3 Speed Management
3.2.4 The Fuzzy Variable Speed Limit Device – Arizona
Impetus for the Activity
Northern Arizona’s crash frequency was increasing in the winter, and many of the crashes were attributed to “speed too fast for conditions.” In northern Arizona, roads average over 1524 m in elevation with several of the interstates reaching 2130 m. Drivers from all directions come from lower elevations in a short period of time and many do not expect to be driving on snow and ice. Arizona DOT (ADOT) thought VSL systems might be the answer, but wanted to somehow tie it to actual weather and road conditions. They wanted to automate the process at the same time to keep it from being subjective for each storm. Involvement from stakeholders was their highest priority and fuzzy logic fit this process perfectly.
ADOT has static signage, warning of ice, etc., but they are not allowed to change speed limits. Even if they were allowed to change speed limits, their current resources would not allow them to be timely in placing signs or folding them down, etc. They wanted a system that was independent, yet responsive and timely.
The expected result of deploying the system is to eventually have an enforceable VSL system.
Criteria for System Selection (Hardware and Software)
The fuzzy variable speed limit system was proposed by a Professor at Northern Arizona University. The concept for research was approved and funded by the Arizona Transportation Research Council for Phase 1. A slightly different funding approach was approved for Phase 2, using ADOT and University of Arizona funding.
Description of the System
The system is currently an experimental prototype, and assuming that adequate funding is received, trial deployments for field-testing will occur in the 2002-2003 fall/winter season. Baseline winter data collection and testing of the reliability of communications, VSL hardware components, and the sensor system will be performed during the fall/winter of 2001-2002. An assessment of the Arizona legal issues relating to the deployment and enforcement of VSLs on State highways will also be performed.
The objective of the project is to develop a VSL system that utilizes “fuzzy” control technology to identify speed limits appropriate for differing environmental conditions.
It is envisioned that this system can be deployed anywhere in any climate. Initial deployment is for roadways that experience winter driving conditions.
As inputs, the system uses road surface condition, average wind speed, wind gust speed, visibility, degree of cross wind, and precipitation intensity. The system also has a manual override input. These inputs are obtained from Road Weather Information System sensors. The inputs are averaged over a 10-min sampling period.
To generate system output - the posted speed limit - the system algorithm uses a “fuzzy logic” system of speed management. Fuzzy logic is multi-valued reasoning – decision-making is neither black nor white. The concept replicates human reasoning processes. Fuzzy logic uses imprecise categories, or a range of values with overlap, to determine the maximum prudent speed based on: road surface condition, average wind speed, wind gust speed, visibility, degree of cross wind, and precipitation intensity. The maximum prudent speeds are based on weighted averages of these components, established by highway agency and State police partners as the calculation basis. Emergency speeds can be set manually.
The posted speed limit is updated 10 min into the program. However, ADOT has not determined how often to change the actual speed limit signs that are shown to the public. That will be determined in Phase 3.
How the posted speed limit information will be made available to the drivers has yet to be determined. The first try might be with VMSs. The posted speed limits are currently advisory.
System Performance
A couple of factors affect system performance, including calibration and accuracy of RWIS system sensors, and timely repair and maintenance.
Phase 1 of the system performed as planned. The performance and reliability of the system had not yet been tested in the field, but will be examined under the next phase of research.
System Effectiveness Evaluation
The effectiveness of the system had not been evaluated since it had not yet been deployed.
Perceived “Hot Spots” for Rural ITS Deployment
Several perceived “Hot Spots” were noted. They included: speed limits, effective weather information, radar systems for knowing what is in front of you, and vehicle guidance systems.
Obstacles and Lessons Learned
ADOT noted a few obstacles. One was funding and another was deploying in a timely manner. ADOT stated that they need this system now, yet there are many hurdles yet to be met, the largest hurdle being their own resistance to trying something new and different. They also noted that law or statute changes may be needed by the Legislature.