Upload
trinhdan
View
216
Download
2
Embed Size (px)
Citation preview
Road Markings for Machine VisionNCHRP Project 20-102(6)
Update to AASHTO SCOTE
June 2016
NCHRP Project 20-102
• Impacts of Connected Vehicles and Automated Vehicles on State and Local Transportation Agencies– 20-102(01) Policy and Planning Actions to Internalize Societal Impacts
of CV and AV Systems into Market Decisions – 20-102(02) Impacts of Regulations and Policies on CV and AV
Technology Introduction in Transit Operations– 20-102(03) Challenges to CV and AV Application in Truck Freight
Operations– 20-102(05) Strategic Communications Plan for NCHRP Project 20-102– 20-102(06) Road Markings for Machine Vision– 20-102(07) Implications of Automation for Motor Vehicle Codes– 20-102(08) Dedicating Lanes for Priority or Exclusive Use by CVs and
AVs– 20-102(09) Providing Support to the Introduction of CV/AV Impacts
into Regional Transportation Planning and Modeling Tools
Cameras to Detect Markings
Adaptive Headlights
• Potential
Benefit
• Demonstrated
Benefit
Lane Departure Warning
• Potential
Benefit
• Demonstrated
Benefit
Highway Safety Statistics
• Roadway departure crashes account for over half of all fatal crashes (51%)
Potential LDW Safety
• Decrease fatalities by 7,529 fatal per year in the US– Journal of Accident Analysis & Prevention, Vol 43, 2011
• Reduce roadway departure crashes 26.1%– NHTSA
• Reduce serious injuries 20.7%– NHTSA
LDW Experiences
• Not effective as other vehicle safety technologies (IIHS 2014)
• “we don’t see any evidence that these systems are helping drivers avoid being in crashes” (IIHS 2014)
• 2/3 of Honda owners turn off their LDW feature (IIHS 2016)
– Annoying warnings
Early TTI Studies
• Simulated Conditions
– Sun glare
– Wet conditions
• Flat v. structured markings
Machine Vision
video
Challenging Conditions
• Uniform road marking maintenance thresholds• Preventive pavement maintenance treatments• Horizontal curves• Roadway lighting• Nighttime conditions• Wet conditions• Snow conditions• Debris• Poor marking removal
Historic Coordination
HighwayInfrastructure
Industry
Vehicle Industry
Standards Standards
Needed Coordination
Standards
HighwayInfrastructure
Industry(FHWA, AASHTO)
Vehicle Industry(NHTSA,
SAE)
NCHRP 20-102(6)
• Road Markings for Machine Vision
• Objectives
– develop information on the performance characteristics of pavement markings that affect the ability of machine vision systems to recognize them
– provide data and recommendations that the AASHTO/SAE Working Group can use to quickly develop guidelines and criteria
Work Plan
• Kick-Off Meeting
• Review Policies & Machine Vision Technologies
• Identify Testing Conditions
• Conduct Closed-Course Testing
• Analyze Results
• Prepare Reports
Current Testing Requirements
• ISO 17361:2007 – No requirements on types of road markings – Lane markings must be in good condition and in accordance
with the nationally defined visible lane markings std– No requirements on the environmental conditions – Visibility range must be greater than 1 km
• NHTSA– High contrast and uniform pavement– Lane marking specifications adhering to MUTCD– Avoiding tests in inclement weather including rain, fog, snow,
hail, smoke, or ash
Field Data
Texas A&M RELLIS Campus
Markings (Level 1)
Markings (Level 2)
Markings (Level 3)
Markings (Level 4)
Markings (Level 5)
Test Markings
• Possible Markings
– Continuous white
– Continuous yellow
– Skip white
– Skip yellow
– Contrast markings
– Raised retroreflective pavement markers
– Raised non-retroreflective pavement markers
Test Conditions
• Daytime
– Weather (dry, wet)
• Nighttime
– Conditions (dry, wet)
– Continuous roadway lighting (on or off)
– Vehicle lighting (tungsten-halogen, LED, other)
Next Steps
• Identify and obtain vehicle industry data
• Synthesize all data sources
• Develop up to 5 marking levels
– Color and retroreflectivity
• Provide 3M and Ennis-Flint specifications
• Install markings at RELLIS
• Conduct testing
Sensor-Enhanced Markings
• Pilot testing of enhanced detection technologies
Making It Happen