A new study at Cornell University demonstrated that robots can be programmed by causally talking with them. Till date, they needed specific instructions to conduct a particular task.
The study was conducted at Robot Learning Lab of Cornell University under Ashutosh Saxena, Assistant Professor of Computer Science. At present, Saxena is involved in teaching the robots to understand instructions in natural language from different speakers, account for missing information and adapt to the environment at hand.
The robot has built-in programming language with certain commands that can translate the human sentences for the robot. For example command find (pan), grasp (pan), carry (pan, water tap), fill (pan, water), carry (pan, stove) and so on and these instructions can be translated for the robot by the software and will be Fill a pan with water, put it on the stove, and heat the water.
Saxena’s robot is equipped with a 3-D camera and can scan its environment and identify the objects in it by using computer vision software and was previously developed in Saxena lab and the robot has been trained to associate objects with their capabilities.
Saxena along with his graduate students namely Dipendra K Misra and Jaeyong Sung will describe their methods from 12 to 16 July 2014 at the Robotics: Science and Systems conference at the University of California, Berkeley.
If you have any Question/Point on the above information, please ask/discuss it in the Current Affairs Group.
Who: Ashutosh Saxena
Where: Cornell University