Brown University’s robotics department, the Brown Robotics Group, recently demonstrated a first generation prototype robot capable of following receiving, interpreting and responding accurately to hand signals, body motion and verbal commands.
The quest to create such a robot allegedly started when the scientists conceived the notion to create a robot that can do whatever a person wanted.
The project is funded by the U.S. Defense Advanced Research Projects Agency Information Processing Techniques Office and by the U.S. Office of Naval Research.
A few target commands include, “Please get me a glass of water,” “Please keep out the trash” and “Dunk these in the washing machine.”
The market for such personal assistants or buddies such as this robot is anticipated to be great once the final product is ready.
In a vivid demo of the robot, one can see how it waits for a person to stand in front of them with their back to the robot at a distance of about 2 meters. It is programmed with voice recognition of commands such as "follow," "halt," "wait" and "door breach."
When the assigned person says, “follow” the robot is able to recognize only this person as its master or buddy. It understands that it must start following when the left hand is raised horizontally outwards and the elbow is bent upwards with a raised fist. A repetition of this command informs it to stop and remain in one place.
Also, it is able to follow the person in narrow corridors, around bends and at a constant distance. When the person stops, even without saying halt or executing a command, the robot also stops, trying to keep the distance constant as much as possible. It is also fairly sensitive about change in speed and adapts its own speed to match.
The team used a mechanized system from iRobot and CSES Swiss Ranger depth camera for visual and distance sensing.
“We have created a novel system where the robot will follow you at a precise distance, where you don’t need to wear special clothing, you don’t need to be in a special environment, and you don’t need to look backward to track it,” said Chad Jenkins, assistant professor of computer science at Brown University and the team’s leader.
Vivek Naik is a contributing editor for TMCnet. To read more of Vivek's articles, please visit his columnist page.Edited by
Stefania Viscusi