bachelor thesis - rvs.unibe.chrvs.unibe.ch/teaching/seminar_spring2017/niklaus.pdf · 2nd...
TRANSCRIPT
![Page 1: Bachelor Thesis - rvs.unibe.chrvs.unibe.ch/teaching/seminar_spring2017/niklaus.pdf · 2nd Presentation Joel Niklaus Supervisors: Dr. ZhongliangZhao, José Luis Carrera Communication](https://reader034.vdocuments.mx/reader034/viewer/2022042222/5ec910d31b72bc186b68c300/html5/thumbnails/1.jpg)
Machine Learning for Indoor Localization2nd Presentation
Joel Niklaus
Supervisors:Dr. Zhongliang Zhao, José Luis Carrera
Communication and Distributed SystemsUniversity of Bern
Bachelor Thesis
![Page 2: Bachelor Thesis - rvs.unibe.chrvs.unibe.ch/teaching/seminar_spring2017/niklaus.pdf · 2nd Presentation Joel Niklaus Supervisors: Dr. ZhongliangZhao, José Luis Carrera Communication](https://reader034.vdocuments.mx/reader034/viewer/2022042222/5ec910d31b72bc186b68c300/html5/thumbnails/2.jpg)
Outline
> Short Recap— Goal of the Project— Proposed Solution
> Progress> Landmarks> Experiment
— Methodology— Environment
> Preliminary Results— With / Without Center— With / Without Magnetic Field
> Further Work> Summary> Questions
![Page 3: Bachelor Thesis - rvs.unibe.chrvs.unibe.ch/teaching/seminar_spring2017/niklaus.pdf · 2nd Presentation Joel Niklaus Supervisors: Dr. ZhongliangZhao, José Luis Carrera Communication](https://reader034.vdocuments.mx/reader034/viewer/2022042222/5ec910d31b72bc186b68c300/html5/thumbnails/3.jpg)
Goal of the Project
> Component to enhance Indoor Tracking System
> Should improve localization at specific points
> Use RSS values and sensor data from smartphone
> Due to complexity: use machine learning algorithms
> Improve accuracy and performance
![Page 4: Bachelor Thesis - rvs.unibe.chrvs.unibe.ch/teaching/seminar_spring2017/niklaus.pdf · 2nd Presentation Joel Niklaus Supervisors: Dr. ZhongliangZhao, José Luis Carrera Communication](https://reader034.vdocuments.mx/reader034/viewer/2022042222/5ec910d31b72bc186b68c300/html5/thumbnails/4.jpg)
Proposed Solution
> Android app tocollect data
> WEKA formachinelearning
> Experiment on whichalgorithmswork best
Receive data(Android
app)RSSI
Sensors Test besttrained Model
(WEKA)
Train Models
(WEKA)
Evaluate bestModel
Location
Data
Models
Mod
el
![Page 5: Bachelor Thesis - rvs.unibe.chrvs.unibe.ch/teaching/seminar_spring2017/niklaus.pdf · 2nd Presentation Joel Niklaus Supervisors: Dr. ZhongliangZhao, José Luis Carrera Communication](https://reader034.vdocuments.mx/reader034/viewer/2022042222/5ec910d31b72bc186b68c300/html5/thumbnails/5.jpg)
Progress
> Until first presentation— Integrated WEKA 3.7.3 into Android app (https://github.com/rjmarsan/Weka-for-Android)— Implemented functionality to collect data— Implemented first performance and accuracy tests
> Additionally until now— Continuous data collection now: allows data points to be collected much faster— Ran experiments for both room recognition and landmark recognition— Ran lots of experiments on which features improve the result— Ran lots of experiments to tweak parameters of ml methods (autoweka)
![Page 6: Bachelor Thesis - rvs.unibe.chrvs.unibe.ch/teaching/seminar_spring2017/niklaus.pdf · 2nd Presentation Joel Niklaus Supervisors: Dr. ZhongliangZhao, José Luis Carrera Communication](https://reader034.vdocuments.mx/reader034/viewer/2022042222/5ec910d31b72bc186b68c300/html5/thumbnails/6.jpg)
Landmarks
> Small area inside a room (around 1m2)
> Why do we need Landmarks?— Higher local accuracy inside a
room— Higher prediction confidence
inside a room— à Provide new starting point
for tracking system where we can be almost absolutely sure of!
— Points far away are very unlikely à exclude them
![Page 7: Bachelor Thesis - rvs.unibe.chrvs.unibe.ch/teaching/seminar_spring2017/niklaus.pdf · 2nd Presentation Joel Niklaus Supervisors: Dr. ZhongliangZhao, José Luis Carrera Communication](https://reader034.vdocuments.mx/reader034/viewer/2022042222/5ec910d31b72bc186b68c300/html5/thumbnails/7.jpg)
Experiment Methodology
> Collected in grid pattern
> Landmark level— About 500 data points per landmark
> With / without center landmark> With / without magnetic field
![Page 8: Bachelor Thesis - rvs.unibe.chrvs.unibe.ch/teaching/seminar_spring2017/niklaus.pdf · 2nd Presentation Joel Niklaus Supervisors: Dr. ZhongliangZhao, José Luis Carrera Communication](https://reader034.vdocuments.mx/reader034/viewer/2022042222/5ec910d31b72bc186b68c300/html5/thumbnails/8.jpg)
Experiment Environment
> Apartment in student accommodation in Exeter, England
> Kitchen area
> 2 specially installed access points + 2 others from university network (location unknown)
AP 1
AP 2
6.8 m
6.4 m
![Page 9: Bachelor Thesis - rvs.unibe.chrvs.unibe.ch/teaching/seminar_spring2017/niklaus.pdf · 2nd Presentation Joel Niklaus Supervisors: Dr. ZhongliangZhao, José Luis Carrera Communication](https://reader034.vdocuments.mx/reader034/viewer/2022042222/5ec910d31b72bc186b68c300/html5/thumbnails/9.jpg)
Preliminary Results
> Training is usually much more computation heavy (except in instance basedmethods like knn)
> Testing performance mostly is very fast (except in instance based methods like knn)
> à Focus on accuracy
![Page 10: Bachelor Thesis - rvs.unibe.chrvs.unibe.ch/teaching/seminar_spring2017/niklaus.pdf · 2nd Presentation Joel Niklaus Supervisors: Dr. ZhongliangZhao, José Luis Carrera Communication](https://reader034.vdocuments.mx/reader034/viewer/2022042222/5ec910d31b72bc186b68c300/html5/thumbnails/10.jpg)
Landmarks With/Without Center (with magnetic field)
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
70.00%
80.00%
90.00%
100.00%
Accuracy
Accuracy with center Accuracy without center
![Page 11: Bachelor Thesis - rvs.unibe.chrvs.unibe.ch/teaching/seminar_spring2017/niklaus.pdf · 2nd Presentation Joel Niklaus Supervisors: Dr. ZhongliangZhao, José Luis Carrera Communication](https://reader034.vdocuments.mx/reader034/viewer/2022042222/5ec910d31b72bc186b68c300/html5/thumbnails/11.jpg)
Landmarks With/Without Magnetic Field (with center)
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
70.00%
80.00%
90.00%
100.00%
Accuracy
Accuracy with magnetic field Accuracy without magnetic field
![Page 12: Bachelor Thesis - rvs.unibe.chrvs.unibe.ch/teaching/seminar_spring2017/niklaus.pdf · 2nd Presentation Joel Niklaus Supervisors: Dr. ZhongliangZhao, José Luis Carrera Communication](https://reader034.vdocuments.mx/reader034/viewer/2022042222/5ec910d31b72bc186b68c300/html5/thumbnails/12.jpg)
Further Work
> Add GPS as a feature (may help at building‘s border)
> Refactor and clean up the code
> Write documentation
![Page 13: Bachelor Thesis - rvs.unibe.chrvs.unibe.ch/teaching/seminar_spring2017/niklaus.pdf · 2nd Presentation Joel Niklaus Supervisors: Dr. ZhongliangZhao, José Luis Carrera Communication](https://reader034.vdocuments.mx/reader034/viewer/2022042222/5ec910d31b72bc186b68c300/html5/thumbnails/13.jpg)
Done
> Data collection
> Weka integration
> Sensor data
TO DO
> GPS Integration
> Refactoring
> Documentation
Summary
![Page 14: Bachelor Thesis - rvs.unibe.chrvs.unibe.ch/teaching/seminar_spring2017/niklaus.pdf · 2nd Presentation Joel Niklaus Supervisors: Dr. ZhongliangZhao, José Luis Carrera Communication](https://reader034.vdocuments.mx/reader034/viewer/2022042222/5ec910d31b72bc186b68c300/html5/thumbnails/14.jpg)
QUESTIONS ?Please ask! J