Hierarchical neuro-fuzzy call admission controller for ATM networks

Chatovich, A; Oktuğ, Sema; Dündar, Günhan

doi:10.1016/s0140-3664(00)00331-5

Cited by 16 publications

(6 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this work, we have used the Generalised Approximate Reasoning-based Intelligent Control (GARIC) architecture of Berenji and Khedkar [24] as the basis of our system, primarily because of the facility it gives to combine actor-critic and neurofuzzy methods. It has been widely employed in similar works on intelligent control (e.g., [25][26][27]). GARIC consists of two neural networks: the Action Selection Network (ASN) operating as the actor and the Action Evaluation Network (AEN) which criticises the actions made by the ASN.…”

Section: Actor-critic Methodsmentioning

confidence: 99%

Adaptive neurofuzzy control of a robotic gripper with on-line machine learning

Domínguez-López

Damper

Crowder

et al. 2004

Robotics and Autonomous Systems

View full text Add to dashboard Cite

Pre-programming complex robotic systems to operate in unstructured environments is extremely difficult because of the programmer's inability to predict future operating conditions in the face of unforeseen environmental conditions, mechanical wear of parts, etc. The solution to this problem is for the robot controller to learn on-line about its own capabilities and limitations when interacting with its environment. At the present state of technology, this poses a challenge to existing machine learning methods. We study this problem using a simple two-fingered gripper which learns to grasp an object with appropriate force, without slip while minimising chances of damage to the object. Three machine learning methods are used to produce a neurofuzzy controller for the gripper. These are off-line supervised neurofuzzy learning and two on-line methods, namely unsupervised reinforcement learning and an unsupervised/supervised hybrid. With the two on-line methods, we demonstrate that the controller can learn through interaction with its environment to overcome simulated failure of its sensors. Further, the hybrid is shown to out perform reinforcement learning alone in terms of faster adaptation to the changing circumstances of sensor failure. The hybrid learning scheme allows us to make best use of such pre-labeled datasets as might exist and to remember effectively good control actions discovered by reinforcement learning.

show abstract

Section: Actor-critic Methodsmentioning

confidence: 99%

Adaptive neurofuzzy control of a robotic gripper with on-line machine learning

Domínguez-López

Damper

Crowder

et al. 2004

Robotics and Autonomous Systems

View full text Add to dashboard Cite

show abstract

“…It obtains the knowledge through trial-and-error and interaction with environment to improve its behavior policy. Because of the advantages above, RL has been played a very important role in the flow control in high-speed networks [4][5][6][7][8].…”

Section: Introductionmentioning

confidence: 99%

A Genetic Based Fuzzy Markov Game Flow Controller for High-speed Networks

2011

2011 Third International Conference on Measuring Technology and Mechatronics Automation

View full text Add to dashboard Cite

For the congestion problems in high-speed networks, a genetic based fuzzy Markov game flow controller (GFMC) is proposed. Because of the uncertainties and highly timevarying, it is not easy to accurately obtain the complete information for high-speed networks. In this case, the fuzzy Markov game, which is independent of mathematic model, and prior-knowledge, has good performance. It offers a promising platform for robust control in the presence of the bounded external disturbances. The genetic operators are used to obtain the consequent parts of fuzzy rules. Simulation results show that the proposed controller can learn to take the best action to regulate source flow with the features of high throughput and low packet loss ratio, and can avoid the occurrence of congestion effectively.

show abstract

“…So it has the ability of self-learning. Because of the advantages above, RL has been played a very important role in the flow control in high-speed networks [4][5][6][7]. The Q-learning algorithm of RL is easy for application and has a firm foundation in the theory.…”

Section: Introductionmentioning

confidence: 99%

A Genetic Based Fuzzy Q-Learning Flow Controller for High-Speed Networks

Li¹,

Jing²,

Jiang³

et al. 2009

IJCNS

View full text Add to dashboard Cite

For the congestion problems in high-speed networks, a genetic based fuzzy Q-learning flow controller is proposed. Because of the uncertainties and highly time-varying, it is not easy to accurately obtain the complete information for high-speed networks. In this case, the Q-learning, which is independent of mathematic model, and prior-knowledge, has good performance. The fuzzy inference is introduced in order to facilitate generalization in large state space, and the genetic operators are used to obtain the consequent parts of fuzzy rules. Simulation results show that the proposed controller can learn to take the best action to regulate source flow with the features of high throughput and low packet loss ratio, and can avoid the occurrence of congestion effectively.

show abstract

Hierarchical neuro-fuzzy call admission controller for ATM networks

Cited by 16 publications

References 15 publications

Adaptive neurofuzzy control of a robotic gripper with on-line machine learning

Adaptive neurofuzzy control of a robotic gripper with on-line machine learning

A Genetic Based Fuzzy Markov Game Flow Controller for High-speed Networks

A Genetic Based Fuzzy Q-Learning Flow Controller for High-Speed Networks

Contact Info

Product

Resources

About