Abstract




 
   

IJE TRANSACTIONS B: Applications Vol. 31, No. 8 (August 2018) 1283-1291    Article in Press

PDF URL: http://www.ije.ir/Vol31/No8/B/17-2862.pdf  
downloaded Downloaded: 26   viewed Viewed: 115

  APPLICATION OF WAVELET NEURAL NETWORK IN FORWARD KINEMATICS SOLUTION OF 6-RSU CO-AXIAL PARALLEL MECHANISM BASED ON FINAL PREDICTION ERROR
 
A. Rahmani
 
( Received: January 17, 2018 – Accepted in Revised Form: March 09, 2018 )
 
 

Abstract    Application of artificial neural network (ANN) in forward kinematic solution (FKS) of a novel co-axial parallel mechanism with six degrees of freedom (6-DOF) is addressed in Current work. The mechanism is known as six revolute-spherical-universal (RSU) and constructed by 6-RSU co-axial kinematic chains in parallel form. First, applying geometrical analysis and vectorial principles the kinematic model is extracted and inverse kinematics solution is done. Due to highly nonlinear characteristic of the model, forward kinematic solution for 6-RSU is so complicated. Therefore, ANN based on wavelet analysis, as a powerful solution, is designed and applied to solve FK problem. The minimum prediction risk principle with using final prediction error (FPE) is applied to find the best and optimum topology of our proposed neural network (WNN) in this paper. Furthermore, proposed wavelet WNN is developed to approximate the specific reference trajectories for manipulated platform of mechanism and the results are obtained. Comparing the extracted results by WNN with closed form solution (CFS) demonstrates the accuracy and efficiency of the proposed WNN.

 

Keywords    Wavelet Neural Network, Kinematic Analysis, 6-RSU Parallel Mechanism, Final Prediction Error

 

چکیده    در این مقاله ابتدا یک مکانیزم موازی جدید بنام HEXAROT که در مقایسه با سایر مکانیزم های موازی دارای فضای کاری به مراتب بیشتری است، معرفی می شود. این مکانیزم 6 درجه آزادی بوده و با توجه به ساختار سینماتیکی آن به مکانیزم 6-RSU نیز معروف می باشد. سپس با استفاده از جبر بردارها و روش ماتریس هموژن مدل سینماتیکی مکانیزم استخراج و حل سینماتیک معکوس مکانیزم به جهت تولید دیتاهای لازم برای آموزش شبکه انجام می شود. از آنجاییکه بدلیل پیچیده بودن مدل سینماتیکی مکانیزم مذکور که ناشی از غیرخطی بودن شدید معادلات سینماتیکی است، محاسبات مربوط به حل سینماتیک مستقیم عملا غیر ممکن است از اینرو مدل شبکه عصبی ویولت (WNN) جهت حل سینماتیک مستقیم طراحی و استفاده شده و نتایج بدست آمده از آن با نتایج حاصل از سینماتیک معکوس صحه گذاری گردیده است. مدل شبکه عصبی مصنوعی ارایه شده در این مقاله یک شبکه سه لایه با نرون های فعالسازی مارلت (Morlet) و سیگمویید (Sigmoied) برای لایه های میانی و خروجی است که توپولوژی بهینه برای آن با استفاده از روش حداقل ریسک پیش بینی (MPR) و پیش بینی خطای نهایی (FPE) به دست آمده است. در این مقاله جهت شبیه سازی و بررسی عملکرد شبکه عصبی طراحی شده دو مسیر با معادلات پیچیده غیرخطی در نظر گرفته شده است. مسیر I جهت تولید دیتاها برای آموزش شبکه و مسیر II جهت صحه گذاری شبکه آموزش داده شده استفاده شده است. مقایسه نتایج بدست آمده از شبکه عصبی با نتایج حاصل از حل دقیق بیانگر دقت و کارآمدی شبکه عصبی طراحی شده می باشد.

References   

1.     Isaksson, M., Gosselin, C. and Marlow, K., "Singularity analysis of a class of kinematically redundant parallel schönflies motion generators", Mechanism and Machine Theory,  Vol. 112, (2017), 172-191.

2.     Li, T., Li, F., Jiang, Y., Zhang, J. and Wang, H., "Kinematic calibration of a 3-p (pa) s parallel-type spindle head considering the thermal error", Mechatronics,  Vol. 43, (2017), 86-98.

3.     Mohan, S., "Error analysis and control scheme for the error correction in trajectory-tracking of a planar 2prp-ppr parallel manipulator", Mechatronics,  Vol. 46, (2017), 70-83.

4.     Pedrammehr, S., Danaei, B., Abdi, H., Masouleh, M.T. and Nahavandi, S., "Dynamic analysis of hexarot: Axis-symmetric parallel manipulator", Robotica,  Vol. 36, No. 2, (2017), 225-240.

5.     Schreiber, L.-T. and Gosselin, C., "Kinematically redundant planar parallel mechanisms: Kinematics, workspace and trajectory planning", Mechanism and Machine Theory,  Vol. 119, (2018), 91-105.

6.     Herrero, S., Pinto, C., Altuzarra, O. and Diez, M., "Analysis of the 2pru-1prs 3dof parallel manipulator: Kinematics, singularities and dynamics", Robotics and Computer-Integrated Manufacturing,  Vol. 51, (2018), 63-72.

7.     Ghasemi, J., "Kinematic synthesis of parallel manipulator via neural network approach", International Journal of Engineering Transactions C: Aspects,  Vol. 30, No. 9, (2017), 319-1325.

8.     Xu, Q., Yang, Y., Jing, Z. and Hu, S., "Forward kinematics analysis for a class of asymmetrical parallel manipulators", International Journal of Advanced Robotic Systems,  Vol. 14, No. 1, (2017), 1729881416678132.

9.     Huang, G., Guo, S., Zhang, D., Qu, H. and Tang, H., "Kinematic analysis and multi-objective optimization of a new reconfigurable parallel mechanism with high stiffness", Robotica,  Vol. 36, No. 2, (2017), 187-203.

10.   Gao, L. and Wu, W., "Forward kinematics modeling of spatial parallel linkage mechanisms based on constraint equations and the numerical solving method", Robotica,  Vol. 35, No. 2, (2017), 293-309.

11.   Sadjadian, H., Taghirad, H. and Fatehi, A., "Neural networks approaches for computing the forward kinematics of a redundant parallel manipulator", International Journal of Computational Intelligence,  Vol. 2, No. 1, (2005), 40-47.

12.   Rahmani, A. and Ghanbari, A., "Application of neural network training in forward kinematics simulation for a novel modular hybrid manipulator with experimental validation", Intelligent Service Robotics,  Vol. 9, No. 1, (2016), 79-91.

13.   Lu, Y., Wang, P., Zhao, S., Hu, B., Han, J. and Sui, C., "Kinematics and statics analysis of a novel 5-dof parallel manipulator with two composite rotational/linear active legs", Robotics and Computer-Integrated Manufacturing,  Vol. 30, No. 1, (2014), 25-33.

14.   Qazani, M.R.C., Pedrammehr, S., Rahmani, A., Danaei, B., Ettefagh, M.M., Rajab, A.K.S. and Abdi, H., "Kinematic analysis and workspace determination of hexarot-a novel 6-dof parallel manipulator with a rotation-symmetric arm system", Robotica,  Vol. 33, No. 08, (2015), 1686-1703.

15.   Qazani, M.R.C., Pedrammehr, S., Rahmani, A., Shahryari, M., Rajab, A.K.S. and Ettefagh, M.M., "An experimental study on motion error of hexarot parallel manipulator", The International Journal of Advanced Manufacturing Technology,  Vol. 72, No. 9-12, (2014), 1361-1376.

16.   Bazoobandi, H., "Wavelet neural network with random wavelet function parameters", International Journal of Engineering-Transactions A: Basics,  Vol. 30, No. 10, (2017), 1510-1516.

17.   Hashemi, S.M.A., Haji Kazemi, H. and Karamodin, A., "Verification of an evolutionary-based wavelet neural network model for nonlinear function approximation", International Journal of Engineering,  Vol. 28, No. 10, (2015), 1423-1429.

18.   Huang, M. and Cui, B., "A novel learning algorithm for wavelet neural networks", ICNC 2005: Advances in Natural Computation,  Springer-Verlag Berlin Heidelberg, (2005), 421-421.

19.   Neshat, N., "An approach of artificial neural networks modeling based on fuzzy regression for forecasting purposes", International Journal of Engineering-Transactions B: Applications,  Vol. 28, No. 11, (2015), 1651-1655.

20.           Moody, J., Prediction risk and architecture selection for neural networks, in From statistics to neural networks. 1994, Springer.147-165.


Download PDF 



International Journal of Engineering
E-mail: office@ije.ir
Web Site: http://www.ije.ir