On-line learning and planning in a pick-and-place task demonstrated through body manipulation

Abstract

Abstract-When a robot is brought into a new environment, it has a very limited knowledge of what surrounds it and what it can do. One way to build up that knowledge is through exploration but it is a slow process. Programming by demonstration is an efficient way to learn new things from interaction. A robot can imitate gestures it was shown through passive manipulation. Depending on the representation of the task, the robot may also be able to plan its actions and even adapt its representation when further interactions change its knowledge about the task to be done. In this paper we present a bio-inspired neural network used in a robot to learn arm gestures demonstrated through passive manipulation. It also allows the robot to plan arm movements according to activated goals. The model is applied to learning a pick-and-place task. The robot learns how to pick up objects at a specific location and drop them in two different boxes depending on their color. As our system is continuously learning, the behavior of the robot can always be adapted by the human interacting with it. This ability is demonstrated by teaching the robot to switch the goals for both types of objects.

Speed Adaptation in Learning from Demonstration through Latent Space Formulation

“…LfD has become an important topic in robotics research with notable applications in relevant sectors, such as motion behaviors, human-robot interaction, artificial intelligence, and goal-based learning. [11][12][13][14] In the current section, we briefly review and highlight representative works from contemporary literature. Accordingly, Gupta et al 15 proposed an algorithm for policy learning and generalization that allows complex dexterous manipulators to learn from multiple human demonstrations.…”

A developmental approach of imitation to study the emergence of mirror neurons in a sensory-motor controller

“…These goals can be associated with sequences of proprioceptive states acquired through passive manipulation of the arm [4]. When the robotic arm moves, its proprioception (joint angles, gripper infrared and force sensor values) is categorized into distinct states.…”

Learning of Social Signatures Through Imitation Game Between a Robot and a Human Partner

“…Second, in the context of learning, imitation reduces the search space of the learner and facilitates the participant-infant (or teacher-learner) interaction. Thus, in the field of robotics, imitation is often considered to be a powerful behavior that corresponds to the ability to learn by observation [14], [24], [25], [49], [76]. For example, a robot can combine known actions to reproduce a behavior that a participant performs.…”