<< Return to Webinar Files
Webinar Presentation
HTML version of the presentation
Image descriptions are contained in brackets. [ ]
Preparing for the Future: Collaborative ADS/IOO Framework
(May 9, 2023)
The Future Begins Here: ADS/IOO Collaborative Framework
Presenter: John Harding
Presenter’s Org: USDOT/FHWA Office of Transportation Management
T3 webinars are brought to you by the Intelligent Transportation Systems (ITS) Professional Capacity Building (PCB) Program of the U.S. Department of Transportation (USDOT)’s ITS Joint Program Office (JPO). References in this webinar to any specific commercial products, processes, or services, or the use of any trade, firm, or corporation name is for the information and convenience of the public, and does not constitute endorsement, recommendation, or favoring by the USDOT.
[The slides in this presentation contain the USDOT Federal Highway Administration (FHWA) logo.]
Slide 1: The Future Begins Here: ADS/IOO Collaborative Framework
PCB Webinar – May 9, 2023
John Harding, Team Leader Connected/Automated and Emerging Technologies
Office of Operations
[This slide contains an aerial view of six cars traveling in various lanes on a highway. Four of the cars are surrounded by concentric circles to indicate that they are connected.]
Slide 2: Disclaimer
The U.S. Government does not endorse products or manufacturers. Trademarks or manufacturers’ names appear in this presentation only because they are considered essential to the objective of the presentation. They are included for informational purposes only and are not intended to reflect a preference, approval, or endorsement of any one product or entity.
This presentation was created and is being presented by the Federal Highway Administration (FHWA). The views and opinions expressed are those of the presenter and do not necessarily reflect those of FHWA or the United States Department of Transportation (USDOT). The contents do not necessarily reflect the official policy of the USDOT.
Slide 3: Agenda
- History/Purpose
- Framework Introduction and Discussion
- Collaborative Taxonomy
- Success Stories
- Q&A/Discussion
Slide 4: Work Zones: A Common Sight Across the U.S.
- Most drivers will have no trouble steering through the lane shift without hitting or crossing the barriers.
- Is the same true for vehicles equipped with Automated Driving Systems (ADS)?
- What about with nonorange barrels, faded lane markings, adverse weather, or low lighting?
- How can ADS testing impart a higher level of public confidence in driverless vehicles?
[This slide contains a graphic of a car passing through a lane shift.]
Slide 5: Collaboration Adds Diverse, Complementary Perspectives
- ADS Developers
- View a system of external sensory inputs
- Speak a language related to software engineering
- Support an iterative testing model
- Infrastructure Owner/Operators (IOOs)
- View a system of safety rules and regulations
- Speak a language related to the Department of Transportation
- Support a comprehensive testing model
- Both aspire to see safe and efficient deployment of ADS!
- Likelihood of success is contingent on productive industrywide collaboration.
- The Federal Highway Administration (FHWA) has recognized that:
- The Federal Highway Administration (FHWA) has recognized that:
- Collaboration between ADS and roadway stakeholders is not adequate.
Slide 6: Collaborative Research Framework for ADS Developers and IOOs
[This slide contains a graphic that is described below.]
- Collaboration
- Institutional/Organizational Issues
- Checklists
- Influence/constraint
- Policies
- Common Ground
- Goals and benefits
- Terminology
- Metrics and measures
- Roles and Responsibilities
- Automated Driver
- ADS capabilities
- ADS responses
- ADS performance
- Success Factors
- Operational readiness
- Definition of test success
- Test Logistics
- Scale
- Where to test
- What to test
- How to test
- Plans
- Stakeholder Engagement/Agreements
- Quality
- Test
- Safety
- Data Management
- Public Road Testing/Permitting/Insurance
- Public Communications
- Risk Management
- Sharing Opportunities
- What to share
- How to share
- When to share
- Pretest
- Stakeholder engagement
- Collaboration opportunities
- Programmatic considerations
- Test Definition
- Problem statement
- Common goals and benefits
- Definition of test success
- Test Execution
- Operational collaboration
- Ongoing communication
- Monitor and adjust
- Posttest
- Data sharing opportunities
- Lessons learned
- Process improvement/calibration
Slide 7: “The Framework’s” Objectives
- Develop a collaborative testing and evaluation framework for ADS technology and the roadway environment.
- Create a safe transportation network that is inclusive of ADS-equipped vehicles.
Key Items
- Foster Collaboration
- Understand Capabilities
- Enable Deployment
Slide 8: Framework Development
- This Framework was developed with extensive engagement and input from both ADS and Roadway stakeholders.
- Stakeholder engagement included document/concept reviews, webinars, and 1:1 interviews, which allowed the project team to get a broad perspective on:
- Ways to foster collaboration
- Concerns and needs during ADS and roadway infrastructure testing
- Essential elements in collaborative ADS testing and evaluation
Slide 9: Framework Language and Taxonomy
- Facilitates common ground by creating a taxonomy for collaboration:
- Compatible with SAE J3016_20104
- Addresses interaction between ADS Vehicles and the surrounding World
[This slide contains a flowchart depicting the Framework Language and Taxonomy, which shows that the ADS Vehicle consists of the Sensing Unit(s), the Processing Unit(s), and the Vehicle Platform, and the World as the Roadway, Environment, and Objects. ]
Slide 10: The Framework
- The Framework addresses nine over-arching themes that represent the key elements of the framework.
- These overarching elements are applied to the different test phases along with contextual examples and real-world lessons learned.
[This slide contains a smaller version of a portion of the image in Slide 6: a flowchart of the nine key elements of the Framework: Collaboration, Institutional/Organizational Issues, Common Ground, Roles and Responsibilities, Automated Driver, Success Factors, Test Logistics, Plans, and Sharing Opportunities.]
Slide 11: Overarching Themes
- The nine overarching aspects represent the key elements of the Framework.
Overarching Themes
- Collaboration
- Institutional/Organizational Issues
- Common Ground
- Roles and Responsibilities
- A New Driver
- Success Factors
- Test Logistics
- Plans
- Sharing Opportunities
[This slide contains nine icons representing the nine key elements of the Framework: three people in conversation, a flowchart, planning documents, a circular flowchart diagram, a target, two computer monitors with arrows pointing to each other, an automated vehicle, a checklist, and a handshake.]
Slide 12: Collaboration
- There are many reasons why collaboration between ADS and IOO stakeholders is critical for successful testing and evaluation.
- One prominent reason is that stakeholder collaboration allows for early detection and resolution of ADS issues related to technical, organization, and strategic test implementations.
[This slide contains an icon of three people in conversation, the symbol for collaboration.]
Slide 13: (No title)
Common Ground
- Common Ground refers to creating a common or a shared working environment so that ADS and IOO stakeholders fully understand each other.
- There are three key components of Common Ground:
- Common goals and benefits
- Common terminology
- Common metrics and measures
Institutional/Organizational Issues
- Since ADS is a relatively new and evolving technology, testing and evaluation will face challenges from multiple fronts.
- Having organizational experts from both the ADS and IOO organizations participate early and throughout the test phases will greatly aid in navigating challenges.
[This slide contains two icons: a handshake, the symbol for common ground, and a flowchart, the symbol for institutional/organizational issues.]
Slide 14: (No title)
A New Driver
- The new driver of tomorrow will be the vehicle.
- Successful testing and operation of ADS-equipped vehicles will address how these new drivers:
- Operate
- Perform
- React
- Adjust
Roles and Responsibilities
- In the process of ADS/Roadway testing and evaluation, it is important to identify who from the various organizations needs to participate, what roles within the organizations are needed, and when and in which phases they are needed.
- Some participants may be involved in only one phase, while others may be essential to all phases.
[This slide contains two icons: an automated vehicle, the symbol for automated drivers, and a circular flowchart, the symbol for roles and responsibilities.]
Slide 15: (No title)
Test Logistics
- Test logistics refers to where to test, what to test, and how to test.
- This includes development of test scenarios, testing methodologies, and test environment.
Plans
- Efficient and effective ADS testing and evaluation is an iterative process that benefits from collaboration among multiple stakeholders, particularly the ADS developers and IOOs.
[This slide contains two icons: a checklist, the symbol for test logistics and planning documents, the symbol for plans.]
Slide 16: (No title)
Sharing Opportunities
- Data is a key issue that requires thorough discussions with IOO and ADS stakeholders to avoid challenges.
- Resource sharing includes sharing of skills and expertise in addition to sharing of information and existing data:
- ADS Vehicle Data:
- Raw feed, processed data, actionable items
- World Data:
- Roadway, Environment, Objects
Success Factors
- Success comes in many shapes and sizes.
- Each phase of testing will have desired and anticipated outcomes.
- The Framework aids in defining test success factors within each test phase.
[This slide contains two icons: two connected computer monitors, the symbol for sharing opportunities, and a target, the symbol for success factors.]
Slide 17: Pretest
- Conduct public outreach.
- Identify stakeholders.
- Negotiate data sharing.
- Assess integration challenges.
- Determine testing grounds.
- Obtain proper authorization.
Objective |
Conduct activities that help identify a problem. Evaluate collaboration opportunities for successful testing and evaluation. |
Inputs |
Technology of interest, initial stakeholder contact |
Key Activities |
- Programmatic considerations
- Stakeholder engagement
- Collaboration opportunities
|
Outputs |
On completion of this phase, you can expect to obtain a clear definition of the goals of collaborative testing. Additionally, you will be able to determine the agreements under which the stakeholders can expect to collaborate. |
Plans |
- Stakeholder engagement/agreements
- Permits, requirements, and insurance
- Preliminary risk management plans
- Preliminary public communications
|
Slide 18: Test Definition
- Develop test plans.
- Identify key metrics.
- Define evaluation criteria.
- Determine data management requirements.
- Schedule resources.
Objective |
Develop a comprehensive test plan, which captures the technical, data, evaluation, and quality facets of collaborative ADS/Roadway testing. |
Inputs |
Problem definition from the Pretest Phase, stakeholder collaboration assessment |
Key Activities |
- Common goals and benefits
- Problem statement
- Definition of test success
|
Outputs |
On completion of this phase, you will obtain a comprehensive test plan. You may use the test plan to facilitate test execution. You will also develop a data management plan (DMP), a quality plan, and definition of test success. |
Plans |
- Test plan
- DMP
- Quality plan
|
Slide 19: Test Execution
- Execute test plans.
- Monitor progress.
- Identify gaps.
- Adapt to changes.
- Communicate issues.
Objective |
Stakeholders execute testing per the defined test plan. They monitor and adjust program direction to meet defined goals and communication on an ongoing basis. |
Inputs |
- Test plan, DMP, quality plans
- Permits and insurance
|
Key Activities |
- Operational collaboration
- Ongoing communications
- Monitor and adjust program direction
|
Outputs |
- Collaborative ADS/Roadway testing and evaluation data
- Deviations in ADS performance and test execution
|
Slide 20: Posttest
- Analyze data.
- Document conclusions.
- Evaluate processes.
- Share results.
Objective |
Extract insights from testing and evaluation data and share these findings with stakeholders. Conduct project closeout activities that end the collaborative testing and evaluation. |
Inputs |
- Collaborative ADS/Roadway testing and evaluation data
- ADS/Roadway performance and test execution not meeting design intent
|
Key Activities |
- Data sharing
- Process improvement and calibration
- Lessons learned
|
Outputs |
- Testing and evaluation data insights
- Lessons learned for future collaborations
|
Slide 21: The Framework Document
- Uses over-arching theme icons throughout Framework phases to highlight opportunities when these components may be useful
- Presents real-world examples of collaborative ADS and IOO testing
- Identifies checklists for various elements
- Utilizes representative scenario examples to highlight concepts
- Includes taxonomy and data appendices
- Accompanied by a Quick Reference Guide as a high-level overview of the Framework
Link to the Collaborative Research Framework for ADS Developers and IOOs
https://ops.fhwa.dot.gov/publications/fhwahop21012/fhwahop21012.pdf
Slide 22: ADS/Roadway Perception Functionality
- Purpose: To develop a functional understanding of ADS and infrastructure testing and evaluation and clearly describe it
- Adds to framework development, to include:
- Understanding of a joint ADS Vehicle/World system
- A language and taxonomy for the ADS Vehicle and World modules and their parameters
[This slide contains three images: (1) a flowchart showing how the ADS Vehicle interacts with the World, (2) a diagram of the three parts of the World: Roadway, Environment, and Objects, and (3) the first flowchart showing the three parts of the ADS Vehicle: Sensing Unit(s), Processing Unit(s), and Vehicle Platform.]
Slide 23: Taxonomy Breakdown
- ADS Vehicle Architecture:
- Sensing
- Processing
- Vehicle Platform
- World
- Work Zone Example:
Slide 24: ADS Vehicle Architecture
[This slide contains a Functional Level 2 Detail diagram of the ADS Vehicle Architecture. The information in the diagram is reproduced below.]
|
Raw Sensor Feed |
Processed Data and States |
Controlled Variables and Planned Actions |
World |
Sensing Units
- Scene Information
- State Information
|
Processing Units
- Localization
(Where is the ADS?)
- Sensor Function
(What is around the ADS?)
- Semantic Understanding
(What does this mean?)
- Decision and Control
(What should the ADS do?)
|
Vehicle Platform
- Brake, Steer, and Accelerate commands sent to actuators
|
World represents both internal and external to the ADS Vehicle |
Sensors collect Scene and State Information about the World |
Single Block/Distributed |
|
Slide 25: ADS Sensing Unit – L3 and L4 Scene Info
[This slide contains a diagram of the Level 3 and Level 4 Scene Info of ADS Sensing Unit(s). The information in the diagram is reproduced below.]
- Sensing Units
- Scene Information
- ADS Vehicle
- Objects
- Roadway
- Environment
- Cloud Cover
- Roadway Lighting
- Subjective Atmospheric Parameters
- State Information
Functional Level 3 consists of ADS Vehicle, Objects, Roadway, and Environment.
Functional Level 4 consists of Cloud Cover, Roadway Lighting, and Subjective Atmospheric Parameters and a rectangle under Level 3 that is labeled “Shape, Size, Pose/Orientation, Feature, Unique Information.”
Slide 26: ADS Sensing Unit - L3 and L4 State Information
[This slide contains a diagram of the Level 3 and Level 4 Scene Info of the ADS Sensing Unit. The information in the diagram is reproduced below.]
- Sensing Units
- Scene Information
- State Information
- ADS Vehicle
- Objects
- Roadway
- Environment
- Cloud Cover
- Roadway Lighting
- Subjective Atmospheric Parameters
Functional Level 3 consists of ADS Vehicle, Objects, Roadway, and Environment.
Functional Level 4 consists of a rectangle under Level 3 that is labeled “Location/Orientation/Form Information, Motion Information, and Condition Information.”
Slide 27: Work Zone Navigation
- ADS with lane centering with adaptive cruise control (SAE L3) that:
- Detects a work zone ahead
- Safely follows construction barrels/cones in work zones
- Test goals:
- Understand how an ADS vehicle reacts when approaching a work zone
- Understand how cone placement affects ADS performance
- Understand how lighting affects ADS performance
[This slide contains a diagram of a car’s path through a lane shift.]
Slide 28: Work Zone Navigation - Program Definition Phase
- Test scenario: Joint ADS/World System
- ADS-equipped subject vehicle:
- ADS user (test speed of 45 mph) activates the L3 system before entering a construction zone with a lane shift.
- ADS user does not provide manual inputs to accelerator, brake, or touch steering wheel.
- ADS detects other road users in path but may be confused when objects are in redirected route.
[This slide contains a diagram of a car’s path through a lane shift.]
Slide 29: ADS Processing Unit(s) - High Level
Functional Level 2
[This slide contains a high level diagram of the ADS processing unit(s) showing Localization, Sensor Fusion, Semantic Understanding, Decision and Control (Trajectory, Energy, and Platform Fault Management), and Vehicle Execution Platform.]
Source: Behere, S. and M. Torngren. 2015. “A Functional Architecture for Autonomous Driving.” In 2015 First International Workshop on Automotive Software Architecture (WASA), pp. 3-10.
Slide 30: ADS Processing Unit(s) - Sensor Fusion
[This slide contains a diagram of the Sensor Fusion section of the ADS Processing Unit(s). The information in the diagram is reproduced below.]
Functional Level 3: Processing Unit - Sensor Fusion
- Sensor Fusion
- ADS Vehicle-related functions
- Internal Scene Information
- State Information
- Condition Information
- World-related Functions
- Object
- Detection
- Association
- Classification
- Motion Estimation
- Roadway
- TCD State Information
- Roadway Design Information
- Roadway Operations Information
- Roadway Condition Information
- Environment
- Atmospheric Data Fusion
- Road-Weather Data Fusion
- State of Connectivity
Fusion of relevant ADS Vehicle and Object data as a pre-step for High-Level Localization (Global Location, Traffic Flow, and Network Analysis)
Slide 31: ADS Processing Unit(s) - Localization
[This slide contains a diagram of the Localization section of the ADS Processing Unit(s). The information in the diagram is reproduced below.]
Functional Level 3: Processing Unit - Localization
- Localization
- ADS Vehicle
- World
- Objects
- Roadway
- Environment
- Strategic
- EGO Waypoint Information for Global Objectives
- Global Level Landmark Extraction
- Global Route Information
- Tactical
- Correct Lane for Global and Local objectives
- Nontravel Lane Object Position
- Nontravel Lane Object Motion Estimation
- TCD Nontravel Lane
- Nontravel Lane - Location & Tracking
- Operational
- In-Lane(s) Position
- Position w/rt Reference
- Traffic Information in Travel Lane
- In-Lane Object Motion Estimation
- In-Lane Object Position w/rt reference
- TCD applicable to Travel Lane
- Travel Lane - Location & Tracking
- Operational Information for Travel Lane
Atmospherics Information, Connectivity Information, Road-Weather Location Information
Creation of multilayered digital map from all information to determine location of ADS Vehicle and World entities
Slide 32: ADS Processing Unit(s) - Semantics
[This slide contains a diagram of the Semantics section of the ADS Processing Unit(s).]
Functional Level 3: Processing Unit - Semantic Understanding
- Strategic
- Road Network Modeling
- Traffic Flow Analysis
- Tactical
- Scene Modeling
- Stationary Environment
- Dynamic Environment
- Situation Analysis
- ODD* Analysis
- Predicted Environment Model
- Operational
- Object Classification
- Object Motion Estimation
- Understand Roadway
- Road Laws and Behavior
- ODD* Limit Analysis
*Operational Design Domain (ODD)
Slide 33: ADS Processing Unit(s) - Decision and Control
[This slide contains a diagram of the Decision and Control section of ADS Processing Unit(s). The information in the diagram is reproduced below.]
Functional Level 3 - Processing Unit Decision and Control
- Strategic
- Global Route Plan
- Navigation Waypoint Generation
- Tactical
- Behavior Planning
- Trajectory Generation
- Path Planning
- Operational Design Domain (ODD) Analysis
- Operational
- Longitudinal Motion Plan
- Lateral Motion Plan
- Vehicle Stabilization Assessment
- ADS & Vehicle Fault Monitoring
Slide 34: World Classification
[This slide contains a flowchart that is described in the multi-level list below.]
- World
- Roadway
- Roadway Design Information
- Roadway Operations
- Traffic Control Devices
- Environment
- Atmospherics
- Road-Weather
- Connectivity
- Objects
- Roadway Users
- Other Objects
Slide 35: Roadway Classification
[This slide contains a flowchart that is described in the multi-level list below.]
- World
- Roadway
- Roadway Design Information
- Roadway Types
- Surfaces
- Edges
- Geometry
- Roadway Operations
- Traffic Control Devices
- Signs
- Markings
- Traffic Control Signals
- Temporary Traffic Control Devices
Slide 36: Environment Classification
[This slide contains a flowchart that is described in the multi-level list below.]
- World
- Environment
- Atmospherics
- Road-Weather
- Connectivity
Slide 37: Object and User Classification
[This slide contains a flowchart that is described in the multi-level list below.]
- World
- Objects
- Roadway Users
- Other Objects
- Non-Roadway Users/Obstacles
- Off-Roadway Structures
Slide 38: Work Zone Navigation Feature - Program Definition Phase
World Description:
Essential Climate Variables |
Ambient Temperature |
32°F - 104°F |
Wind Speed |
Less than 22 mph |
Precipitation |
No inclement weather |
Visibility |
Distance |
3 miles |
Time of Day |
Full sun in line with vehicle |
World: Roadway: Roadway
Type of Roadway |
Highway |
Number of Travel Lanes |
Three |
Roadway Width |
N/A |
Lane Width |
10-12 ft |
Shoulder Presence |
Yes, but coned off/not accessible |
Surface Type |
Concrete |
Radius of Curvature |
Straight Lanes |
Grade |
Between Level and 1% |
[This slide contains a diagram of a car’s path through a lane shift.]
Slide 39: Work Zone Navigation Feature - Program Definition Phase
World: Roadway: Roadway - Traffic Control Devices
Signing |
Work Zone Ahead Sign
Reduced Speed Limit (45 mph) |
Lane Marking Type
Right to Left |
Solid White and Barrels
Initial Travel Lane (Ends)
Double Solid Yellow and Barrels
New Travel Lane
Dashed White and Barrels
Oncoming Travel Lane
Solid White |
Lane Marking Condition |
Retro-reflectivity of lane markings are degraded, other aspects meet or exceed recommendation |
Lane Marking Color |
Yellow and White mentioned as above
(Acceptable per National Institute of Science and Technology Reference) |
Lane Marker Width |
4-6 in |
[This slide contains a diagram of a car’s path through a lane shift.]
Slide 40: For Additional Information
John Harding
FHWA Connected/Automated Vehicles and Emerging Technologies
John.Harding@dot.gov
202‒366‒5665
Collaborative Research Framework for Automated Driving System Developers and Infrastructure Owners and Operators
Publication Number: FHWA-HOP-21-012
https://ops.fhwa.dot.gov/publications/fhwahop21012/fhwahop21012.pdf
↑ Return to top