Reinforcement{driven Adaptation of Control Relations
نویسندگان
چکیده
Hans Arno Jacobsen Joachim Weisbrod Berkeley Initiative in Soft Computing Institut f ur Programmstrukturen und Datenorganisation University of California Universitat Karlsruhe Berkeley, CA 94720-1776 USA Deutschland [email protected] [email protected] Abstract The conceptual framework of a hybrid control system architecture is brie y motivated. It employs neural and fuzzy techniques on a side{by{side basis using each one for the task it is best suited for. In this paper, our main interest is with the adaptation of the fuzzy control knowledge. The adaptation algorithm is based on reinforcement signals and directly optimizes the global fuzzy relation representing the complete knowledge base. The new approach is experimentally evaluated.
منابع مشابه
Reinforcement-driven adaptation of control relations - Fuzzy Information Processing Society, 1996. NAFIPS., 1996 Biennial Conference of the North America
T h e conceptual f ramework of a hybrid control s y s t e m architecture i s briefly motivated. It employs neural and f u z z y techniques on a side-by-side basis using each one for the task it i s best suited for. I n this paper, our m a i n interest i s with the adaptation of the f u z z y control knowledge. T h e adaptation algorithm i s based o n reinforcement signals and directly optimizes...
متن کاملAdaptive RBF network control for robot manipulators
TThe uncertainty estimation and compensation are challenging problems for the robust control of robot manipulators which are complex systems. This paper presents a novel decentralized model-free robust controller for electrically driven robot manipulators. As a novelty, the proposed controller employs a simple Gaussian Radial-Basis-Function Network as an uncertainty estimator. The proposed netw...
متن کاملEvaluation of Static and Dynamic Methods for Determining the Bearing Capacity of the Driven Pipe Piles (TECHNICAL NOTE)
Determination of the pile capacity is always a major concern for the safe geotechnical engineering design. Different direct and indirect methods have been proposed for determining pile bearing capacity such as static and dynamic approaches. Each method depends on the condition of the soil and type of pile as well as accessibility to necessary information. In this study bearing capacity of drive...
متن کاملLightweight Adaptation in Model-Based Reinforcement Learning
Reinforcement learning algorithms can train an agent to operate successfully in a stationary environment. Most real-world environments, however, are subject to change over time. Research in the areas of transfer learning and lifelong learning addresses this problem by developing new algorithms that allow agents to adapt to environment change. Current trends in this area include model-free learn...
متن کاملLearning Complex Swarm Behaviors by Exploiting Local Communication Protocols with Deep Reinforcement Learning
Swarm systems constitute a challenging problem for reinforcement learning (RL) as the algorithm needs to learn decentralized control policies that can cope with limited local sensing and communication abilities of the agents. Although there have been recent advances of deep RL algorithms applied to multi-agent systems, learning communication protocols while simultaneously learning the behavior ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 1996