DCSIMG

MIT researchers develop semiautonomous system to improve vehicle safety

MIT researchers develop semiautonomous system to improve vehicle safety

Like what you see? Tell a colleague
RSS Feed

News & Events - Engineering News

People have long been calling for a solution to help limit the prevalence and magnitude of car accidents, and they may eventually see significant progress in this regard due to research done by scientists from the Massachusetts Institute of Technology.

According to the MIT News Office, a new semiautonomous safety system is being developed by Sterling Anderson, a PhD student in the school's Department of Mechanical Engineering, and Karl Iagnemma, a research associate with MIT's Robotic Mobility Group.

The two researchers are relying on the school's wealth of engineering resources to develop this technology, as they are looking to create a system that would correct human error in regards to driving an automobile.

The system uses a laser rangefinder and an onboard camera to identify hazards that exist in the vehicle's environment. According to the news outlet, the research team devised and developed an algorithm to analyze data and identify areas that would be safe for passage. This would help to expose items that could interfere with driving, such as other cars or barrels in the road.

MIT News Office reported that the system allows a driver to control the vehicle, but it takes the wheel when the driver is about to exit a safe zone. Anderson has been testing the device on a course in Saline, Michigan, relying on an obstacle course and a series of tests to work out the kinks in the technology.

Anderson describes the system as an "intelligent co-pilot" that monitors a driver's performance. It then uses data taken from this monitoring to make adjustments behind the scenes, preventing the car from colliding with obstacles and helping to direct it to a safe environment.

"The real innovation is enabling the car to share [control] with you," Anderson told the news outlet. "If you want to drive, it’ll just…make sure you don’t hit anything."

The two researchers presented the details of their system at the Intelligent Vehicles Symposium in Spain, where they highlighted the "minimally-invasive" control method associated with the result of their engineering research.

Anderson and Iagnemma compared part of their system to the technology behind self-parking cars, but the researchers noted that when it comes to driving, such a simple system with only one or two paths would not work.

"The problem is, humans don’t think that way," Anderson told MIT News Office. "When you and I drive, [we don’t] choose just one path and obsessively follow it. Typically you and I see a lane or a parking lot, and we say, 'Here is the field of safe travel, here’s the entire region of the roadway I can use, and I’m not going to worry about remaining on a specific line, as long as I’m safely on the roadway and I avoid collisions.'"

PC Magazine reported that the shared-control system represents significant progress in trying to limit crashes, as the more than 1,200 tests runs for the device have produced results that indicate a finalized system could be in the near future for MIT researchers.

According to the news outlet, people may initially worry that the system takes too much control away from the individual, but Anderson noted that if the driver stays within the boundaries of the algorithmically determined "corridor of safety," the human will be responsible for driving.

Only dangerous maneuvers, according to Anderson, will engage the co-pilot's control system, as the car will be steered clear of the obstacles and into the "safe zone."



Share
Read more like this