Posted in | News | Consumer Robotics

KAIST Researchers Develop Robot Archetype Operated by Action-Based Commands

Scientists at the Korea Advanced Institute of Science and Technology (KAIST) have developed a robot, which can precisely understand and react to human non-verbal commands.

As part of demonstration, a researcher pushed a toy car as well as tossed a ball towards the archetype. The robot got little confused and displayed a question mark on a screen linked to it. The researcher did not reply orally, instead he just made a throwing gesture using his hand. The archetype immediately reacted by throwing the ball to him, neglecting the car.

The core system of the robot is developed to replicate the movements of a human being. Human brain includes mirror neurons which make humans to mimic the same action performed by others.

The archetype observes an action, split it into easy verbal explanation and then stores it in a simulated database. Soon after, when the same action is repeated by a human, the robot will refer to its memory bank and then expresses a reactive movement that best suites the observed action. If its database suggests two probable intentions, then it will ask for a second action from the human. Every action details are stored by the robot to bring down the reaction time if it faces the same action in the future.

Ji-Hyeong Han has stated that robots can understand human thoughts by interpreting verbal commands. However, humans need to give continuous precise instructions to the robot. Non-verbal instructions will enable to bring down error numbers that occur while communicating the message to the robot.

The developers are looking forward to conduct another round of assessment after including visual sensors in the robot for identifying and recalling additional human actions.

Tell Us What You Think

Do you have a review, update or anything you would like to add to this news story?

Leave your feedback
Your comment type
Submit

While we only use edited and approved content for Azthena answers, it may on occasions provide incorrect responses. Please confirm any data provided with the related suppliers or authors. We do not provide medical advice, if you search for medical information you must always consult a medical professional before acting on any information provided.

Your questions, but not your email details will be shared with OpenAI and retained for 30 days in accordance with their privacy principles.

Please do not ask questions that use sensitive or confidential information.

Read the full Terms & Conditions.