Better driving decisions = improved fuel economy

Dr. Neil Canter, Contributing Editor | TLT Tech Beat November 2011

A smartphone system alerts drivers with information to better move through intersections.

KEY CONCEPTS
• One parameter that has not been addressed in striving to improve fuel economy is the driver.
• A new Green Light Optimal Speed Advisory System (GLOSA) has been developed to help drivers better move through intersections under urban traffic conditions.
• The GLOSA system relies on a network of mobile phones in each car to capture images of the road ahead and detect the schedule of traffic signals. Use of this system led to a 20% improvement in fuel economy.

Fuel economy improvement remains a major market driver for the automotive industry and, as a consequence, for the lubricants industry. The U.S. government has proposed that the corporate average fuel Economy (CAFE) be raised to 54.5 miles per gallon by 2025. Heavy-duty trucks also are involved in this process, as the U.S. has proposed that fuel economy for this segment be raised by 20% by the 2018 model year.

With these motivations, research is underway to expand ways to improve vehicular fuel efficiency. In a previous TLT article, a new thermoelectric material was discussed that can convert automotive exhaust heat to electricity (1). This is an important finding because approximately 40% of the fuel energy produced is lost as heat. The researchers who developed the thermoelectric material hope that fuel economy can be improved by at least 15% to 20%.

One additional aspect that needs to be taken into consideration is the driver. If the driver is provided with more useful information while driving in situations such as urban traffic, can this lead to improved fuel economy? Many of us are dependent on the automobile to get to work. We have all at some point been in traffic which has led to a situation where fuel consumption and emissions increase. Any means that can be developed to improve traffic flow also should lead to improved fuel economy.

Emmanouil Koukoumidis, doctoral student at Princeton University in Princeton, N.J., and visiting researcher at MIT in Cambridge, Mass., says, “Traffic signals are very important in developed countries to allow competing lines of cars to safely pass through intersections. But the high volume of cars can lead to stop-and-go traffic that can increase fuel consumption by 17% and carbon dioxide emissions by 15%, plus lead to high levels of driver anxiety.”

Technologies have been developed that can help assist drivers in moving through traffic signals. They are collectively known as Green Light Optimal Speed Advisory (GLOSA) systems. Koukoumidis says, “GLOSA systems help drivers determine the optimal speed at which they should go as they approach and pass through a specific intersection.”

Only a few GLOSA systems have been developed worldwide. Many of these systems are very limited in that they are dependent upon roadside message signs that are wired to traffic signals. Koukoumidis says, “These systems are very costly to implement and require maintenance. Their deployment is also often not practical, particularly in busy cities.”

Vehicular-traffic-signal countdown timers constitute an alternative advisory system that has been deployed in a few cities around the world. Koukoumidis adds, “However, drivers can only see these timers when they are within 50 meters of a traffic signal, which is often too late for them to adjust their speed.”

The GLOSA concept is fundamentally a good one that can help improve fuel economy. A simpler system though is needed that can work when the driver is further away from the intersection and is not costly to use. Such a system has now been developed.

SMART MOBILE PHONES
Koukoumidis, in collaboration with Li-Shiuan Peh, associate professor in the department of electrical engineering and computer science at MIT, and Margaret Martonosi, professor in the department of computer science at Princeton University, N.J., have developed a new approach to provide drivers with the needed information to better move through intersections. Koukoumidis says, “We have developed a GLOSA system that is based on the use of mobile phones placed on the windshield by drivers. One of the keys is that each of the mobile phones can communicate with each other, enabling a driver far away from the intersection to get information about the traffic signal ahead so that the speed of the automobile can be adjusted to pass through the intersection.”

This GLOSA system is known as SignalGuru and relies on each of the mobile phones to capture images of the road ahead and detect the schedule of the traffic signals. Figure 3 shows an image of the smartphone on the windshield of an automobile. One major advantage of this system is it does not require any additional infrastructure.


Figure 3. A cell phone placed on the car’s windshield can be part of a network to help drivers better move through intersections under congested traffic conditions and potentially increase their fuel economy. (Courtesy of the Massachusetts Institute of Technology)

Koukoumidis indicates that the mobile phone system gathers three pieces of information needed to calculate the optimal driving speed. He says, “Our system can predict the amount of time until the traffic signal at the next intersection turns green, the location of the intersection and the vehicle’s current location. We should also augment the system to tell how many automobiles are ahead of our vehicle.”

Weather can be one of the limitations of this smartphone system. Koukoumidis says, “We initially evaluated our system during sunny days. By adjusting the exposure time, we can get the system to work at night and under adverse weather conditions. The one environment which we have not yet solved is foggy conditions.”

Two studies were conducted in Cambridge and Singapore to evaluate the effectiveness of the smartphone system. Koukoumidis says, “We conducted the study in Cambridge using five vehicles that followed a specific urban route through three traffic lights over a three-hour period from the early afternoon until the beginning of rush hour. The smartphone system was able to correctly predict the timing for a green traffic signal with an accuracy of 98.2%.”

In Singapore, eight taxis were used to travel through two routes in a downtown area of the country. The system worked well to predict timing for the traffic-adaptive traffic lights with an accuracy of 96.3%.

Fuel efficiency was calculated using a 2001 2.4-liter Chrysler PT Cruiser in Cambridge. Koukoumidis says, “When this car did not use our system, it stopped at 1.7 out of every three lights. In contrast, the car did not stop at any of the three traffic lights when the smartphone system was used. We measured that this led to a 20% reduction in fuel usage.”

Koukoumidis believes the smartphone camera-based technology can be used in other driver-related applications such as determining where gasoline stations are located in a specific neighborhood and finding the best price for gasoline. A second application is to determine the availability of parking spaces in a specific area. 

Further information can be obtained in a recent article (2) or by contacting Koukoumidis at koukou@csail.mit.edu. 

REFERENCES
1. Canter, N. (2011), “Electricity from Automotive Exhaust Heat,” TLT, 67 (3), pp. 10-11.
2. Koukoumidis, E., Peh, L. and Martonosi, M. (2011), “Signal-Guru: Leveraging Mobile Phones for Collaborative Traffic Signal Schedule Advisory,” Association for Computing Machinery’s MobiSys 2011 Conference, Washington, D.C., July 2011.


Neil Canter heads his own consulting company, Chemical Solutions, in Willow Grove, Pa. Ideas for Tech Beat items can be sent to him at neilcanter@comcast.net.