WelcomeTeachingPublicationsInfo for prospective studentsCitizen ScienceGeocachingLinksBioTAP

by Dmitry Serov

We live in a world in which your wristwatch can give you a stock quote or a weather forecast, in which robotic, human-like assistants help the elderly in nursing homes, and in which a robotic team is expected to be able to defeat the best soccer team in the world by the year 2015. Yet in this same world, commuters everywhere all too often utter the phrase “traffic was awful.” Traffic congestion limits the average speed on commuter roads near large U.S. cities to about 36 miles per hour during rush hour. This leads to approximately $50 billion in losses in national productivity (Ashley, 1998, p.1). In the past, the normal method of alleviating automotive congestion was building larger highways, yet this is no longer a realistic alternative as the financial and environmental burdens of such projects have become unbearable (Ashley, 1998, p.1). In our modern era of technological achievement, many researchers believe that the answer to this dilemma is artificially intelligent cars. Scientists have been dabbling in this field and examining the practicality of “smart cars” and automated roads for decades. This technology has been developed and experimented and while concerns have been raised, the potential benefits of an automated traffic system far outweigh these problems, which are expected to be resolved with continued research. As of now, the implementation of this innovation is on the horizon and such a development would revolutionize the world of transportation.

In this realm of technologically advanced forms of traffic, studies have examined everything from automated highways to “free agent” cars, or autonomous vehicles (Ashley, 1998, p.3). One such possibility, projected to triple highway capacity, is to bury magnets in the highway roadbed and in the cars themselves, install magnetometers, devices used to detect the magnets in the roads, to stay in the lanes and steer. Cars would travel in convoys, close together, communicating through wireless networks and constantly exchanging information about road conditions, speed, etc. This would all be closely monitored by computer stations set up along the highway. Cars driving this close together can cut the drag in half by drafting off each other, as drafting is the act of moving closely behind a fast-moving object so as to take advantage of its slipstream. A driver wishing to merge onto such a highway would first pass through a “verification lane”, where a computer would make sure everything is in order, establish the destination and charge toll fees from the driver’s account. The reverse process would also involve an in-between lane where the computer would make sure the driver is able to regain full control of the vehicle and take appropriate measures if the driver is asleep, injured, or dead (Ashley, 1998, p.2,3).

There are several major concerns about an automated highway program. One of these concerns is the production of such technology by commercial vendors. A constant framework needs to be developed for manufacturers to work within, but also a framework that doesn’t scare manufacturers away. Ivy Renga, manager of the Intelligent Vehicle and Highway System Program for Chrysler Corp. brings up the example of electronic toll booth passes. “Today, there are 20 companies with 20 versions of the technology, and none are compatible. Trucks must have a mosaic of electronic tags to use these systems when they go cross country (Ashley, 1998, p.4).” Thus, all manufacturers must make their automated driving products universally compatible. Furthermore, people are often hesitant to trust computers to such extremes. However, the beauty of this system is that even if something malfunctions and a collision occurs, damage would be minimal, as all the vehicles are driving at the same speed. Perhaps the most pressing concern at this juncture is the issue of cost. It is estimated that the installation of this technology would cost merely $10,000 per mile, an amount much lower than that spent on highway expansion. Thus, the catch is in vehicle costs, as the additional gadgets that control and monitor the car could increase its cost significantly (Ashley, 1998, p.4).

However, the potential benefits of such a system are great. Research done by the California Partners for Advanced Transit and Highways (PATH) at the University of California, Berkeley, indicates that this style of traffic automation could perhaps triple the capacity of certain highways. The capacity of a single highway lane could increase from 2,000 to 6,000 cars per hour (Ashley, 1998, p.4). More importantly, a great deal of the massive amounts of exhaust toxins emitted into the environment could be avoided. If traffic flows smoothly, vehicles running at a constant speed would reduce emissions and increase gas mileage. The drafting affect of the car convoys could decrease emissions by as much as 20%. Arguably the most important benefit of this system is its safety. In police reports, 90% of all traffic accidents are caused by driver error and most of these accidents are the direct result of driver fatigue. Furthermore, one third of all automotive accidents are single-car (Ashley, 1998, p.8). If the driver could be replaced by a computer that never gets tired and has senses far superior to those of any human, automobile deaths could be greatly reduced. Preliminary studies show that systems avoiding rear end, lane change, and off ramp crashes could potentially reduce accidents by 1.2 million a year (Ashley, 1998, p.8).

The most promising aspect of the idea of traffic and automobile automation is the fact that this technology is more realistic than most people realize. The Carnegie Mellon University Navigation Laboratory has been working on artificially intelligent cars since 1984. In 1986, they designed and customized a Chevy van into NavLab 1, a self-driving automobile with a maximum speed of 2 miles per hour (Navlab Website, 2003). By 1995, they had improved their technology to the point where they could take a Pontiac Trans Sport (NavLab 5) and drive it from San Diego to Pittsburgh with computerized steering. The car drove a total of 2,849 miles and the intelligence equipment controlled the steering wheel 98.2% of the way. The car used cameras to detect lane markers and other aspects of the open road to designate the appropriate driving area (“No Hands Across America”, 1995, p.3). This sensory equipment has evolved to include lasers and other such detectors. NavLab 11, a 2000 Jeep Wrangler Sport, is their latest project and is equipped with GPS, gyroscopes and magneto-meters, proximity laser scanners, omni-directional cameras, a laser line striper, and so forth (NavLab Website, 2003). Also, in 1997, the National Automated Highway System Consortium installed magnets along a 7.6 mile stretch of Interstate 15 and equipped eight Buick LeSabres with $200,000 worth of equipment. Data was transferred between the cars at a rate of 50 times per second by 166-megahertz Intel Pentium processors as the cars ran back and forth along the strip. The cars drove a total 8,000 miles and transported 4,000 passengers without any problems (Ashley, 1998, p.6).

As of now, liability prevents this technology from being implemented. It will be a while before scientists are sure enough of their technology to trust it with the lives of our drivers and passengers. Also, computers are constantly getting faster, technological devices are getting smaller and more efficient, and most importantly, all this equipment is getting less expensive. Artificially intelligent vehicles will benefit aspects of driving ranging far beyond traffic congestion and the potential problems of these systems can be worked out through further experimentation and tweaking. Regardless of the current status, smart cars and automated highways are imminent and inevitable, and will have a great impact on the transportation world.


Ashley, Steven. (1998, May). Smart cars and automated highways [Electronic Version]. Mechanical Engineering. Retrieved January 25th, 2003, from http://www.memagazine.org/backissues/may98/features/smarter/smarter.html

Carnegie Melon University Official Press Release. (1995). “No Hands Across America Official Press Release”. Retrieved January 25th, 2003, from http://www-2.cs.cmu.edu/afs/cs/user/tjochem/www/nhaa/official_press_release.html

Carnegie Melon Navigation Laboratory. (2002, September). Safe Robot Driving. Retrieved January 25th, 2003, from


About the Author

Enter content here

Enter content here