Learning to Coordinate Controllers - Reinforcement Learning on a Control Basis

Huber, Manfred; Grupen, Roderic A.

Learning to Coordinate Controllers - Reinforcement Learning on a Control Basis

Download BibTeX Cite

Proceedings of the Fifteenth International Joint Conference on Artificial Intelligence. Robotics. Pages 1366-1370.

Abstract

Autonomous robot systems operating in an uncertain environment have to be reactive and adaptive in order to cope with changing environment conditions and task requirements. To achieve this, the hybrid control architecture presented in this paper uses reinforcement learning on top of a Discrete Event Dynamic System (DEDS) framework to learn to supervise a set of basis controllers in order to achieve a given task. The use of an abstract system model in the automatically derived supervisor reduces the complexity of the learning problem. In addition, safety constraints may be imposed a priori, such that the system learns on-line in a single trial without the need for an outside teacher. To demonstrate the applicability of the approach, the architecture is used to learn a turning gait on a four legged robot platform.

Manfred Huber, and Roderic A. Grupen. 1997. Learning to Coordinate Controllers - Reinforcement Learning on a Control Basis. In Proceedings of the Fifteenth International Joint Conference on Artificial Intelligence, IJCAI-97. IJCAI, 1366–1370. https://ijcai.org/proceedings/1997/197

@inproceedings{ijcai1997p197,
  title     = {Learning to Coordinate Controllers - Reinforcement Learning on a Control Basis},
  author    = {Huber, Manfred and Grupen, Roderic A.},
  booktitle = {Proceedings of the Fifteenth International Joint Conference on
               Artificial Intelligence, {IJCAI-97}},
  publisher = {International Joint Conferences on Artificial Intelligence},
  editor    = {Martha E. Pollack},
  pages     = {1366--1370},
  year      = {1997},
  note      = {Robotics},
  url       = {https://ijcai.org/proceedings/1997/197},
}

TY  - CPAPER
TI  - Learning to Coordinate Controllers - Reinforcement Learning on a Control Basis
AU  - Huber, Manfred
AU  - Grupen, Roderic A.
T2  - Proceedings of the Fifteenth International Joint Conference on Artificial Intelligence, IJCAI-97
PB  - International Joint Conferences on Artificial Intelligence
PY  - 1997
SP  - 1366
EP  - 1370
UR  - https://ijcai.org/proceedings/1997/197
ER  -