Sequential Adaptive Sensor Management

1
Sequential Adaptive Sensor Management

• Alfred O. Hero III
• Dept. of Electrical Engineering and Computer
  Science, The University of Michigan

2
Sequential Sensor Resource Allocation

• Progress (since June 05)
• Theory of information gain (IG) scheduling
• Result IG bounds risk (Kreucher CDC05).
• Implication IG is a universal surrogate
• Classification reduction for RL
• Result Generalization error bounds (Blatt,
  Thesis-06).
• Implication Minimum samples and
  model/measurement complexity
• Adaptive energy allocation and waveform selection
• Result LARS reduction of optimal adaptive
  waveform selection policy (RangarajanICASSP06)
• Implication linear-complexity solution to
  exponential-complexity problem
• IRIS sensor management for STW
• Result IRIS adaptive illuminator placement
  strategy w/ confidence maps
• Implication Information-directed path planning
  for STW (Marble06)

Predict performance for each possible sensing
action
Time update information state under each
available action model
Compute expected improvement for each sensing
action
Deploy action with best predicted performance
improvement
Measurement update info state
3
Progress Highlighted Today

• Adaptive energy allocation and waveform selection
  RangarajanRaichHero
• Iterative Redeployment of Illumination and
  Sensing (IRIS) for STW MarbleRaichHero

4
Progress 1 Adaptive Waveforms

• Sequentially illuminate a medium and measure
  backscatter using an array of sensors.
• Applications to mine detection, ultrasonic
  medical imaging, foliage penetrating radar,
  nondestructive testing, communications, and
  active audio.
• GOAL Optimally design a sequence of waveforms
  using an array of transducers
• To image a scatter medium (Estimation).
• To track targets (Tracking)
• To discover strong scatterers (Detection).

5
Progress 1a Energy allocation for DE

• Let past observations be

2. Active Waveform Design
1. Adaptive Energy Management under
average energy constraint

• Energy allocation question Given transmission
  of certain waveforms , how much can we gain
  through optimal energy allocation between various
  time steps (Rangarajan2005)?

R. Rangarajan, R. Raich and A. O. Hero, "Optimal
experimental design for an inverse scattering
problem," ICASSP-2005.
6
Gains more than 5dB!!!!

• RESULT We prove through optimal energy designs,
  we can achieve at least 5dB gain (compared to
  one-step strategy) for estimation problems
  (imaging).
• How much can we gain for target detection??

7
Results for target detection

• Two-step energy design procedure
• 2dB gain or 20 decrease in average error for
  same SNR.
• How much improvement can be achieved
  asymptotically with time? (Work in progress)

8
Progress 1b Active waveform selection

• M possible waveforms
• Can only send p out M, p lt M1
• Design criterion
• Optimal solution subset selection, is intractable

9
Simplification via rule ensembles

• We approximate the decision statistic at receiver
  (detector, estimator, classifier) by a weighted
  sum of non-linear functions (rule ensembles
  (Friedman2005)) of subsets of q measurements at
  time t
• Special case (GAM) for estimating state variable
  s
• Reduced GAM waveform selection criterion

Friedman, J. H. and Popescu, B. E. "Predictive
Learning via Rule Ensembles." (Feb. 2005)
10
Solution via convex relaxation

• Convex relaxation (Tibshirani1994) of waveform
  design criterion (Rangarajan2006)

Tibshirani, R. "Regression selection and
shrinkage via the lasso" Technical Report (June.
1994). R. Raghuram, R. Raich and A.O. Hero,
"Single-stage waveform selection for adaptive
resource constrained state estimation," IEEE
Intl. Conf. on Acoustics, Speech, and Signal
Processing, Toulouse France, 2006.
11
Summary comparisons

• HMM diffusion with bi-level variance
• Diffusion measured in Gaussian additive noise
  with one of possible subsets of n5
  waveforms

12
Numerical results

• Future Directions
• Sensor network localization/tracking problem.
• Combine optimal energy allocation with waveform
  selection.
• (Work in progress)

13
Progress 2 Iterative Redeployment of
Illumination and Sensing (IRIS)

• Elements of IRIS strategy
• Initial illumination with physical antenna
  array
• Antenna array is deployed at an initial location
  and illuminates the region of interest.
• Sparse reconstruction image reconstruction
  (Ting2006) is performed
• Form Confidence Map of Image
• Confidence map (Raich2005) is computed using
  initial image and side information
• Select a region of low confidence from confidence
  map
• Simulate external energy/resolution field
  induced by virtual transmitter
• Place virtual transmitter in low confidence
  region and apply FEM, MoM, PO to estimate
  electric field distribution outside the building
• Compute induced energy or gradient field (wrt
  perturbation of virtual transmitter location)
• Re-illuminate with physical antenna array at
  maximum of simulated field

M. Ting, R. Raich and A.O. Hero, "Sparse image
reconstruction using a sparse prior," ICIP 2006
R. Raich and A.O. Hero, "Sparse image
reconstruction for partially unknown blur
functions," ICIP 2006
14
IRIS Illustration Sensor Illumination
Chair
Table
Initial Sensor Position/Configuration
Transmitter
Sink
Weapons Cache
Point Scatterer
Wall
15
IRIS Illustration Confidence Map
Iterative image reconstruction (FesslerHeroTIP95
) Sparsity constrained deconvolution (Nowaketal
TSP03) Image confidence maps (RaichHeroICASSP0
6)
Sink
Weapons Cache
Wall
Low confidence region
16
IRIS Illustration Virtual back-illumination
Sink
Weapons Cache
Wall
17
IRIS Illustration Predict Energy/Resolution MAP
Sink
Weapons Cache
Wall
18
Sparse image reconstruction and confidence mapping

• MAP-EM Formulation
• Separates deconvolution from denoising
• EM-MAP iterations for image x and confidence
  map
• Properties
• Iterates monotonically increase likelihood
• Deconvolution (E) only involves adjoint of
  forward operator
• Fast implementation with wavenumber migration
  approx for H

Ting,M, Hero,A.O., Sparse Image Reconstruction
Using a Sparse Prior, ICIP 2006.
19
Sparse Reconstruction Example

20
IRIS illustration for STW
Accessible Region
Simulated Scene
Inaccessible Region
Inaccessible Region
External Wall Permittivity 10
Thickness 0.2m Length 10m

Accessible Region
21
IRIS for STW Iteration 1 Sparse
reconstruction
Standard Wavenumber Migration
Sparse iterative Reconstruction (10
iterations) (MarbleRaichHero06)
1m Aperture
1m Aperture
22
IRIS for STW Iteration 1 Confidence Mapping
Sparse Prior
w 0.25 a 0.5

• Confidence Map shows pixels
• that have high confidence of
• being empty space.
• Quantitative map

Ambiguous pixels
23
IRIS for STW Iteration 1 Insert virtual
transmitter and simulate field
KL Mapping
Energy Mapping
24
Spectral Information Gain
KL Divergence Information Gain
Reference Field
Observation Location

• Electric Field From
• Transmitter k.

Horizontal Perturbation Field
Virtual Transmitters
Div Map Div(E1,E3) Div(E1,E2)
Vertical Perturbation Field
KL Divergence is a measure of Discrimination
Error Probability
25
IRIS for STW Iteration 2 Insert virtual
transmitter and simulate field
26
IRIS for STW Iteration 3 Insert virtual
transmitter and simulate field
Energy Mapping
Cross Range m
27
IRIS for STW Comparisons to fixed aperture
1m
1m
2
3
1
4
1m Aperture
10m Aperture
1m
28
Personnel on A. Heros sub-Project (2005-2006)

• Raviv Raich, post-doctoral researcher
• BS Tel Aviv University
• PhD Georgia Tech, May 2004
• Neal Patwari, post-doctoral researcher
• BS Virginia tech
• PhD, Univ of Michigan, Sept. 2005
• Doron Blatt, 4th year doctoral student
• BS Univ. Tel Aviv
• PhD Univ of Michigan, May 2006
• Raghuram Rangarajan, 5th year doctoral student
• BS IIT Madras
• Dept. Fellowship/MURI GSRA
• Jay Marble, 5th year doctoral student
• BS UIUC
• MURI GSRA
• Presently employed at NVRL

29
Pubs Since June 2005

• Theses of students funded on MURI
• "Performance Evaluation and Optimization for
  Inference Systems Model Uncertainty, Distributed
  Implementation, and Active Sensing," PhD Thesis,
  The University of Michigan, May 2006.
• Journal
• Adaptive Multi-modality Sensor Scheduling for
  Detection and Tracking of Smart Targets, C.
  Kreucher, D. Blatt, A. Hero, and K. Kastella,
  Digital Signal Processing,vol. 15, no. 4, July
  2005.
• "Multitarget Tracking using the Joint Multitarget
  Probability Density," C. Kreucher, K. Kastella,
  and A. Hero, IEEE Transactions on Aerospace and
  Electronic Systems, 39(4)1396-1414, October 2005
  (GD Medal winner 2005) .
• "Convergent incremental optimization transfer
  algorithms application to tomography", S. Ahn,
  J.A. Fessler, D. Blatt, and A. Hero, IEEE Trans.
  on Medical Imaging, vol. 25, no. 3, pp.283-296,
  March 2006

30
Pubs Since June 2005

• Conference
• "Sequential Design of Experiments for a Rayleigh
  Inverse Scattering Problem," R. Rangarajan, R.
  Raich, and A.O. Hero, Proc. Of IEEE Workshop on
  Statistical Signal Processing (SSP), Bordeaux,
  July 2005.
• "APOCS a convergent source localization
  algorithm for sensor networks," D. Blatt and A.O.
  Hero, IEEE Workshop on Statistical Signal
  Processing (SSP), Bordeaux, July 2006
• "Incremental optimization transfer algorithms
  application to transmission tomography", S. Ahn,
  J.A. Fessler, D. Blatt, and A. Hero, IEEE Conf
  on Medical Imaging, Oct. 2005.
• "A Comparison of Task Driven and Information
  Driven Sensor Management for Target Tracking," C.
  Kreucher, A. Hero, and K. Kastella, 44th IEEE
  Conference on Decision and Control (CDC) Special
  Session on Information Theoretic Methods for
  Target Tracking, December 12-15 (Invited)
• "Single-stage waveform selection for adaptive
  resource constrained state estimation," R.
  Raghuram, R. Raich and A.O. Hero, IEEE Intl.
  Conf. on Acoustics, Speech, and Signal
  Processing, Toulouse France, June 2006.
• "Optimal sensor scheduling via classification
  reduction of policy search (CROPS)," D. Blatt and
  A.O. Hero, 2006 Workshop on POMDP's,
  Classification and Regression (Intl Conf on
  Automated Planning and Scheduling (ICAPS)),
  Cumbria UK, June 2006. (Invited)

31
Synergistic Activities and Awards since June 2005

• Sensip Nov 2005, A. Hero plenary speaker
• Member of ARO MURI (John Sidles PI) awarded in
  2005 for MRFM sensing and image reconstruction
• Member of AFOSR MURI (Randy Moses PI) awarded in
  2006 for multi-platform radar sensing
• Member of ISP team (Harry Schmit PI)
• General Dynamics, Inc
• K. Kastella collaboration with A. Hero in sensor
  management, July 2002-
• C. Kreucher former doctoral student of A. Hero,
  continued collaboration
• M. Moreland Melbourne collaborator in area of
  sensor management
• Ben Shapo MS student collaborator in area of
  sensor management
• Mike Davis MS student collaborator in area of
  satellite MIMO
• ARL
• NRC ARLTAB A Hero is member of NAS
  oversight/review committee
• ARLTAB SEDD A. Hero participated in yearly
  review
• Night Vision Lab Jay Marble spent two weeks of
  Aug 2005 with Steve Bishop
• EIC of Foundations and Applications of Sensor
  Management (Springer - 2006)
• Contributor, IEEE Proceeedings Special Issue on
  Large Scale Complex Systems, Editor S. Haykin.

32
Synergistic Activities (ctd)

• In May 2005 UM Student Jay Marble was at
  Georgia Tech (working with Waymond Scott)

• In Aug 2005 Jay Marble was at Night Vision Lab
  (working with Steve Bishop)
• Indirectly support the Autonomous Mine
  Detection System (AMDS)
• Identify new data sets for algorithm
  validation Check Test 1 (April 2005)
• Apply multi-stage reinforcement learning
  algorithms to Army problems.
• Further develop demonstration software for
  illustrating algorithm performance.
• A. Hero visited AFRL Rome (B. Bonneau) in Nov.
  and gave invited presentation at Sensip on sensor
  management and at the Old Crows Conference at
  AFRL.
• Collaboration with Eric Michielsson on IRIS
  started in fall 2005 led to several proposals
  to DARPA, ARO.

33
Transitions

• Transition of SM methods to control of sensor
  swarms (GD) resulted in GD sensor net demo.
• Marble visited NVRL for 1 month in summer 2005 to
  demo UM mine detection software
• June 2006 Marble is now full-time employee at
  Night Vision Research Laboratory