Descripción del proyecto
Diseño de robots tan resilientes como los animales
Los robots son el futuro, pero, a pesar de decenios de investigación, siguen teniendo un gran defecto: son máquinas frágiles que pueden averiarse fácilmente en condiciones difíciles. Sin embargo, hay una forma de crear robots de bajo coste que puedan recuperarse de forma autónoma (e inmediata) de daños imprevistos. El proyecto ResiBots, financiado por el Consejo Europeo de Investigación, revolucionará nuestro planteamiento de la tolerancia a fallos y producirá robots tan resistentes y adaptables como los animales. En concreto, el equipo del proyecto empleará algoritmos de aprendizaje por ensayo y error que permitan que los robots descubran rápidamente comportamientos compensatorios sin necesidad de sensores caros o planes de contingencia predefinidos. El objetivo general es aumentar sustancialmente la vida útil de los robots sin incrementar su coste. En el proyecto se allanará el camino a nuevas vías de investigación para máquinas adaptativas.
Objetivo
Despite over 50 years of research in robotics, most existing robots are far from being as resilient as the simplest animals: they are fragile machines that easily stop functioning in difficult conditions. The goal of this proposal is to radically change this situation by providing the algorithmic foundations for low-cost robots that can autonomously recover from unforeseen damages in a few minutes. The current approach to fault tolerance is inherited from safety-critical systems (e.g. spaceships or nuclear plants). It is inappropriate for low-cost autonomous robots because it relies on diagnostic procedures, which require expensive proprioceptive sensors, and contingency plans, which cannot cover all the possible situations that an autonomous robot can encounter. It is here contended that trial-and-error learning algorithms provide an alternate approach that does not require diagnostic, nor pre-defined contingency plans. In this project, we will develop and study a novel family of such learning algorithms that make it possible for autonomous robots to quickly discover compensatory behaviors. We will thus shed a new light on one of the most fundamental questions of robotics: how can a robot be as adaptive as an animal? The techniques developed in this project will substantially increase the lifespan of robots without increasing their cost and open new research avenues for adaptive machines.
Ámbito científico
Not validated
Not validated
- engineering and technologyelectrical engineering, electronic engineering, information engineeringelectronic engineeringsensorsoptical sensors
- natural sciencescomputer and information sciencesartificial intelligencemachine learningreinforcement learning
- engineering and technologyelectrical engineering, electronic engineering, information engineeringelectronic engineeringroboticsautonomous robots
- natural sciencesearth and related environmental sciencesphysical geographynatural disasters
Programa(s)
Régimen de financiación
ERC-STG - Starting GrantInstitución de acogida
78153 Le Chesnay Cedex
Francia