Intelligent Fuzzy Q-Learning Control of Humanoid Robots

Er, Meng Joo; Zhou, Yi

doi:10.1007/11427469_34

Cited by 3 publications

(1 citation statement)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The reward evaluation can be defined by scaled values as "-1" for failure, "+1" for success and "0" for neutral. Moreover, continuous functions and fuzzy evaluation can also be adopted for the reward function as well [131,132]. If an action is followed by a satisfactory state of affairs or an improvement in the state of affairs, the tendency to produce that action is strengthened or reinforced (rewarded).…”

Section: Basic Elements Of Reinforcement Learning Systemmentioning

confidence: 99%

Automatic generation of fuzzy neural networks via reinforcement learning with applications in path planning of mobile robots

Zhou¹

Self Cite

View full text Add to dashboard Cite

Clustering and Q − value based Genetic Algorithm learning schemes f or F uzzy system design DF N N Dynamic F uzzy N eural N etwork DF QL Dynamic F uzzy Q − Learning DP Dynamic P rogramming DSGF QL Dynamic Self − Generated F uzzy Q − Learning DSGF QL − CA Dynamic Self − Generated F uzzy Q − Learning and Continuous Action Q − learning EBF Ellipsoidal Basis F unction EDSGF QL Enhanced Dynamic Self − Generated F uzzy Q − Learning F IS F uzzy Inf erence System F N N F uzzy N eural N etwork F QL F uzzy Q − Learning GA Genetic Algorithm GDF N N General Dynamic F uzzy N eural N etwork M F M embership F unction M OM M ean Of M aximum M OM O M ulti − Input M ulti − Output M OSI M ulti − Input Single − Output x

show abstract

Section: Basic Elements Of Reinforcement Learning Systemmentioning

confidence: 99%