Robust indoor loca/on tracking via interval analysis - PowerPoint PPT Presentation

Robust ¡indoor ¡loca/on ¡tracking ¡ 
 via ¡interval ¡analysis ¡ � Mohamed-­‑Hédi ¡AMRI, ¡Yasmina ¡BECIS, ¡ 
 Didier ¡AUBRY ¡& ¡ ¡Nacim ¡RAMDANI ¡ Université ¡d’Orléans, ¡Bourges, ¡France. ¡ � SWIM ¡2015, ¡Praha ¡ 
 9-­‑11 ¡June ¡2015

Outline n Motivations n Set membership estimation n Indoor location tracking n Experimental evaluation n Research directions 2

Monitoring for Healthcare 3

Smart Homes 4

Motivations n Smart home sensors + Robust data fusion 
 = Indoor location tracking, 
 = Activity Dailing Living characterization. � � n Indoor location tracking 
 = set-membership state reconstruction n Robust to sensor failures 5

Outline n Motivations n Set membership estimation n Indoor location tracking n Experimental evaluation n Research directions 6

Classical Estimation n Classical estimation is probabilistic C onfi de nce sets y s p 1 y s e(p) Optimisation of J ( e ( p )) f ( p ) p 2 n Yield valid results only if Perturbations, errors and model uncertainties with statistical properties known a priori Model structure is correct, no modeling errors 7

Set Membership Estimation n Unknown but bounded-error framework Solution set p 1 Y Y Set Membership Algorithm Set membersip algorithm f ( p ) p 2 n Hypothesis Uncertainties and errors are bounded with known prior bounds A set of feasible solutions S = { p ∈ P | f ( p ) ∈ Y } = f − 1 ( Y ) ∩ P 8

Set Membership Estimation n State estimation with continuous systems l Prediction - Correction / Filtering approaches ‣ (Kieffer, et al., 1999) … 9

Set Membership Estimation n Set inversion. Parameter estimation l Branch-&-bound, branch-&-prune, interval contractors … 
 (Jaulin, et al. 93) (Raïssi et al., 2004) 10

Outline n Motivations n Set membership estimation n Indoor location tracking n Experimental evaluation n Research directions 11

Binary sensors 12

Binary sensors 13

14

System modeling 15

Predictor-Corrector Approach Prediction step Correction step 16

Prediction step : random walk 17

Prediction step, 
 no motion detected 18

Use of RFID sensors 19

Correction step 20

q-Relaxed intersection 21

q-Relaxed intersection 
 (Jaulin, 2009) 22

q-Relaxed intersection 
 (Jaulin, 2009) 22

q-Relaxed intersection 
 (Jaulin, 2009) 22

q-Relaxed intersection 
 (Jaulin, 2009) 22

Outline n Motivations n Set membership estimation n Indoor location tracking n Experimental evaluation n Research directions 23

Location Tracking using 
 binary sensors only 24

Location Tracking using 
 binary sensors + RFID RSSI 25

Location tracking of 
 single inhabitant (IEEE ICRA 2015) 26

Location tracking of 
 single inhabitant (IEEE ICRA 2015) 26

Location tracking of 
 two inhabitants (IEEE CASE 2015) 27

Location tracking of 
 two inhabitants (IEEE CASE 2015) 27

Outline n Motivations n Set membership estimation n Indoor location tracking n Experimental evaluation n Research directions 28

Research directions n Use forward-backward predictions n Extend to multiple inhabitants n Use with multi-modality n Apply to FDI (IFAC SafeProcess 2015) � n Combine set-membership and stochastic modeling of errors. 29

Focused References n M.H. Amri, Y. Becis, D. Aubry, N. Ramdani, M. Fränzle, 
 Robust Indoor Location Tracking of Multiple Inhabitants Using Only Binary Sensors. 
 IEEE CASE 2015, Gothenburg, Accepted. n M.H. Amri, D. Aubry, Y. Becis, N. Ramdani, 
 Robust Fault Detection and Isolation applied to Indoor Localization. 
 IFAC SafeProcess 2015, Paris, Accepted. n M.H. Amri, D. Aubry, Y. Becis, 
 N. Ramdani, Indoor Human/Robot Localization using Robust Multi-modal Data Fusion, 
 IEEE ICRA 2015. Accepted. 30