Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat,...
-
Upload
simon-peters -
Category
Documents
-
view
213 -
download
0
Transcript of Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat,...
![Page 1: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/1.jpg)
Doorjamb: Unobtrusive Room-level Tracking of People in Homes
using Doorway Sensors
Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin WhitehouseU of Virginia
Presenter:SY
![Page 2: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/2.jpg)
About This Paper
• Unobtrusive room-level tracking – People in homes
• Doorway sensors– Ultrasound sensor
• Method– Estimates the height and direction
![Page 3: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/3.jpg)
Technical Problems
• Multi-target tracking – Data association
• Noise– Person’s posture, multipath reflections, and the
natural undulation of gait• Algorithms– Crossing event detection– Tracking
![Page 4: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/4.jpg)
Contributions
• Hardware– Design and prototyping– Lesson learned
• In-depth analysis of the sources errors– Present signal processing algorithm
• Data association challenges– Tracking algorithm
• Proof-of-concept implementation, deployment, and empirical evaluation
![Page 5: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/5.jpg)
Outline
• Hardware design• Signal processing algorithm• Tracking algorithm• Evaluation• Conclusion
![Page 6: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/6.jpg)
Hardware
• Features– Cost effective– Battery powered– Wireless
• Design– Detect height • Measure the distance to the top of the head
– Detect walking direction• Angled into one room more than the other
![Page 7: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/7.jpg)
Doorway Sensor
• Parallax PING ultrasonic range finders• Passive infrared sensors • Magnetic reed sensors• Custom-designed power module • Synapse Wireless SnapPY RF100 module
![Page 8: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/8.jpg)
Achieving Doorway Coverage
• Requirements– 1 cm resolution– Heights ranging from 151 cm to 189 cm– Walking speeds up to 3 m/s^2– Doorways range: 90-300 cm wide, 213-275 cm tall
• Parallax PING ultrasonic– 40 degree beam angle– Min: 2 cm; Max: 300 cm
![Page 9: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/9.jpg)
Achieving Doorway Coverage
• Tallest person– Gap between the head and doorway 24cm– 40 degree beam Sensing diameter of 17 cm– Speed of 3 m/s, a head that is 15 cm diameter• Pass sensing region in about 100 ms
– 50 Hz sample rate – one module at a time
![Page 10: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/10.jpg)
Doorway size
• Typical doorway width of 90 cm– Sensing diameter – 17 cm– Head radius – 7 cm– Two sensors should be enough
• Higher door frames require fewer sensor• 300 x 275 cm– 4 range finders– Sampling rate 12.5 Hz– Cannot support wide and short
![Page 11: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/11.jpg)
Early Prototypes and Lessons Learned
Audible click
![Page 12: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/12.jpg)
Outline
• Hardware design• Signal processing algorithm• Tracking algorithm• Evaluation• Conclusion
![Page 13: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/13.jpg)
Signal Processing
• Input: stream of height value• Output: doorway events D (tj,hj, vj)• Four algorithms– Doorway crossing detection– Noise filtering– Height estimation– Direction estimation
![Page 14: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/14.jpg)
Signal Captured
![Page 15: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/15.jpg)
Doorway Crossing Event
• Find timeout, multi-path, measurement events • Within 400 msecs of each other
![Page 16: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/16.jpg)
Noise Filtering
Extend 200ms
Define clusters
![Page 17: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/17.jpg)
Noise Filtering -- Obstacle
• Extends 30 seconds on either side– Remove any height measurement that is positive and identical
![Page 18: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/18.jpg)
Height Estimation
• Multi-path reflections– Maximum measurement may fail– Typically only occur once
• Height estimation– If maximum height cluster exist• Max of the cluster
– Else • Maximum height
![Page 19: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/19.jpg)
Direction Estimation
• Sensor tilts into the doorway• Three algorithms– Line slope– Compare max height timestamp to median– Compare min height timestamp to median
• Vote– Each algorithm estimate: +1, -1, 0– Sum all: [-3,3]
![Page 20: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/20.jpg)
Outline
• Hardware design• Signal processing algorithm• Tracking algorithm• Evaluation• Conclusion
![Page 21: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/21.jpg)
Tracking
• Input: sequences of detection events D• Output: Corresponding room states S, (r1i, r2i)• Ambiguity– False detections, miss detections
• Key insight– Ambiguities can often be resolved by future
observations
![Page 22: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/22.jpg)
MHT Algorithm
• Multiple hypothesis tracking approach– Multiple alternative tracks are considered
simultaneously
• As new events are processed– Tracks that are not consistent with the new
information are evicted
![Page 23: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/23.jpg)
Overview
• Initial– All tracks created with identical weight– For 2 persons + K rooms, K2 tracks are created
• Update– For each doorway event
• Update track• Update weight (based on prior training study)
• Merging and Evicting– Evicting low weight tracks– Merging duplicate tracks
![Page 24: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/24.jpg)
Prior Training Study
• Find conditional probabilities– p(H|O) – a height measurement given the origin– p(V|O) – a direction measurement given the origin– p(H = ) – probability of missed detection
• Origin -- Person A, or B, or false detection• Training period– Each individual walks under each doorway
multiple times
![Page 25: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/25.jpg)
Creating Tracks
• Initial tracks – every possible combination
• For each new doorway event– Between rooms i and j– Five new states are possible• a/b move to room i/j + false detection
– Duplicate every track 5 times
![Page 26: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/26.jpg)
Weighting Tracks
• New weight is– Old weight multiply by– Probability of the origin moved through doorway
m given height measurement– Probability of moving from room p to m given the
direction measurement– Probability of moving from the last observed room
m-1 to p without having detected
![Page 27: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/27.jpg)
Merging and Evicting Hypotheses
• “N-best” eviction policy– Keep the n best
tracks• Problem –
duplicate tracks• Track merging
algorithm
1
2
3
4
![Page 28: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/28.jpg)
Outline
• Hardware design• Signal processing algorithm• Tracking algorithm• Evaluation• Conclusion
![Page 29: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/29.jpg)
Experimental Setup
• Built 43 ultrasonic doorway sensors– Deployed across 4 different homes – Periods of 6-18 months– Used for development, testing, and iterative design
• For this evaluation– Performed 3 controlled experiments– 3 different pairs of testers– Randomly walk around– Collect ground truth with handheld device– 3000 unique doorway events
![Page 30: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/30.jpg)
Evaluation Metric
• Type 1: correct state• Type 2: wrong person• Type 3: false room transition• Type 4: missed room transition
![Page 31: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/31.jpg)
Tracking Accuracy
![Page 32: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/32.jpg)
False Detections and Missed Detections
• Precision:– The number of false detections divided by the number of total detections
• Recall – Number of missed detections divided by the number of true doorway
crossing events
![Page 33: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/33.jpg)
Height Measurement Accuracy
![Page 34: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/34.jpg)
Direction Measurement Accuracy
![Page 35: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/35.jpg)
Systems Performance
• Average 24 states, max 55 states per track• Real time, online
– With 500 ms look-ahead window
![Page 36: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/36.jpg)
Limitations
• Fall short of true in-situ experiments– Controlled experiments
• Do not capture long-term effects• A proof-of-concept for Doorjamb tracking• Scalability– Typical homes with 3-4 people
• Requires calibration and training• Does not detect children
![Page 37: Doorjamb: Unobtrusive Room-level Tracking of People in Homes using Doorway Sensors Timothy W. Hnat, Erin Griffiths, Ray Dawson, Kamin Whitehouse U of Virginia.](https://reader035.fdocuments.in/reader035/viewer/2022062713/56649cee5503460f949bc88f/html5/thumbnails/37.jpg)
Conclusion
• Track people in homes with room-level accuracy
• Unobtrusive• Achieve 90% tracking accuracy• My opinions– Well written complete work– Not so sexy– Has it’s own selling points