We wrote last week that while collision avoidance systems have had a beneficial effect on accident statistics in the US, collision avoidance systems appear to have had the opposote effect.

New technology being researched by MIT’s Department of Mechanical Engineering in the US, however, combines these two approaches for even safer driving.

The system works a little like an aircraft autopilot in reverse. Commercial aircraft auto-pilot systems take control of flight under normal conditions and only hand control back to the pilot when a situation becomes too complex to handle — the presumption being that a human knows how to best handle extreme conditions.

MIT’s system, on the other hand, assumes the driver wants to control the vehicle under normal driving conditions and will only step in when it looks like they’re about to lose control — to avoid a collision, in other words.

The system uses an onboard camera and a laser rangefinder to identify hazards in a vehicle’s immediate environment and plot a ‘safe zone’. The system then acts as an intelligent ‘co-pilot’ and only takes an interest when it thinks the drive is about to leave the safe zone.

[tl:movies size=450×253]

The system is being developed by Sterling Anderson, a PhD student in MIT’s Department of Mechanical Engineering, and Karl Iagnemma, a principal research scientist in MIT’s Robotic Mobility Group.

The difference with this system is that, unlike Google’s self-driving cars, it leaves the driver in full control of the vehicle. Nor does it attempt to detect road markings in order to determine when a vehicle is being driven safely.

Explaining this different approach, Anderson said: “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.’”

More than 1,200 trials of the system have so far resulted in few collisions and most were the result of glitches with the vehicle’s camera failing to identify an obstacle.

This system also has the added advantage of being much simpler to implement than similar self-driving technologies being developed by car manufacturers.

“You could stick your cellphone on the dashboard, and it would use [its onboard sensors] to provide the feedback 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.”

[MIT]