Characterization of Unpaved Road Conditions

snowpeaschocolateManagement

Nov 18, 2013 (3 years and 8 months ago)

76 views

www.mtri.org

Colin N.
Brooks,
Michigan
Tech Research Institute (MTRI
)

Christopher
Roussi
,
MTRI

Dr. Tim
Colling
,
P.E.
,
Michigan Tech Center for Technology and Training (CTT)

Caesar Singh
, P.E., US Department of Transportation (USDOT) Research & Innovative
Technology Administration (RITA)


www.mtri.org/unpaved

RITARS
-
11
-
H
-
MTU1




Characterization of Unpaved Road Conditions

through the Use of Remote Sensing



Friday, May 2
nd
, 2013


2
nd

Technical Advisory Committee meeting



Characterization of Unpaved Road Conditions

Goal of the Project:
Extend available Commercial Remote Sensing & Spatial
Information (CRS&SI)
tools to enhance & develop an unpaved road assessment
system by developing a sensor for, & demonstrating the utility of remote sensing
platform(s) for unpaved road assessment
.


Commercially viable in that it can measure inventory and distress data at a
rate and cost
competitive with traditional methods


Rapid ID & characterization
of unpaved roads


Inventory level with
meaningful metrics


Develop a sensor

for, &
demonstrate the utility

of remote sensing platform(s) for
unpaved road assessment


Platform could be a
typical manned fixed
-
wing aircraft, UAV, or both
; depends on
relative strengths & weaknesses in meeting user community requirements


Simplify mission planning
, control of sensor system, & data processing fitting for a
commercial entity or large transportation agency


Demonstrate prototype
system(s) to stakeholders for potential implementation
developed through best engineering practices


Develop a decision support system to
aid the user
in asset management and planning




2



Project web page

http://www.mtri.org/unpaved

http://www.mtri.org/unpaved
/


3



Project Partners

Partners:


Michigan Tech Center for Technology & Training:
Gravel roads & Decision Support Tool software
expertise


Transportation Asset Management Council of
Michigan (TAMC)


shared PASER data, provide
advice (briefed 1/9/13 on progress, pleased with
results


SEMCOG (Southeastern Michigan Council of
Governments)


shared aerial imagery, provide
advice, inventory needs


RCOC (Road Commission for Oakland County)


provide advice, local expertise on unpaved roads
management needs


USDOT
-
RITA


Program Manager, advice,
transportation expertise


Michigan Tech Research Institute (MTRI)


project lead, remote sensing, engineering, UAVs,
software coding, image processing

4



Assessment Method: Dept. Army

Unsurfaced Road Condition Index



Representative Sample Segment (approx. . 100’ long)


2 Part Rating System (per distress)


Density


Percentage of the sample area


Severity


Low


Medium


High



Drawback: typically takes significant time to


complete manual assessments by traditional


methods



Road Characteristics

6


Unpaved roads have common characteristics


Surface type


Surface width


Collected every 10', with a precision of +/
-

4”


Cross Section (Loss of Crown)


Facilitates drainage, typically 2%
-

4% (up to 6%) vertical change, sloping
away from the centerline to the edge


Measure the profile every 10' along the road direction, able to detect a
1% change across a 9'
-
wide lane


Potholes


<1', 1'
-
2', 2'
-
3', >3‘ width bins


<2”, 2”
-
4”, >4” depth bins


Ruts


Detect features >5”, >10' in length, precision +/
-
2”


Corrugations (
washboarding
)


Classify by depth to a precision of +/
-
1”


<1”, 1”
-
3”, >3”


Report total area of the reporting segment affected


Roadside Drainage


System should be able to measure ditch bottom relative to road surface
within +/
-
2”, if >6”


Detect the presence of water, elevation +/
-
2”, width +/
-
4”


Float aggregate (
berms
)






Inventory: Surface Type


How many miles of unpaved road are there? Not all counties have this.


Need to able to determine this inventory


c. 43,000 (1984 estimate)


but no up
-
to
-
date, accurate state inventory exists


c. 800 miles in Oakland County estimate


We are extracting this from recent, high
-
resolution aerial imagery, focusing on
unincorporated areas


attribute existing state Framework roads layer


Completed Oakland, Monroe Counties


ready to share with SEMCOG; working
on Livingston, St. Clair, Macomb, Washtenaw Counties

7




Motivation for Phenomenology
Approach: Understand
how the
physical properties of the road surface
distresses manifest themselves in
observable
ways



Color
(inc. need for balancing)



Texture



Patterns



profile
(inc. 3D structure)



Polarization



Sensor Nikon
D800


full
-
sized (FX)
sensor, 36.3 Mp, 4 fps
-

$
3,000; 55
mm prime & 105mm
lense
, 200 mm
planned

Sensing Unpaved Road Conditions

8



Flight factors for remote control aircraft


Forward speed
must be
low
to be able to image with
the required scene overlap at the maximum rate of the
sensor


Low speed → rotary wing aircraft, since fixed
-
wing
would stall


Must be able to loft 5kg of sensor, controller, and
batteries


Must be able to fly for 20min under full load,
we’re
staying below 100’, in
sight of safety
pilot;




Selected initial aircraft: Bergen
Tazer

800

10



Flight Safety & Effectiveness Inspection

Evaluate site for safe flight operations,
suitable for aerial collection


High
-
voltage towers, restricted airspace, visual
obstructions

Manned vs. unmanned:


Manned: licensed pilot review, FAA
regs

followed,
safety margins included


Unmanned: more possible instructions

11



Flight trajectory planning

Ground Station Control program / tool


create flight trajectory

Includes ability to automatically take off, fly, auto
-
land; operator has
joystick control at all times

Includes Google Earth / Maps information

12

Typical view of opening screen in Ground Station program



Data Collection


unmanned helicopter

Totally autonomous flight.

Flight time for a 200 m section: 4 minutes

During collects helicopter is flown at 2 m/s
and at an altitude of 25 m (82’) and 30 m
(98’)


FAA ceiling of 400’

13

Example flight at
http://www.youtube.com/watch?v=KBNQzM7xGQo




Piotter

Rd. and
Garno

Collect

November 8, 2012



Helicopter Data


Piotter

Rd.

25 m Altitude



Other Example
Image

Taken from 25m altitude,
2m/s (1
st

photo); 30m
(2
nd

)



Ground data being collected for all roads
being flown for assessment

17



Fixed
-
wing Choice


FAA restrictions on fixed
-
wing flight


>500ft altitude


Sensor cannot be attached to aircraft without FAA review


Any small aircraft meet SWAP and flight requirements


While charter costs can be up to $1600
-
$
2500/
hr
, we
flew last fall in a Cessna 172 for $280 for 1.2 hours of
flight time


F
ly
to site, collect data, and fly back


Trial flight 2012; more planned for 2013 after we
consulted with President of Professional Aerial
Photographers Association, Chuck Boyle


Recommended John Sullivan of AAP Inc. at Ann Arbor
Airport



Aerial Collect



Software Architecture


Because we are incorporating legacy code, third
-
party tools, and custom code, we need a flexible
architecture


Developed in C, C++, Python, bash


Flexible control, with tools calling each other as needed



Algorithm

Use Structure from Motion (SIFT+ Bundler +
PVMS) to turn 2D images into 3D point
-
cloud reconstruction


SIFT = scale
-
invariant feature transform


PVMS = patch
-
based multi
-
view stereo

Form a surface from the 3D point
-
cloud


Form grid, Fourier Filter, Marching Cubes to
triangulate

Find the depth/height map of the surface


Singular Value Decomposition (SVD)


Rotate so z
-
axis is

up


(depth)



Algorithm

Find and select the road in the scene


Image entropy measure (road is

smoother

)

Rotate extracted road into new coordinate system


Makes it easier to take cuts along and across road

Analyze for features of interest


Gabor Filtering, Circular Hough Transform, Cuts for
profiles of road and drainage

Convert to PASER
-
like metrics (Pavement Surface
Evaluation and Rating System)

Generate XML output suitable for
RoadSoft

GIS decision
support processing



Example 3D Reconstruction


15 images use to form point cloud

Bundler output

Densified

point cloud

3D surface from point cloud

Height
-
field from surface



3D data examples

Important to categorizing distresses by severity

Obtaining 0.9 cm ground sample distance


24



Input to Crown Measurement

Across Road

Along Road

Example
crossection

plot

(
vert

meters



URCI Density, Severity, Deduct

26

Distress Type

Density

Severity

Deduct

Value

Improper

Cross Section

5.6

L

13

Corrugations

50.0

M

29

Dust

NA

M

4

Potholes

1.44

M

34

Ruts

65

H

44

Total Deduct Value = 124

q = # of deduct values = or > 5

q

= 4



URCI Assigned Distress ID & Ranking in
the
RoadSoft

GIS DSS

27



Where next with the project?

Larger set of field deployments along rural roads (Del. 7
-
A)


Both unmanned RC helicopter and Cessna flights


Demonstrate
hexacopter

capability vs. single rotor helicopter


Cessna flights shooting at nadir (we have the door that can hold the
camera internally)


Integrating of results into
RoadSoft

GIS

Write up formal Performance Evaluation (Del. 7
-
B)


How well did we do? How capable is the system? Useful metric
generated… Where is technology going towards practical
deployment & usage,
inc.

cost?

Avenues for practical usage by transportation agencies


in
-
house model (buy equipment, software)


Contracted services model (company performs data collections &
analysis for transportation agency; end
-
to
-
end system licensed to
company)



28



29

Contact Info

Colin
Brooks
colin.brooks@mtu.edu

Desk: 734
-
913
-
6858, Mobile: 734
-
604
-
4196

Michigan Tech Research Institute, MTRI

3600 Green Court, Suite 100

Ann Arbor, MI 48105

www.mtri.org


Tim
Colling
, Ph.D., P.E.

tkcollin@mtu.edu

Chris
Roussi

croussi@mtu.edu


Rick Dobson
rjdobson@mtu.edu

David Dean

dbdean@mtu.edu



DISCLAIMER:
The views, opinions, findings and conclusions
reflected in this presentation are the responsibility of the authors
only and do not represent the official policy or position of the
USDOT/RITA
, or any State or other entity.