For decades, scientists worked to create control algorithms for autonomous helicopters — robotic helicopters that pilot themselves. Dozens of research teams have competed in a series of autonomous-helicopter challenges posed by the Association for Unmanned Vehicle Systems International (AUVSI); progress has been so rapid that the last two challenges have involved indoor navigation without the use of GPS.
Now MIT’s Robust Robotics Group — which fielded the team that won the last AUVSI contest — have created a robotic airplane that can fly at high speeds, through obstacles, using only on-board sensors. And it’s doing all that without benefit of GPS. They created an “autonomous plane navigation in confined spaces”.
The MIT researchers completed a serious of flight tests around pillars in the parking garage under their labs. The plane worked like a charm — gliding and threading, almost blindly, around the structures. Of course it’s not blind, the plane, using a laser rangefinder, calculates 15 different variables while in flight — keeping it from crashing into things.
The plane’s navigation system is powered by an Intel Atom processor, the same consumer grade processor that may be inside your smartphone or tablet right now.
The MIT group wants to now set its sights on building algorithms that will allow the plane to map out its flight environment on the go.