MIT researchers created a system that helps robots complete intricate tasks, named Unvertian Specifications, or PUnS. The system is able to teach robots to even set a dinner table. The PUnS system operates by instructing the robot to have ‘beliefs’ using a language. From there the robot can figure out ways to complete complex tasks.
In the future, the MIT researchers will work towards using the system to have robots learn by watching and also have the ability to respond to feedback. This would mean that humans would be able to correct the robot when it makes a mistake. Overall the new system could have large implications, as it is a step toward robotic systems that can operate as assistants.
Image Credit: Christine Daniloff, MIT
What Makes This Trend Stand Out
- Robotic Task Teaching
- Teaching robots to complete intricate tasks using language and having the ability to learn from human feedback.
- Belief-based Robotics
- Creating systems that uses beliefs to help robots figure out ways to complete complex tasks.
- Assisting Robotic Systems
- Developing robots that can operate as assistants alongside humans.
Sectors Adopting This
- Hospitality Industry
- Implementing robotic task teaching to the hospitality industry where robots can set up tables and perform other tasks.
- Manufacturing Industry
- Belief-based robotics can be used in manufacturing industries to be more efficient and precise in completing intricate tasks.
- Healthcare Industry
- Developing assisting robotic systems for healthcare professionals to perform basic tasks such as lifting, transferring and repositioning patients.
