A minimum phase output based tracking control of ball and plate systems

Kan, Dehui; Xing, Bin; Xie, Wei; Zhu, Linquan

doi:10.1007/s40435-021-00824-1

Search citation statements

Order By: Relevance

Paper Sections

Select...

Introduction1

Citation Types

Supporting

Mentioning

Contrasting

Year Published

2023

2024

Publication Types

Select...

Article2

Relationship

Self Cite0

Independent2

Authors

Journals

Cited by 2 publications

(1 citation statement)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since the ball-on-plate system has attracted great interest in the field of non-linear controls, many methods have been developed. PD or PID type controls, sliding mode control, Fuzzy, Neural PID and Neural Fuzzy, as well as many other adaptive control methods are some of the most notable examples [13][14][15][16][17][18][19][20].…”

Section: Introductionmentioning

confidence: 99%

Two-Dimensional Positioning with Machine Learning in Virtual and Real Environments

Kóczi

Németh²,

Sárosi

2023

Electronics

View full text Add to dashboard Cite

In this paper, a ball-on-plate control system driven only by a neural network agent is presented. Apart from reinforcement learning, no other control solution or support was applied. The implemented device, driven by two servo motors, learned by itself through thousands of iterations how to keep the ball in the center of the resistive sensor. We compared the real-world performance of agents trained in both a real-world and in a virtual environment. We also examined the efficacy of a virtually pre-trained agent fine-tuned in the real environment. The obtained results were evaluated and compared to see which approach makes a good basis for the implementation of a control task implemented purely with a neural network.

show abstract