Researchers at MIT, have brought to us a new object recognition system that works on SLAM (simultaneous localization and mapping) technology. This robot sees objects are a three-dimensional plane and not just in one view, as is the case in conventional image recognition systems.

“Ideally we want robots to be cleaning our dishes at some point in the future. We want recognition systems where it does in fact see the objects that the robot should care about and manipulate them,” says Sudeep Pillai, lead author of the study.

SLAM-aware robots hold advantages over SLAM-unaware robots as they can easily identify an object from a clutter of objects. It has been seen that the SLAM-aware system identified objects with only 86.1% accuracy, as compared to SLAM-unaware identifying it with a 92.8% accuracy. But, the former wins over latter in terms of startup time. SLAM-aware system takes a mere 1.6 seconds to start, whereas, classical systems might take up to four seconds.

This new system works sans infrared light, which the old recognition systems need to function. Often, the conventional systems could not be used outdoors. Now, both indoors and outdoors can be are included in its area of activity.

Presently, the team at MIT is working to solve the “loop closure” issue that exists in classical robots, in which the system fails to recognize a previously visited place, or a previously performed activity.

[adinserter block=”7″]

Author:Technology Blog