Improving a robot’s self-awareness by giving it proprioception

Two roboticists from the Munich Institute for Robotics and Artificial Intelligence (MIRMI), Technical University of Munich, Germany, have discovered that it is possible to give robots a certain degree of proprioception using machine learning techniques. In their study reported in the journal Scientific roboticsFernando Díaz Ledezma and Sami Haddadin have developed a new machine learning approach to allow a robot to learn the specificities of its body.

Giving robots the ability to move in the real world involves equipping them with technology like cameras and pressure sensors. Data from these devices is then processed and used to direct the legs and/or feet to perform appropriate actions. This is very different from the way animals, including humans, accomplish their work.

In animals, the brain is aware of the state of its body: it knows where the hands and legs are, how they work, and how they can be used to move or interact with the environment. Such knowledge is known as proprioception. In this new effort, researchers conferred similar capabilities on robots using machine learning techniques.

The idea behind their system is to add sensors to the body that give information about each part of the body. For example, the sensors know where the knee is, which direction it is bending, and the degree of flexion at any given moment. Researchers have found that the overlap between sensors and the data they send to a central processor allows for greater overall awareness of the body’s state.

They also discovered that a robot could, to some extent, learn to understand its body without pre-learned data. Instead, they simply cause what they describe as “motor chatter” where all the servo motors used to power a robot are fired at random at the same time. This allows the robot to begin building a base of information that can be used to learn how its parts work.

The researchers then tested their approach on several types of robots, including a six-legged spider robot, a humanoid, and one arm. They found that their approach allowed all types of robots tested to develop some sense of their own bodies, their parts, and how they worked together.

© 2023 Science X Network