Descripción del proyecto
Una metodología mejorada para el aprendizaje de los robots
Debido a la gran utilidad de la robótica para varios sectores, así como para su automatización, se están dedicando muchos esfuerzos al desarrollo y la programación de robots. En la actualidad, los métodos con los que se programan y enseñan a los robots son el aprendizaje por imitación (el robot aprende imitando al profesor) y el aprendizaje por refuerzo (el robot aprende por prueba y error). Por desgracia, ambos consumen mucho tiempo y son caros, incluso para tareas sencillas. El objetivo del proyecto TERI, financiado con fondos europeos, es revolucionar el aprendizaje de los robots al introducir una nueva metodología, la interacción intermitente entre el robot y el profesor, un método que los humanos usamos con eficacia. Esto permitirá al profesor guiar y corregir mejor al robot durante su proceso de aprendizaje, con lo que se conseguirán unos resultados mejores y más rápidos y baratos.
Objetivo
Programming and re-programming robots is extremely time-consuming and expensive, which presents a major bottleneck for new industrial, agricultural, care, and household robot applications. My goal is to realize a scientific breakthrough in enabling robots to learn how to perform manipulation tasks from few human demonstrations, based on novel interactive machine learning techniques.
Current robot learning approaches focus either on imitation learning (mimicking the teacher’s movement) or on reinforcement learning (self-improvement by trial and error). Learning even moderately complex tasks in this way still requires infeasibly many iterations or task-specific prior knowledge that needs to be programmed in the robot. To render robot learning fast, effective, and efficient, I propose to incorporate intermittent robot-teacher interaction, which so far has been largely ignored in robot learning although it is a prominent feature in human learning. This project will deliver a completely new and better approach: robot learning will no longer rely on initial demonstrations only, but it will effectively use additional user feedback to continuously optimize the task performance. It will enable the user to directly perceive and correct undesirable behavior and to quickly guide the robot toward the target behavior. In my previous research I have made ground-breaking contributions to the existing learning paradigms and I am therefore ideally prepared to tackle the three-fold challenge of this project: developing theoretically sound techniques which are at the same time intuitive for the user and efficient for real-world applications.
The novel framework will be validated with generic real-world robotic force-interaction tasks related to handling and (dis)assembly. The potential of the newly developed teaching framework will be demonstrated with challenging bi-manual tasks and a final study evaluating how well novice human operators can teach novel tasks to a robot.
Ámbito científico
Programa(s)
Régimen de financiación
ERC-STG - Starting GrantInstitución de acogida
2628 CN Delft
Países Bajos