By Kalwinder Kaur
Sterling Anderson, a PhD student in MIT’s Department of Mechanical Engineering and Karl Iagnemma, MIT Robotic Mobility Group’s principal research scientist have developed a new semiautonomous safety system.
Based on an onboard camera and laser rangefinder, the system detects the obstacles in the environment. The team developed an algorithm for data analysis and detection of safe zones. Using this system, the driver can efficiently maneuver the vehicle. The system enables the driver to have control over the wheels when the vehicle is on the point of exiting a safe zone.
This system has been depicted as an “intelligent co-pilot” by Anderson, who has been testing the system in Michigan since September 2011. This system monitors the driver’s performance in addition to facilitating out of sight adjustments. The vehicle is therefore made secure within a safe environment, preventing collision with obstacles.
The details of the safety system were presented recently at the Intelligent Vehicles Symposium in Spain. Lately, robotics research has highlighted systems that can be leveraged by either humans or robots. Most of those systems function with respect to preprogrammed paths.
Anderson describes the technology underlying self-parking cars. Here, parallel park can be attained by flipping a switch followed by removing the driver’s hands off the wheel. After a preplanned path owing to the distance between adjacent cars, the car can make automatic parking.
This human perspective was incorporated into the robotic system by Anderson and Iagnemma. The team devised an approach for detecting safe zones, or “homotopies”. In contrast to mapping individual paths along a roadway, a vehicle’s environment can be classified into triangles, where the triangle edges may represent a lane’s boundary or an obstacle.
An algorithm that constrains obstacle-abutting edges has been formulated by the researchers, through which drivers can safely navigate across any triangle edge excluding those constrained ones. When the driver is out of control, the system takes control of the vehicle, steering the car back to a safer environment.