On the Feasibility of WiFi-Based Material Sensing October 24, 2019 - - PowerPoint PPT Presentation

▶
on the feasibility of
SMART_READER_LITE
LIVE PREVIEW

On the Feasibility of WiFi-Based Material Sensing October 24, 2019 - - PowerPoint PPT Presentation

Mar 29, 2023 360 likes •552 views

On the Feasibility of WiFi-Based Material Sensing October 24, 2019 Diana Zhang*, Jingxian Wang*, Junsu Jang, Junbo Zhang*, Swarun Kumar* *Carnegie Mellon University, MIT Media Lab Drones are increasingly useful in obstacle-rich

slide-1
SLIDE 1

On the Feasibility of WiFi-Based Material Sensing

October 24, 2019

Diana Zhang*, Jingxian Wang*, Junsu Jang†, Junbo Zhang*, Swarun Kumar*

*Carnegie Mellon University, †MIT Media Lab

slide-2
SLIDE 2

Drones are increasingly useful in obstacle-rich environments.

Urban Settings Warehouses Disaster Sites

2

slide-3
SLIDE 3

3

Drones must make obstacle-specific responses to maximize utility

The sensing system must be infrastructure-free and contained entirely on the drone.

slide-4
SLIDE 4

Obstacle Type Identification Non-Line-of-Sight Sensing

?

Vision-based (LIDAR, Infrared,…) RF-Based (WiFi, RADAR,…)

4

Current infrastructure-free sensing solutions cannot enable this.

slide-5
SLIDE 5

A complementary WiFi-Based sensing system that can detect material of obstacles in line-of-sight and non-line-of-sight settings.

  • Uses existing WiFi radio already on many drones
  • Does not assume infrastructure
  • Applies beyond drones – vehicles, product testing, disasters, etc.

5

IntuWition

slide-6
SLIDE 6
  • 1. Localization
  • 2. Material Sensing

wood metal (1, 2) (3, 6) (9, 2) human

  • 1. Localization
  • 2. Material Sensing

6

IntuWition comprises two major parts:

Tx Rx

slide-7
SLIDE 7

α wood =180°

Different α values can be used to distinguish materials

7

Radar Polarimetry can measure material-specific responses

α metal =90°

slide-8
SLIDE 8

To bring Radar Polarimetry to WiFi, a vertically polarized signal must be transmitted and received

8

α wood =180° α metal =90°

slide-9
SLIDE 9

Challenge #1: Multi-Bounce

9

Vs.

slide-10
SLIDE 10

Since the α values of multi-bounce are related to the single-reflection, these can be removed

10

Algorithm looks for alpha and locations that are consistent with physics of multi- bounce, to eliminate them as spurious (details in paper) α wood =180° α metal =90°

slide-11
SLIDE 11

Challenge #2: Several Variations in Material

11

slide-12
SLIDE 12

Solution: Devise machine learning models

ML model accounts for additional challenges: location, texture – details in the paper.

12

73 76 88 69 94 10 20 30 40 50 60 70 80 90 100 PCA kNN SVM NB MLP

Accuracy (%) Machine Learning Model

Material Classification Accuracy by Model

slide-13
SLIDE 13

IntuWition’s System Overview:

IN SUMMARY…

13

slide-14
SLIDE 14

We tested polarimetry as a material identification feature across a variety of materials and platforms.

14

slide-15
SLIDE 15

Our system showed high classification rates for five classes of materials.

5 classes, sheets of material

15

slide-16
SLIDE 16

Further, our system also worked well for classifying real-life objects as wood or metal.

Wood vs. Metal Classification of Real-Life Objects

16

slide-17
SLIDE 17

Limitations

  • Can’t detect signal when too weak, too occluded, or too many multi-bounce

effects

  • Cannot distinguish materials of similar polarization characteristics
  • May respond excessively to surface characteristics (e.g. clothing)

17

slide-18
SLIDE 18

IntuWition is a system that explores sensing the material and location of occluded objects

  • Uses COTS WiFi radios
  • Our evaluation demonstrates promising accuracy in material classification
  • Applies broadly beyond drones: vehicles, disaster response, product testing, etc.
  • Future work includes more objects, on-board processing, and sensor fusion

www.witechlab.com/intuwition

18