www.mtri.org colin n. brooks, michigan tech research institute (mtri) christopher roussi, mtri dr....
TRANSCRIPT
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/unpavedRITARS-11-H-MTU1
Characterization of Unpaved Road Conditions through the Use of Remote Sensing
Friday, May 2nd, 2013 – 2nd 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 pagehttp://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. ArmyUnsurfaced 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
12Typical 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’
13Example flight at http://www.youtube.com/watch?v=KBNQzM7xGQo
Piotter Rd. and Garno CollectNovember 8, 2012
Helicopter Data – Piotter Rd.25 m Altitude
Other Example Image
Taken from 25m altitude, 2m/s (1st photo); 30m (2nd )
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
– Fly 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 examplesImportant to categorizing distresses by severity
Obtaining 0.9 cm ground sample distance
24
Input to Crown Measurement
Across Road
Along R
oad
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 [email protected]: 734-913-6858, Mobile: 734-604-4196
Michigan Tech Research Institute, MTRI 3600 Green Court, Suite 100
Ann Arbor, MI 48105www.mtri.org
Tim Colling, Ph.D., P.E. [email protected] Roussi [email protected] Rick Dobson [email protected] Dean [email protected]
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.