MIT’s Robotic Co-Drivers Aim to Prevent Car Accidents
In recent years, inventors have come up with many features to decrease the number of traffic accidents. Emily Finn at MITnews elaborates in her excellent article “‘Smart cars’ that are actually, well, smart”:
Engineers have developed myriad safety systems aimed at preventing collisions: automated cruise control, a radar- or laser-based sensor system that slows a car when approaching another vehicle; blind-spot warning systems, which use lights or beeps to alert the driver to the presence of a vehicle he or she can’t see; and traction control and stability assist, which automatically apply the brakes if they detect skidding or a loss of steering control.
Still, more progress must be made to achieve the long-term goal of ‘intelligent transportation’: cars that can ‘see’ and communicate with other vehicles on the road, making them able to prevent crashes virtually 100 percent of the time.
But despite all of the above-mentioned technical systems, humans will continue to drive vehicles as long as older vehicles are still in use. To get around the human factor as a cause of auto accidents, mechanical engineers at Massachusetts Institute of Technology (MIT) have been developing an intelligent transportation system based on a new algorithm that takes into account models of human driving behavior in order to warn drivers of possible collisions and even take control of the vehicle to prevent a crash.
MIT scientists Rajeev Verma, who was a visiting Ph.D. student at MIT this academic year, Domitilla Del Vecchio, assistant professor of mechanical engineering, and W. M. Keck, Career Development Assistant Professor in Biomedical Engineering, will publish a paper about their work in the journal IEEE Robotics and Automation Magazine.
Del Vecchio said it’s a challenge to design a system that is safe without being hypersensitive and always reacting to the worst-case scenarios. That type of system, she told MITnews, would give “you warnings even when you don’t feel them as necessary. Then you would say, ‘Oh, this warning system doesn’t work,’ and you would neglect it all the time.”
Following the lead of many other researchers, Del Vecchio and Verma developed their algorithm by reasoning that drivers’ actions consist mainly of accelerating and braking, and that there is a finite set of places where a vehicle could be in the future, based on predictive models of human behavior, like where and when drivers slow down or speed up at intersections. The program they created can compute, for any two vehicles on the road nearing an intersection, an area in which two vehicles might collide.
As Alan Kotok, editor and publisher of Science Business, an online news service about the process connecting scientific discovery to the marketplace, explains in his article, “Engineers Developing Human Driving Model to Reduce Crashes”:
The car with a system using the algorithm then calculates a game-theory based decision; using information from its on-board sensors as well as roadside and traffic-light sensors, the system tries to predict what the other car will do, and reacts accordingly to prevent a crash.
And, as Finn writes for MITnews:
When both cars are ITS-equipped, the ‘game’ becomes a cooperative one, with both cars communicating their positions and working together to avoid a collision.
Now that the researchers have already begun to test their system in full-size passenger vehicles with human drivers, their plans include considering human reaction time data to determine when the system can give a passive warning to drivers and when it should take control of the vehicle, building in sensors for weather and road conditions, and accounting for manufacturing details specific to different makes and models of vehicles.