep periods are wrongly detected as Active • Comfort penalty: Active periods are wrongly detected as Sleep or Away 11 /18

Experimental Setup 61% of the day outside home, or sleep 12 /18

Evaluation 30% lower energy penalty 32% lower comfort penalty 13 /18

Evaluation WalkSense HMM Baseline Actual Actual Predicted Predicted 95% vs. 83% Accuracy 14 /18

Analysis Training Size Effect 15 /18

Conclusion • WalkSense: Occupancy Detection algorithm based on Walkway Sensing • Higher Accuracy • Fewer Sensors • 96% Accuracy with 30% and 32% lower energy and comfort penalty • Limitation : Assumes direct relationship between Occupancy States & Occupancy zones. • TV • Study Desk 16 /18

Conclusion: Generalizable 17 /18

Thank You! 18