An algorithm developed at Brown University will improve robots' ability to ask clarifying questions and more effectively retrieve objects, an important task for future robotic assistants.
“Fetching objects is an important task that we want collaborative robots to be able to do,” said Brown University computer science professor Stefanie Tellex said. “But it’s easy for the robot to make errors, either by misunderstanding what we want, or by being in situations where commands are ambiguous."
Tellex’s lab had previously developed an algorithm that enables robots to receive speech commands as well as information from human gestures, such as pointing.
The new algorithm enables the robot to quantify its certainty in knowing what a user wants retrieved. When certainty is high, the robot will simply hand over the object as requested. When less sure, the robot will make its best guess about what the person wants, then request confirmation by hovering its gripper over the object and asking, “This one?”
The development also allows a robot to make inferences. If a robot, for example, chooses the wrong wrench out of two possible choices, the technology will deduce that a user requires the other remaining one.
In future work, Tellex and her team would like to combine the algorithm with more robust speech recognition systems, which could further increase the system’s accuracy and speed.
Also: Read more Robotics tech briefs.