Weather Applications and Products Enabled Through Vehicle Infrastructure Integration (VII)
Appendix B
VII Weather Applications Workshop I
Boulder, Colorado
Wednesday, June 21, 2006
Meeting Notes
This meeting focuses on new technology to determine where weather fits into the whole Vehicle Infrastructure Integration picture. The subject matter is problematic ranging from purely political to purely economical. We are focusing on the technical aspect but should be cognizant of the politics and economy; it is important that we know the constraints. An open and relaxed exchange of ideas is encouraged. We are in information-gathering mode and are anxious to hear what other people have to say. The number of opportunities and possibilities for VII are tremendous. The plan is to produce a VII weather applications feasibility report.
An ITS General overview presentation was given with focus on DOT vision for the VII initiative, the five phases of the Initiative, application development, Day-1 use cases, researching the data characteristics, key deliverables and outcomes of data characterization research, and building and testing applications.
Additional comments:
- Operators will benefit from the VII application for crash avoidance, the ultimate goal being to make driving safer. Pothole detection uses the same technology as some weather detection devices. The main thrust is near-term analytical emphasis on the support of Day 1 applications. Data uses can be "grown" using the tools we have, making them better and better.
A VII Consortium update presentation was given and included discussions on an overview of VIIC, near-term activities, the Proof of Concept test, contributions to weather information, probe data collection, probe data distribution, uses of weather information, an in-vehicle display application example, a navigation application example, development test environment for the west, central and east sections, and weather community participation
Additional comments:
- Any of the vehicle manufacturers can join. A cooperative agreement is in place between USDOT and VIIC. Ford, Nissan, BMW, Honda, DCX and VW are some of the manufacturers participating. VIIC concentrates on the physical elements installed in vehicles, and considers all vehicle equipment and applications. There will be a Proof of Concept test in 2007 with a fleet of 20 to 30 vehicles dedicated. Testing will begin with rudimentary data such as elevation, position, heading, time, speed, etc. and start with small amounts of information.
- VII takes in a wealth of information, and the primary purpose is to exercise the system, and its ability to do and not do certain things. Relevancy and human factors are not a part of this effort. The laying out work will take place after these exercises are completed. There are a number of engineering versus policy decisions to make. A proper balance between technology and privacy issues is necessary; in the public sector privacy needs to be taken into account.
- VII is already being used extensively, i.e. cell phones, etc. The premise is to choose a "dumb" on-board vehicle/"smart" off-board system; the on-board system can be informing the off-board of position. There will not be an ubiquitous connection from end to end - 2 environmental stations.
- An official VII test bed in California is working independently, and would be complimented by the NYC test bed. The issue of type of vehicles to use is being worked out, but probably heavy vehicles will not be used. Light trucks and vehicles are more likely candidates.
A VII system architecture overview was given and discussion included system architecture, VII system services and contact information. The system architecture chart was reviewed, as were national access points, and service delivery nodes.
A presentation on VII Probe message processes was given with emphasis on the intent, content, structure and header of probe messages, position and time, vehicle and weather status elements, the generation of snapshots, transmission, concept, examples and management of probe messages and current status.
Additional comments:
- Users of probe messages will be able to control how often and where to use information. It cannot be assumed that all cases have all elements, or that all data is coming from all cars. The data will be different from different manufacturers and numbers will be large, so items that are the same across the board should be selected. The intent is to connect to a consistent unit. Vehicles will be more consistent after the year 2012. Goals need to be attainable. Multiple data points, not just one, will be used to make a decision. The challenge is arbitration - deciding the appropriate time for messages.
- Periodic snapshots could be generated by distance alone, not speed. The owner of the road would be one of those who manage. Important snapshots could not be dropped. If there are no snapshots taken between stop and start, we know that it's because traffic is stopped or something is broken.
A presentation was given on the VII tasks, overview and update of data characteristics for traffic management and discussion focused on scope, objectives, staffing and coordination, approach, VII data characteristics task, key deliverables, the strengths and weaknesses of observed floating car and observed data sets, the strengths and weaknesses of simulated vehicle trajectories, a walk-through of the default VII probe message process, and preliminary observations and analysis.
Additional comments:
- Details such as spacing/deleting of snapshots, stops and starts, creep times and high congestion were discussed. The analytical process will be finalized in January.
A presentation on Weather Related VII Use Cases was given with focus on timeline, comment summary, primary changes, draft highly conceptual WDT, current status, improving weather observing and forecasting – Clarus drill-down and other USDOT Day 1 use cases
Additional comments:
- It's up to the operational entity to decide what the priorities will be.
A presentation was given on vehicles and mobile meteorological platforms with emphasis on distribution of ASOS observations, numbers, Mitretek research objectives, Mobile Wireless Laboratory (MoWL), sensor placement, test domains, vehicle platoon formation, data collection segments on the DTR, data synchronization and availability, single point data samples, data run history, DTR thermal profile (air temp), Question1 (Q1): data accuracy, Q2: temperature vs. vehicle speed, congestion modifying road temps, Q3: sensor placement, Q4: temp profiles of like vehicles, Q5: external weather effects, Manual Observations from the MoWL, External Effects, precipitation, and summary challenges.
Additional comments:
- The appropriateness of using probe data from passenger vehicles was discussed. The positioning of various probes on certain vehicles and its relationship to the types of data being collected is significant.
A presentation was given on human factors and VII-enabled applications with discussions including UMTRI, combining human factors and engineering domains of research, driver assistance systems research, origins of the naturalistic data, UMTRI naturalistic data with driver assistance systems, data acquisition and remote monitoring, recent FOT data scope, integrated data collection, data analysis and warehousing, video and visualization tools, overlaying vehicle data with crash and roadway data, naturalistic examination of windshield wiper usage, the data set, wiper utilization by month, wiper speed selection: ambient light level, naturalistic use of high-beam headlamps, high-beam usage, headlamp usage, results, naturalistic ABS events, the data set, ABS and precipitation/temperature, ABS and road class, ABS and speed, video samples, and the implications for VII weather.
Additional comments:
- Data is now ripe for mining, and there are heavily equipped fleets of vehicles with data acquisition systems in place. Wipers should not be a VII event even though wiper use can be modeled. Care needs to be taken to avoid using data improperly.
- Concerning the current state of the system, we do not know the initial state of the atmosphere very well. Greater numbers of observations will help with short-term forecasts but we cannot expect to improve on them by decreasing the grid size. New information acquired will lead to new science, which will complicate the decision making process and become a catalyst. The return value to customers is really the prognostic. Most customers do not expect anything right now and do not yet know about VII.
