Technical Comparisons

Comparisons Between
Roadrunner RF and Other Sensor Systems

More Accurate Than Floating Car Studies (FCS)

;
floatingCar

Floating car studies–as distinct from Floating Car Data–were for years the standard for collecting travel time data. Probe vehicles were fitted with GPS devices that log coordinates or a passenger manually recorded travel times. Manuals on traffic studies recommended that in any particular study a vehicle passed as many vehicles as those that passed it. The study should be conducted in good weather, and it should not occur whenever an incident is present. For obvious reasons this isn’t practical for day-to-day, hour-to-hour, minute-to-minute operations. Furthermore, it consists of the smallest sample size possible: 1.

802_b

802.11b Waveform.

Roadrunner RF Matrix Sampling, however, is a floating car study, because measures are tracked via cell phone MAC address. It should be emphasized, however, that the concern here is not to trace an entire root. Roadrunner RF is a Matrix Sample tool: it breaks the test form, which is a proposed route, into a series of segments, which in GIS are called “Complete Chains.” A Complete Chain is a polyline with an associated Sensor Segment. At any particular time these Complete Chains can be assembled into Poly Chain Links, which can be considered as a complete transit from Point A in a metropolitan area to Point B, such as from the court house to the airport. These Poly Chain Links can be assembled into a round-trip polygon: From the court house to the airport, through the airport past the terminals and back to the court house again.

802.11g Waveform.

802.11g Waveform.

Rather than representing a complete transit that might take 30 or 40 minutes (and the data would be 30 to 40 minutes old), Roadrunner RF presents the most current chains assembled into a current route and thus presents the most up-to-date information on the state of the system.

More accurate than Radar, Inductive Loops, Video Cameras, and Magnetic Sensors

Inductive loops were originally designed to measure a single incident of detection, a count or mathematical point. Six foot by six foot loops can be used to determine speed, as can the Honeywell chipset in magnetic sensors, using difficult algorithms. Video cameras can also measure counts, but aren’t particularly useful for measuring vehicle speeds. Nevertheless, the important distinction here is this: A count is a point, whereas the speed is a vector measurement, it has direction as well as magnitude. Using two vectors, one can calculate an average speed from Point A to Point B, but each of these vectors is disassociated from the other. There is no real correlation.

Roadrunner RF measures the amount of time it takes for an object with a MAC address to transit from Point A to Point B. It is, in effect, an Affine Space with certain properties: a distance x of a Complete Chain (road segment) and a specifically denoted speed of y. It is not an abstract mathematical construct, it is a measurement. And, most importantly, it is not a displacement measurement.

More Commonplace Than
Radio-Frequency Identification Toll Tags

RFID toll tag technology has been used for some time to record travel times, and it has the benefit of using existing toll tag infrastructure. The downside here is clear: most drivers don’t use toll tags. In those areas an RFID system wouldn’t be able to obtain a sufficient number of samples for up-to-the-minute reporting, much less daily logs, weekly data, and monthly aggregations. In areas where toll tags are common (SF Bay Area, Seattle, etc) the capital costs for a single detector site for a six-lane highway is estimated at $18,000 to $38,000, with annual operating costs ranging from $4,000 to $6,000 for each RFID reader site.

More Accurate & Detailed
Than Cellular Floating Car Data

Floating car data (FCD) rests on using the position data of smartphones. The position data are collated by the service provider via Code Division Multiple Access standards (CDMA); the Global System for Mobile Communications (GSM) or G2; Universal Mobile Telecommunications System (UMTS) or 3G, and General Packet Radio Service (GPRS), which is a packet service based on 2G and 3G. No special devices/hardware are necessary: every switched-on mobile phone becomes a traffic probe and is as such an anonymous source of information. The data retrieved, however, are coarse. Google Maps is visualized by the three colors red, orange and green. A red road points to a traffic jam or stop-and-go traffic, orange indicates heavy traffic and green points to clear roads.

The downside is obvious. It does not provide any quantifiable measurements, and thus isn’t useful for progressive timing and

  1. cannot provide current traffic characteristics for optimized signalization; and
  2. cannot quantify traffic flow or uncharacteristic surges due to accidents, road closures, or special events.

More Accurate Than
Magnetic Signature Re-Identification

Magnetic Signature Re-Identification is achieved by matching representations of the magnetic fields that cars generate at two points on a roadway. The representations are reduced to their particular features, such as the outlines of the automobile, and compared. Given test results of between 70% and 90% on preliminary tests for MSRI, this tagging method looks promising. Results, however, vary and the signatures are complex to interpret. The Seimens patent describes their methodology as follows:

“] Image after Seimens Patent (Veröffentlichungsnummer US20080294401 A1; Publikationstyp Anmeldung;Anmeldenummer; US 12/122,800; Veröffentlichungsdatum 27. Nov. 2008)

“…receiving an image that includes a vehicle; and constructing a three-dimensional (3D) model of the vehicle, wherein the 3D model is constructed by: (a) taking a predetermined set of base shapes that are extracted from a subset of vehicles; (b) multiplying each of the base shapes by a parameter; (c) adding the resultant of each multiplication to form a vector that represents the vehicle’s shape; (d) fitting the vector to the vehicle in the image; and (e) repeating steps (a)-(d) by modifying the parameters until a difference between a fit vector and the vehicle in the image is minimized.”

An alternative, using Sensys Networks sensors, is described in Michael T. Volling’s paper ARTERIAL TRAVEL TIME USING MAGNETIC SIGNATURE RE-IDENTIFICATION THEORY OF APPLICATION AND ITS DEPLOYMENT IN SAN DIEGO, in which he states:

“…typical arterials produce a non-Gaussian travel time distribution with a large variance to mean ratio. This is due to cars traveling through arterials with varying speeds, driving habits and the various impacts of traffic signals. A large number of re-identified vehicles are required to accurately describe these distributions. If the results are required in real time or near real time, a high re-identification or penetration rate is essential. For example, if one assumes that 25 valid “probe” data points are required for statistical significance in determining an arterial travel time measurement on an arterial with 500 vehicles per hour, it will require a system with 1% penetration 2.5 hours to determine this measurement, while it will take just over 5 minutes for a system with 50% penetration.”

Re-identification of magnetic signature, sanpshot, from Volling 2009.

Re-identification of magnetic signature, sanpshot, from Volling 2009.

Alternatively, Roadrunner RF matches the MAC address of  an 802.11 signal. It’s a tracking technology that harkens back to 1906 when Marconi first mounted a radio on a steam-driven car, and the Loran A, Loran B, and Loran C tracking systems pioneered first during World War II and perfected in the 1950s.