by
Gail Payne (MIT)
Bicyclists can benefit from Intelligent Transportation Systems (ITS) both directly and
indirectly. The following three ITS programs are most apt to benefit bicycling directly:
ATIS will help bicyclists and transportation planners in obtaining crucial bicycle transportation system information through an "Intelligent Bicycle Routing Expert System." ATMS will make bicycling easier through signal preemptions for bicyclists via advanced traffic sensors. AVCS will help reduce vehicle-bicycle accidents through obstacle avoidance, warnings/tutorings, automatic speed reductions and vision enhancements.
The following two ITS programs will benefit bicyclists indirectly:
APTS makes transit and bicycle coordination more attractive due to improved transit information availability through ATIS. ETC could be used to deter travelers from using their vehicles because it establishes a mechanism that allows highway administrators to increase road or congestion pricing.
In general, ITS programs help link the traveler, infrastructure and transport mechanism (e.g., bicycle, transit, vehicle, etc.). As stated in a report by the Transportation Research Board, ITS "provides an infrastructure for the collection and communication of information to support a coordinated, decentralized approach to the provision of transportation services." ITS improves the linkage between the user and the facilities by monitoring the transportation system and by conveying system information to travelers and to traffic management systems.
Note: Intelligent Transportation Systems (ITS) was formerly called Intelligent Vehicle Highway Systems (IVHS) in the United States. In Europe, ITS is called Road Transport Informatics (RTI) or Advanced Transport Telematics (ATT). In the ITS literature, bicycles fall under the category of "vulnerable road users."
ATIS strives to achieve dynamic route guidance in order to assist travelers with their route choice. For motorists, the main objective of such a system is to reduce unnecessary vehicle miles traveled (VMT). For bicyclists, the main objective of such a system is to improve bicyclists' safety.
Safety is a top concern for bicyclists because in most U.S. cities, a continuous bicycle transportation system does not exist. For instance, in the Minneapolis/St. Paul area, "76 percent of urban arterials and corridors have been rated substandard for bicycling." Currently, bicyclists use the "trial-and-error" method of figuring out one's route. Ideally, a bicycle-related ATIS system could provide advanced knowledge of existing bikeways and potential hazards. Knowledge Management, Inc. titles this type of system an "Intelligent Bicycle Routing Expert System" in an article written in the Transportation Research Record. The Intelligent Bicycle Routing Expert System would allow registered users, system managers and the traffic management system to post both static and dynamic bicycle-related information to be viewed via the Internet or public kiosks.
Concerning dynamic information, the traffic management system would post real-time data on traffic and road conditions. The preferred route-guidance system from a bicyclists' perspective is the infrastructure-based route-guidance system because information such as maps and system algorithms are stored at the roadside traffic control system as opposed to the autonomous route-guidance system which stores the information in vehicles. A PROMETHEUS (Program for European Traffic with Highest Efficiency and Unprecedented Safety) project is experimenting with a dual-mode route-guidance system which provides users access to information from both the traffic control and the in-vehicle systems. A bicycle routing system could also obtain traffic system information from this type of dual-mode system.
Input from users (e.g., bicyclists who use the bicycle routing system) allows for more up-to-date information on the system. Only registered users would have access to the system and their comments could be edited by the system managers before being posted on the system. User comments would also help inform transportation planners of bikeways in need of repair. User contributions are critical for such a labor-intensive system that requires minute details about routes (e.g., hazardous drainage gates and ferocious dogs).
The bicycle routing system would include the following dynamic information:
The system would act as a central forum for cyclists in the community. The posted information would range from bicycle-related events and tours to cycling companion programs and bicycle equipment for sale. The bulletin board would also provide an on-line bicycle routing system evaluation which would help to ensure that the system meets the needs of the users.
The system would provide a link to real-time public transit and carpooling information. Special attention could be given to bus routes that allow bikes-on-buses or buses that have bike racks as well as to train lines that allow bikes on the cars. The user would also be able to access information about a "Bike Share" or "Bicycle Buddy" program. These programs are similar to carpools in that bicyclists form riding groups for safety reasons. These same groups could also carpool together during the winter months or bad weather days.
Route The system would match riders' profile (e.g., bicycle speed), preference (e.g., bike lane vs. bike path), time of day, origin and destination with routing information. An example of a rider profile question is as follows: "Do you ride with children?" An example of a preference question could include: "Is heavy traffic a concern?" The routing information would consist of estimated trip duration given the current weather and road conditions, alternative routes, fastest or most direct route considering current conditions, corresponding transit routes and times, bicycle parking locations, etc.
Regarding static bicycle routing information, registered users and system managers would post static information about bicycle-related issues such as registration (e.g., bicycle and transit permit), route information and safety.
The system would provide bicycle registration and transit permit information and would also allow the user to register for them while on-line. On-line registration would help increase bicycle registrations which would in turn increase registration revenue and the identification of recovered, stolen bicycles. The registration mechanism would also increase intermodal access to bicyclists who use or would use transit.
The route information would rely on existing bikeway inventories and registered-user comments. The static route information would consist of locating railroad tracks for crossing purposes, unsafe drainage gates, aesthetics, neighborhood safety via crime statistics, speed limits, grade, pavement width, etc.
The system could post the rules-of-the-road such as hand signals, helmet usage and safety statistics which indicate dangerous locations.
In conclusion, an intelligent bicycle routing expert system manager would have to obtain the system information from a number of sources, primarily registered users, transportation planners and traffic-control systems. The system would be provided on the Internet for easy access to bicyclists before their rides. Bicyclists could also access the system via conveniently-placed kiosks at tourist, transit and bicycle activity centers, via carpool and transit agencies and through an 800 number.
ATMS strives to improve the operations of traffic systems. For the motorist, the main objective of ATMS is to reduce congestion and increase the throughput of the transportation system. For the bicyclist, the main objective of ATMS is to reduce the number of stops at intersections in order to maintain the bicyclists' momentum. Susan Jourdain, who writes about urban intersections, confirmed the importance of sustaining one's momentum when she states, "Cyclists generally choose the minimum energy route with a stop weighing heavily against a route." Momentum can be maintained by implementing detectors or advanced traffic sensors that respond to bicycles.
Traffic engineers should identify the location of the bicycle-specific detector via a bicycle logo painted on the pavement. Bicyclists would then know where to pass in order to trigger the signal. The bicycle logo would also help to improve the reliability of bicycle counts because more bicyclists would ride over the detector if they knew that it would help trigger, preempt or extend the signal. The placement of the detector is also crucial if bicyclists are to maintain their momentum. An ideal setback (from the intersection) would allow bicyclists to maintain their speed and would also encourage them to slow down to a reasonable speed when crossing the intersection. A study by Kell and Fullerton suggests a detector setback of 45 feet when vehicle miles per hour (mph) equals 15. Being that the average bicycling speed ranges between eight and fifteen mph, a maximum setback of 45 feet would be suitable. (Note that the Kell-Fullerton setback estimation is based on vehicle reaction times, deceleration rates, etc. so the comparison is not exact.)
There are three different types of conventional detectors: inductive loop, magnetic and magnetometers. Out of these three detectors, only the inductive loop is able to respond to bicycles because the other two detectors require a ferromagnetic vehicle to trigger them. Once triggered, engineers could design the loops to preempt signals in favor of bicyclists, to indicate the presence of cyclists or to extend the green light for bicyclists. If more signals could detect the passage or presence of bicycles then, cyclists would not have to lose momentum at every intersection. This added convenience of bicycle-oriented signal timings allows for not only a much more pleasant ride but a safer one, void of risk taking at intersections.
The long loop of the inductive loop detector can detect the passage and presence of vehicles whereas short loops can only detect the passage of vehicles. A modified long loop detector equipped with additional turns of wire, powerheads and angled powerheads would be able to detect bicycles. "To detect bicycles and smaller motorbikes, the wire is wound twice to give a double-layer design termed a "2-4-2" installation. The advantages are gained by the configuration of the wire; any "amplifier" can be used." (Figures 1 and 2:did not transfer) An additional longitudinal wire along the center of the configuration would improve bicycle or small vehicle detection. The advantages and disadvantages of the inductive loop are shown below in Figure 3. (did not transfer)
Ideally, advanced traffic sensors would be able to identify bicycles by transferring messages between transponders (or tags) located on bicycles and roadside receivers (or readers). This type of functionality could be called "Automatic Bicycle Identification" (as opposed to Automatic Vehicle Identification) and would act as a bicycle position tracking system similar to a vehicle position tracking system. It is difficult to decipher which advanced traffic sensors respond best to bicycles so, I simply listed the advantages and disadvantages of each advanced traffic sensor. (Appendix A: did not transfer)
With Automatic Bicycle Identification (ABI), an increased quantity of bicycle-related
information could be made available to transportation planners, intelligent bicycle
routing system providers and bicyclists. An ABI system could track the following
bicycle-related information: bicycle volumes, route origin, destination, length and exact
roadways and bikeways used, number of bicycle trips per day, average speed and speed
variation. In general, very few comprehensive studies have concentrated on
bicycle-transportation statistics so most urbanized areas have an inadequate understanding
of localized bicycle patterns. This additional bicycle-related information would help
transportation planners perform the following tasks:
AVCS warns drivers of potential hazards and relieves them of partial or total control over their vehicles. For the motorist, the main objectives of AVCS are to increase the convenience and safety of driving. For the bicyclist, the main objective of AVCS is to increase the safety of bicycling by reducing the number of vehicle-bicycle collisions.
The risk of riding a bicycle is much higher than driving or riding in a vehicle. "In Minnesota, fatality rates are more than 3 times the rates of automobiles, and injury rates are more than 41 times those for automobiles." Bicycle accidents are more common in urbanized areas yet, rural bicycle-related accidents are more severe. In urbanized areas, approximately 60 to 70 percent of bicycle accidents occur at intersections, according to several European studies. Out of these accidents, the majority took place at non-signalized intersections.
AVCS can help reduce bicycle-vehicle collisions by automatic intervention or warnings.
The in-vehicle mechanism is similar to an advanced cruise control and is even referred to
as an Autonomous Intelligent Cruise Control (AICC). Two-way communication would be
necessary to achieve optimal AVCS results and actually could be possible through the
coupling of inductive loops. AVCS functions can be categorized as follows:
Warnings to motorists are achieved through conflict-zone monitoring. Conflict-zone monitoring tracks "the total maneuvering zone of their vehicle as well as other trafficants and their possible trajectories to detect possible conflicts. This includes detection of obstacles (e.g., vulnerable road users and static objects)." In order for the conflict-zone monitoring and warnings to be effective, the driver needs sufficient time to respond yet, the response time is minimal in high-activity (e.g., urbanized) areas where the majority of accidents occur. Due to this accuracy problem, some ITS experts are skeptical about the success of monitoring and in-vehicle warnings. One ITS scholar writes, "Urban traffic is extremely varied and hazards, in the form of vulnerable road users, can appear at random, so no artificial intelligence to detect hazards, or forecast dangerous behaviour, can be regarded as totally reliable." Furthermore, warnings may distract the driver causing some experts to emphasize the importance of tutoring for certain instances. A tutoring system would inform drivers about mistakes after they occur. The driver would access the tutoring system at their own convenience. Despite these reservations, warnings to drivers could still be useful in detecting driver fatigue, in assisting with driving during low visibility (e.g., fog and darkness) and in tracking vulnerable road users in certain circumstances.
Speed reduction is attractive to cyclists on arterials with heavy traffic and high speeds since these roads tend to be more hazardous to bicyclists. One author states that "Injury reductions of up to 60-80 per cent are quite possible (Hyden, 1990)." Speed reduction is also of particular interest on residential streets or arterials where the street serves a multifunctional purpose. These streets act as a living space for neighborhood residents so speed enforcement for vehicles is of great concern. AVCS could help make these streets safer through the implementation of "an automatic speed-limiting function". Vehicles would be forced to travel at a designated maximum speed by way of roadside controls. Hazardous-speed monitoring is another mechanism that could be used to deter high speeds. "Current estimates indicate that the use of the speed monitoring device may save approximately 10,000 lives and 500,000 injury accidents per year." A speed-monitoring device would give warnings to drivers when their speed is in excess and then would proceed to issue tickets if the driver does not alter his behavior.
Obstacle avoidance mechanisms apply brakes or acceleration in order to avoid a
collision. These techniques will help improve bicyclists' safety by reducing the number of
bicycle-vehicle accidents. Early systems focus on in-vehicle mechanisms while later
systems (25 years or more) will rely on specially-equipped roadside facilities. Currently,
researchers are primarily exploring three different techniques:
Engineers with the European PROMETHEUS program have established a prototype anti-collision vehicle called ALERT and have also experimented with obstacle detection in the Eureka and VITA (Vision Technology Application) projects.
The PROMETHEUS experts describe three types of obstacle-detection needs:
For close range detection (or VisionBumper), the PROMETHEUS project uses inverse-perspective mapping "which effectively compensates the stereo disparities of the ground plane. After the mapping operation, the two transformed images are compared and local mismatches are interpreted as possible obstacle locations." Thus, the vision bumper detects obstacles that will immediately collide with the respective vehicle.
To detect other vehicles as potential obstacles (otherwise known as CarTrack), the PROMETHEUS systems have used mirror symmetry and neural network techniques. The mechanism of mirror symmetry is used to detect vehicle shapes at long-range distances. As the technology advances, experts could adapt it to recognize the shapes of bicyclists. A more technical explanation of mirror symmetry follows: "Mirror symmetry with respect to a vertical axis is one of the most striking generic shape features available for object recognition in a car-following situation. Initially, we use an intensity-based symmetry finder to detect image regions that are candidates for a leading car...Edge points that do not have a mirror symmetric counterpart are suppressed."
Neural networks also detect vehicles for obstacle avoidance. This neural object detection system simulate driver reactions to pre-designated situations. To assist with bicycling safety, scenarios could be devised to replicate human reactions in order to avoid colliding with a bicycle. The bicycle-friendly neural network collision avoidance system would configure algorithms that would contain rules on how to intervene when approaching a bicyclist in a hazardous manner. A more technical description of the neural network follows: "Recognition of a scenario is achieved by acquiring data about a scene from a variety of sensors. "Visual data is preprocessed and features are extracted using a real-time image processing system, while microwave radar provides obstacle information and distances."
For detecting obstacles alongside vehicles (e.g., SideView), wide-angle cameras are used. The PROMETHEUS project specifically uses fan cameras that "try to get information if there is anything in the field of view which cannot be part of the expected road surface. Both sides of the test car have a pair of fan cameras attached to the bodywork so that there are large binocular fields of vision..." These cameras would detect bicycles located alongside vehicles and would then indicate to the braking, accelerating and maneuvering systems an alternative course.
In conclusion, bicycling can directly benefit from ITS via three major systems: ATIS
(e.g., "Intelligent Bicycle Routing Expert System"), ATMS (e.g., Automatic
Bicycle Identification) and AVCS (e.g., warnings, automatic speed reductions and obstacle
avoidance). Bicycling can indirectly benefit by way of the following two ITS programs:
APTS would positively affect bicyclists because it makes transit easier to use. As transit becomes more appealing, ridership will increase and VMT will decrease. Fewer vehicles on the road allows for a much more aesthetically-pleasing bicycling experience. Furthermore, ATIS technology could be used to improve coordination between transit and bicycle transportation systems. The City of Denver, Colorado has implemented Denver SmartBus which will eventually include an ATIS system via kiosks, cable television or computer terminal.
Electronic Toll Collections (ETC) use wireless communications between roadside sensors and transponders mounted on vehicles, buses and trucks to track vehicles for the purpose of toll payments. This technology is more formally known as Automatic Vehicle Identification (AVI) and can be used for road- or congestion pricing policies. This technology is already in use on the Dallas North Tollway in Dallas, Texas, on the Broad Causeway in Florida, on the Grosse Ile Bridge in Michigan, etc.
In general, research on the bicycling benefits of ITS is sparse. Nevertheless, even though the concentration of ITS is primarily auto oriented, the actual implementation of ITS projects depends highly on the needs of individual localities. Communities like Davis, California or Boulder, Colorado, that are committed to bicycling, may find creative ways to use this technology to benefit their bicycling constituents. With air quality problems persisting in most major U.S. cities and with dwindling transportation funding sources, more localities may opt to choose ITS projects that help increase other modes of transportation besides the automobile.
Advanced Vehicle and Highway Technologies, Transportation Research Board, National Research Council, Special Report 232, Washington, D.C., 1991.
Manzoor A. Arain, Raglan Tribe, Edgar An, Chris Harris, "Action planning for the collision avoidance system using neural networks," ," Proceedings of the Intelligent Vehicles `93 Symposium, July 14 - 16, 1993, Tokyo, Sponsored by IEEE Industrial Electronics Society.
Joe Betz, Jim Dustrude, and Jill Walker, "Intelligent Bicycle Routing in the United States," Transportation Research Record, No. 1405: Operations and Safety, Pedestrian, Bicycle, and Older Driver Research, National Research Council, 1993.
Michael E. Brauckmann, Christian Goerick, Jurgen Grob, and Thomas Zielke, "Towards All Around Automatic Visual Obstacle Sensing for Cars," Proceedings of the Intelligent Vehicles `94 Symposium, October 24-26, 1994, Paris, France, Sponsored by IEEE Industrial Electronics Society, p.80.
M.Draskoczy and C. Hyden, "Cyclists' problems: can RTI help?", Driving Future Vehicles, Edited by Andrew M. Parkes and Stig Franzen, Published by Taylor & Francis, London and Washington, D.C., 1993.
Electronic Toll and Traffic Management (ETTM) Systems, A Synthesis of Highway Practice, National Cooperative Highway Research Program, Transportation Research Board, National Research Council, NCHRP Synthesis 194, 1993.
Generic Intelligent Driver Support: A Comprehensive Report on GIDS, Edited by John A. Michon, Taylor & Francis, London and Washington, D.C., 1993.
Intelligent Vehicle Highway Systems Review of Field Tests, Road Transport Research, Organization for Economic Co-operation and Development, 1992.
Susan Jourdain, Urban Intersection Control, The Book Guild Ltd., Sussex, England, 1992.
Alan R. Kaub and Thomas Rawls, "Automatic Speed Monitor: An Intelligent Vehicle Highway System Safe-Speed System for Advance Warning or Hazardous Speed Monitoring," Transportation Research Record, No. 1408, Highway Operations, Capacity and Traffic Control: Intelligent Vehicle Highway Systems, Transportation Research Board, National Research Council, 1993.
James H. Kell, Iris J. Fullerton, Manual of Traffic Signal Design, Second Edition, Institute of Transportation Engineers, Prentice Hall, Englewood Cliffs, New Jersey.
SPIE Technology Tutorial, Highway-Based Vehicle Sensors, Alan Chachich, Nov. 1, 1994.
Richard Whelan, Smart Highways, Smart Cars: Intelligent Vehicle Highway Systems, Artech House, Inc., Boston and London,1995.
| Go to top of this page, | | home page |