Edit: This is sarcasm. I don't expect anyone to try to translate this into anything.
Translate an old essay of mine.
Smart Traffic
Present Technology
Today, the coordination of traffic lights works on three basic systems: timed, sensor actuated, and sensor actuated with a timer backup. Timed systems tend to be used more on older, low-volume traffic grids. They are simple and reliable; however, they are poorly adapted for coping with very heavy or light traffic loads. During “rush hour”, timed systems may become overloaded with long queues of cars, while during a period of very light traffic, a solitary driver may be stuck at a red light for a minute or more even when there is no one else near the intersection. Sensor actuated grids tend to be much better at handling varying loads. Sensor/timer combinations are of differing versatility, depending on their implementation. For the most part, modern traffic light systems work well enough, though they definitely have room for improvement. What is most worrisome is their performance in the future if the general public does not start to use mass transit in abundance. If this does not happen, traffic coordination will definitely need upgrading.
History
The first traffic light was installed in London outside Westminster Palace on December 10, 1868. This had semaphore arms and gas lamps for night use and was manually controlled by a police officer. However, the first electrically-operated traffic light was invented in 1912 by Lester Wire of Salt Lake City, Utah, and one remarkably similar to the form used by present-day signals came eight years later, invented by William Potts. Since then, most traffic lights have used the same basic forum and color system – a vertical or horizontal box with three lights: green, yellow/amber, and red. Except for automation, only minor changes, such as adding tints of orange to the red and blue to the green to assist the red-green colorblind, have been widely adopted since 1920.
Future Technology
Considering the 20th-century trend from manual control (policemen at the corner) to timers to sensor actuation once the previous method was considered to have hit a dead end, it seems reasonable that traffic lights will continue to become more and more automated. However, this will require major efforts on the part of governments willing to undertake such a task. In contrast to simply changing out the traffic lights themselves, as was done in the past, increased automation will need major infrastructure modifications and additions. Our team’s idea for accurate tracking of traffic flow involves heavy use of computers and trilateration (subtly different from triangulation, which only works in 2 dimensions). The most common example of this is GPS (Global Positioning System) units, so that is what this paper will mostly refer to. Ideally, each car on the road would be fitted with a GPS or other accurate position transmitter, with different classes of vehicle as necessary – car, heavy truck, tractor-trailer, etc. (Older cars could be retrofitted.) During and after the process of installing these tracking units, global road infrastructure would need to be changed quite a bit. First of all, centralized, well-protected data collection centers would need to be built to handle the massive task of processing the signals. To make the task easier and more organized, the GPS trackers would constantly send coordinates to a small, local, receiver/transmitter battery, which would automatically and constantly relay the signals up a hierarchy all the way up to a larger control center. The relay batteries would be located about a mile away from each other in a roughly hexagonal overlapping array, preferably on buildings, towers, or other high vantage points. Inside the overlapping areas, the GPS units on the individual cars would send the radio signals to the battery with the least load on it. The aforementioned control center, after receiving the signals, would use a map stored in its internal memory to determine what roads were being used and in what volumes, and adjust light timings – the central premise of the technology - accordingly. For example, in Washington, D.C., at the intersection of 11th Street NE (a north-south road) and H Street NE (an east-west road), if 11th St was being used far more than H St for an evacuation to the south of the city, then traffic traveling on 11th St would be given priority of traffic signal timing over traffic traveling east-west on H St, while the lights signaling for a southward turn on H St would be given priority over turns from 11th onto H. In addition, in an emergency such as the evacuation example given above, it is conceivable that 11th St traffic flow could be modified so that both lanes flowed southwards, which would require human intervention at the traffic control center which had jurisdiction over that particular area (probably a single center for the entirety of Washington, D.C.) However, the system would be easily adaptable for awkward situations, examples of which are below.
- Emergency services: The proposed traffic system would also allow for “preemption” by emergency vehicles, such as fire engines and ambulances, using a special transmitter carried by those vehicles.
- Remote areas: On roads with extremely low volumes of traffic (such as the highways running through uninhabited areas of the southern United States), GPS tracking would not be needed. Instead, the last stations on either side would track outbound vehicles, look up the average speed of people driving on that specific road, and send it to the data collection center. From there, the information would go out electronically to the nearest center on the opposite side of the road, which would necessarily be located before any major junctions.
- Tunnels: Substations, the lowest in the hierarchy of relays, would be located inside tunnels. Multiple substations could be used in extremely long tunnels, such as the so-called “Chunnel” linking Great Britain and France. These substations would relay their data to the nearest transmitter battery by the most convenient method.
Breakthroughs
As with all “revolutionary” ideas, this idea is simply a plan until some major hurdles are cleared. Some are physical, but the majority of them are political and social.
- Signal frequency: As of late January 2007, the only unallocated (e.g. unused) radio frequencies in the United States are those between 3 and 9 kilohertz (KHz). However, this range of frequencies in the Very Low Frequency (VLF) range has extremely poor bandwidth, which is why the frequencies between 11.905 and 24 KHz are generally only intentionally used for navigational and time signal purposes, with a number of frequencies being used for unidentified purposes. Anything below 3 KHz is either audible or too low to be useful, and anything above 300 GHz is either infrared (interfering with infrared “heat-detection” systems), visible or harmful. An international standard range of frequencies for this system with very high bandwidth would be needed.
- Privacy concerns: Fears of the public over their personal privacy (e.g. “Big Brother” from the novel 1984) would in all likelihood be raised over a network like this. Those fears would have to be alleviated or ignored if this system were to take place.
- Tamper-resistance: Related to “privacy concerns”. People of certain mindsets would be attempting to find ways to interfere with the radio/GPS transmissions. Critical information, such as the locations of the GPS units on vehicles and the locations of the relay batteries, would have to be closely guarded.
- Satellites: More GPS satellites than the 24 currently in space could possibly be needed. Barring political difficulties, this could be fixed by combining the GPS (American), GLONASS (Russian), Galileo (mainly European), and Beidou (Chinese) satellite networks to make a total of 113 (24 + 24 + 30 + 35) usable satellites.
- Cost: Unless the GPS units, relay batteries, and computers to handle the data were mass-produced quickly, efficiently, and out of inexpensive materials, the cost of simply creating the proposed system would kill it before it even started.
- Global Positioning System (GPS): Currently, there are several sources of error for GPS location calculations. Taken together, they can result in a significant margin of error. In densely packed suburban and urban areas where streets are often within the publicly given 50m radius, this could lead to poor street recognition and result in flawed pattern timing. In addition, there is a little-known fact involving the accuracy of the positioning: the more accurate precise (P-) code, which can be up to 50 and on average 3 times more accurate than the “Coarse/Acquisition” (C/A) code available to the public, is very limited and is generally only available to the military. Getting them to let go of the code will likely not be an easy task, as shown in the 1990s when the military consistently refused to eliminate a feature called Selective Availability (SA), which introduced intentional errors of up to 100m into the public (but not military) signals. Finally, GPS signals, like all radio signals, can be jammed or misled. The effects of malicious intentional errors could stop city traffic cold. Thus we can conclude that the Global Positioning System in place needs a great deal of refinement before it is accurate enough to use for such a vital (to the Western and Eastern industrial world) system. In fact, other, more local trilateration systems may be in theory more suited for this purpose, although much more expensive. A system could refer to and use these instead of the GPS satellites, which by virtue of distance naturally introduce a larger margin of error. Since GPS is only the means to an end, it could be replaced with any other comparable location system, leaving the satellites free for other tasks.
- Traffic light control: In order for the entire system – car, GPS, relays, control centers, and traffic lights – to work, the traffic lights themselves would have to be changed to accept and use the signals from their respective data centers. This could perhaps take up to two or even three decades on rural roads.
Design Process
- One alternative design originally considered by our team was to, instead of using absolute location, have cars with laser scanners drive over bar codes painted on the road, and thus establish position relative to the road. From here on, the system would work much like the previously described final network. It was dismissed as impractical after considering the effects of wear and tear, weathering, and road maintenance.
- As noted at the end of the “Breakthroughs” section, local trilateration systems – which would work on the same principle as GPS – are also a possibility. This was seriously considered after the satellite option, and while more accurate, was eventually rejected too because the infrastructure did not already exist, while GPS satellites already are in space.
- One of the very first ideas was similar to this, but tracking airplanes and helicopters on air routes instead. However, this idea was dropped when we did some further research on the topic and found that instead of it being promising, it was actually already done.
Our team’s design process was not a long one. Once we had started, we met and discussed general ideas. About 20 minutes in, we decided that “smart traffic lights” was the best idea, seeing as it was unusual, pervasive in everyday life, and highly unlikely to have been used before. After this first meeting, we continued to develop and refine our ideas on our own. Eventually, after continually exchanging ideas over e-mail and instant messaging, we decided on this particular proposal.
Consequences
Points (pros):
Quicker travel times for most road travelers
Uses already-existing GPS satellites
Standardized worldwide
More accurate P-code already exists
Constant monitoring of traffic
Relay stations: reduce need for long-range transmissions
Accurate, worldwide GPS tracking of all vehicles
Radio signals
Counterpoints (cons):
Inconvenience for those on rarely-traveled roads intersecting busy roads
Clogs up GPS satellites
Single frequency range can be jammed easily
P-code is currently highly restricted
Privacy, surveillance, and totalitarian government concerns
Can be located and “bugged” or tapped
Must be carefully considered politically
Can be easily intercepted or jammed
For each point, there is a counterpoint. Some points may be stronger than their respective counterpoints, and vice-versa. The real and perceived strength of the points will be determining factor in any kind of future global or even widespread adoption of this proposal.
Last edited by k30dxedle (2007-04-04 12:30:18)