Washington, Sep 4 (IANS) The way infants spend their initial months trying to figure their way around could serve as a model for training robots, an Indian American researcher says.


A team led by Ashutosh Saxena at Cornell University Personal Robotics Lab, is training robots to handle objects as infants do to find their way around in their new environments.

A computer can be 'trained' to look up cups of all shapes and sizes and find what is common to all of them. A similar process can teach a robot to find a cup's handle and grasp it correctly.

Saxena, an Indian Institute of Technology-Kanpur alumnus, has found that placing objects is harder than picking them up.

A cup is placed upright on a table, but upside down in a dishwasher, so the robot must be trained to make those decisions, according to a Cornell statement.

"We just show the robot some examples and it learns to generalize the placing strategies and applies them to objects that were not seen before," Saxena, assistant professor of computer science, explained.

"It learns about stability and other criteria for good placing for plates and cups, and when it sees a new object -- a bowl -- it applies them," said Saxena.

After training, their robot placed most objects correctly 98 percent of the times when it had seen the objects and environments previously, and 95 percent of the times when working with new objects in a new environment.