2020
DOI: 10.1007/978-3-030-44289-7_29
|View full text |Cite
|
Sign up to set email alerts
|

Deep Learning Based Kinematic Modeling of 3-RRR Parallel Manipulator

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

0
4
0
2

Year Published

2020
2020
2023
2023

Publication Types

Select...
4
2
1

Relationship

0
7

Authors

Journals

citations
Cited by 21 publications
(6 citation statements)
references
References 21 publications
0
4
0
2
Order By: Relevance
“…Table 1 shows the validation of the network and the error obtained between the trained network and the data obtained from the Newton-Raphson method. In order to obtain an improved neural network, experimentation was carried out taking into account and varying the number of layers (2,3,4), the number of neurons per layer (3,4,5), the number of epochs (1000, 2000, 3000) and the yields (0.001, 0.0001, 0.00001). The total number of runs was 81.…”
Section: Training Of a Neural Networkmentioning
confidence: 99%
See 1 more Smart Citation
“…Table 1 shows the validation of the network and the error obtained between the trained network and the data obtained from the Newton-Raphson method. In order to obtain an improved neural network, experimentation was carried out taking into account and varying the number of layers (2,3,4), the number of neurons per layer (3,4,5), the number of epochs (1000, 2000, 3000) and the yields (0.001, 0.0001, 0.00001). The total number of runs was 81.…”
Section: Training Of a Neural Networkmentioning
confidence: 99%
“…Machine learning is an Artificial Intelligence (AI) technique that is currently being used to solve robot kinematic models as an alternative to Newton-Raphson, such as neural networks [4]. Other AI algorithms have been used for applications in robot kinematics, for example, in [5] a model of a Neuro-Fuzzy inference system was built to predict the position of the end-effector of a parallel RRR-type robot within the workspace and tuned with a particle swarm optimization and a genetic algorithm. In [6], the direct kinematic problem of a Stewart platform was solved using soft computing and, subsequently, a particle swarm optimization method was used and a multilayer neural network was trained to solve the forward kinematics problem.…”
Section: Introductionmentioning
confidence: 99%
“…Selanjutnya, dengan mensubstitusikan persamaan (10) dan (11) ke dalam persamaan (7) untuk i = 1 dapat diperoleh hubungan secara matematis yaitu 𝐹 𝑥 2 + 𝐹 𝑦 2 + 2(𝐸 1𝑥 𝐹 𝑥 + 𝐸 1𝑦 𝐹 𝑦 )𝐹 𝑥𝑦 + 𝐸 10 𝐹 𝑥𝑦 2 = 0. (12) Koefisien E1x, E1y, E10, Fx, Fy, dan Fxy pada persamaan (12) merupakan fungsi φ. Dengan memanfaatkan hubungan tangen setengah sudut untuk φ c𝜑 = 1−𝜏 2 1+𝜏 2 ; s𝜑 = 2𝜏 1+𝜏 2 ; 𝜏 = tan 𝜑 2 dapat dituliskan persamaan (12) ke dalam suatu polinom berderajat 12 dalam variabel bebas τ.…”
Section: Kinematika Langsungunclassified
“…Tentunya kedua arsitektur ini dilatih (training) berdasarkan lintasan spiral yang dihasilkan oleh suatu fungsi matematika tertentu. Hal senada juga dilakukan oleh Sayed, dkk [10] yang mengimplementasikan neuro fuzzy inference system (NFIS) dengan pentuning-an melalui particle swarm organization (PSO) dan genetic algorithm (GA). Mereka memberikan jawab kinematika langsung dalam respons waktu setelah proses training dan tuning pada arsitektur NFIS yang digunakan.…”
unclassified
“…The outcome demonstrates that the suggested ANN model can be utilized in place of the sophisticated and time-consuming GA to determine the optimal parameters of the gimbal joint. The AI methods are used for kinematic analysis, path generation, and control of the various parallel manipulators [30][31][32]. A fuzzy-logic control method has been developed for a dynamically uncertain robotic manipulator [33] to improve its trajectory tracking accuracy.…”
Section: Introductionmentioning
confidence: 99%