TECH/CAR SPECIAL...MIT' s co pilot tech averts car crashes

MIT' s co pilot tech averts car crashes 

Unlike fully automatic systems, this vehicle safety technology works behind the scenes and assumes control of a car only when it senses a risk of collision

    Engineers at the Massachusetts Institute of Technology (MIT) have developed a semi-autonomous vehicle safety system capable of taking control of a car to avert an imminent crash.While Google’s self-driving car is programmed to assume control completely, MIT’s “intelligent co-pilot” system doesn't replace the human driver.
    It takes over a car when it detects a risk of collision, manoeuvring it around danger and then handing the control back to the driver. This, MIT researchers say, is the key benefit of the technology.
    "The real innovation is enabling the car to share [control] with you. If you want to drive, it'll just … make sure you don't hit anything," said Sterling Anderson, a PhD student in MIT's Department of Mechanical Engineering.
    He has developed the system with Karl Iagnemma, a principal research scientist in MIT's Robotic Mobility Group. They have been testing the system in Michigan, US, since September last year.
COURSE CORRECTION
The co-pilot system uses an onboard camera and laser range-finder to map the area around a car and identify potential obstacles or hazards in its path. The team devised an algorithm to analyse the data and identify safe zones - avoiding, for in stance, barrels in a field or other cars on a roadway, MIT News reported.
    According to Anderson, that the system monitors a driver's performance and makes behind-the-scenes adjustments to keep the vehicle in a proper lane.
    “When you are driving safely, the system runs in the background. But if you make a mistake, one serious enough to cause a collision or loss of control, it intervenes,” he explained.
    The MIT pair have conducted more than 1,200 test runs of the technology using a modified jeep-like vehicle in an open-field obstacle track. Over 30 different drivers were instructed to steer the vehicle into various obstacles.
KEEPING THE HUMAN TOUCH
Anderson told MIT News that their research marks a departure from traditional auto-pilot technology, where the human operator has to turn on the system and hand over full control to the robotics, which then follows a preplanned course.
    “The problem is that humans don’t think that way. When you and I drive, [we don’t] choose just one path and obsessively follow it,” he said. “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’.”
    Anderson and Iagnemma integrated this human perspective into their robotic system. They came up with an approach to identify safe zones, or “homotopies”, rather than spe cific paths of travel, MIT News reported Instead of mapping out individual paths along a road, they divided a vehicle’s environment into triangles, with certain triangle edges rep resenting an obstacle or a lane’s boundary.
    The researchers are now exploring ways to tailor the system to different levels of driving experience. They are also hoping to pare down the system to identify obstacles using single mobile.
    “You could stick your cellphone on the dashboard, and it would use the camera accelerometers and gyro to provide the feed back needed by the system,” Anderson said “I think we’ll find better ways of doing it that will be simpler, cheaper and allow more users access to the technology.” The team believes the co-pilot technology could quite useful for people who are learning to drive. AGENCIES

MM120717