Title :
Learning optimal robotic tasks
Author :
Lima, Pedro U. ; Saridis, George N.
Author_Institution :
Inst. Superior Tecnico, Lisbon, Portugal
fDate :
4/1/1996 12:00:00 AM
Abstract :
A reinforcement learning system for machine-intelligence robots introduces a performance measure that balances an algorithm´s reliability and computational cost. The system uses this measure to learn the best among a set of alternative tasks capable of executing a command communicated to the intelligent machine
Keywords :
intelligent control; learning (artificial intelligence); optimal control; algorithm reliability; command execution; computational cost; intelligent machine; machine-intelligence robots; optimal robotic task learning; performance measure; reinforcement learning system; Computational intelligence; Design methodology; Force control; Intelligent control; Intelligent robots; Manipulators; Robot kinematics; Robot sensing systems; Service robots; Stress measurement;
Journal_Title :
IEEE Expert
