NEURAL NETWORK SOLUTIONS FOR THE CONTROL OF HEXAPOD FOR NVIDIA JETSON EMBEDDED PLATFORM
Abstract
This research is a part of the work implemented by BSTU "Voenmeh" under the financial support of the Ministry of Education and Science of the Russian Federation for design and devel-opment of a precision mechanism with the parallel kinematics called "Hexapod". New released embedded platforms of artificial intelligence involve the interest of research engineers to imple-ment modern control algorithms at a new qualitative level. The purpose of this work is to obtain neural network solutions for hexapod control problems for the modern NVIDIA JETSON embed-ded platform. The control problems of hexapod are presented, which include solving the forward and inverse kinematics, controlling the forces at the hexapod's legs based on the computing of the inverse model of dynamics implementing the desired trajectory in Cartesian coordinates. We pro-pose to apply the neural networks for solving the forward kinematics problem and approximating Jacobi inverse matrices in the problem of computing the inverse model of dynamics. We used theNeural Network Toolbox Matlab for train neural networks and testing the proposed algorithms. The results of the training of neural networks for solving the forward kinematics problem with an accuracy of more than 10 times greater than the specified error of the control system in all work-space are presented. The architecture of the neural network for approximating the Jacobi inverse matrix is presented. The mathematical description of the neural network control algorithms is implemented. An approach to creating software for the NVIDIA JETSON embedded platform is described. The CUDA implementation of the developed algorithms for the JETSON TX1 platform was performed, testing of which showed the triple superiority of parallel algorithms in the speed of solving the forward kinematics problem compared to the traditional iterative approach based on the Newton-Raphson method.
References
2. Merlet J.P. Parallel Robots. Springer, 2006, 420 p.
3. Korotkov E.B., Matveev S.A., YAkovenko N.G. Puti povysheniya kachestvennykh pokazateley sistemy upravleniya mekhanizmom s parallel'noy strukturoy (geksapod, tripod) na baze rossiyskikh i mirovykh dostupnykh elektronnykh komponentov [The paths of increase quality rating of control system of machine with strut-type structure as hexapod and tripod on based russian and global available electronic component], Voprosy radioelektroniki [Questions of ra-dio electronics], 2016, No. 8, pp. 85-91.
4. Artemenko Yu.N., Agapov V.A., Dubarenko V.V., Kuchmin A.Yu. Gruppovoe upravlenie aktuatorami kontrreflektora radioteleskopa [Group control actuators of contraflexure radio telescope], Informatsionno-upravlyayushchie sistemy [Information-control systems], 2012, No. 4, pp. 2-9.
5. Pesternikov A.A., Komarov S.A., Boyko S.O., Kharitonov S.G. Ustroystvo povorota reflektora [The device of rotation of reflector], Reshetnevskie chteniya [Reshetnev's readings], 2010, Vol. 1, No. 14, pp. 80-81.
6. Rybak L.A., Gaponenko E.V., Malyshev D.I. Razrabotka algoritmov i upravlyayushchikh programm dlya realizatsii dvizheniy vykhodnogo zvena robota-geksapoda dlya 3d-pechati pretsizionnykh izdeliy [Development of algorithms and control programs for the implementa-tion of the robot-hexapod output link movements for 3D printing of precision products], Mekhatronika, avtomatizatsiya, upravlenie [Mechatronics, automation, control], 2016, Vol. 17, No. 12, pp. 821-827.
7. Stewart D. A platform with six degrees of freedom, Proc. of the Institution of mechanical en-gineers, 1965, Vol. 180, pp. 371-385.
8. Zhukov Yu.A., Korotkov E.B., Slobodzyan N.S. Sistema upravleniya mekhanizmom s parallel'noy kinematikoy dlya peremeshcheniya bortovykh priborov KLA na baze sovremennogo otechestvennogo radiatsionno-stoykogo mikrokontrollera s protsessornym yadrom Cortex-M4F [Radiation resistant microcontroller with Cortex-M4F core based control system of parallel kinematics mechanism designed for space-craft’s onboard devices move-ments], Voprosy radioelektroniki [Questions of radio electronics], 2017, No. 7, pp. 48-54.
9. Gavrilenko V.A., Zhukov Yu.A., Moroz A.V. Realizatsiya zadach kinematiki na mikroprotsessore ARM-arkhitektury dlya mekhatronnykh sistem upravleniya geksapodom [Kinematics Solutions on the ARM microprocessor for mechatronic control systems of hexa-pods], Voprosy radioelektroniki [Questions of radio electronics], 2016, No. 8, pp. 92-98.
10. Campa R., Bernal J., Soto I. Kinematic Modeling and Control of the Hexapod Parallel Robot, Proceeding of American Control Conference (ACC), 2016, pp. 1203-1208.
11. Cardona M. N. A new Approach for the Forward Kinematics of General Stewart-Gough Plat-forms, Proceedings of the 2015 IEEE Thirty Fifth Central American and Panama Convention (CONCAPAN XXXV), 2015, pp. 1-6.
12. Geng Z., Haynes L. Neural network solution for the forward kinematics problem of a Stewart platform, Proceedings of IEEE International Conference on Robotics and Automation, 1991, Vol. 3, pp. 2650-2655.
13. Choon seng Yee, Kah-bin Lim Forward kinematics solution of Stewart platform using neural networks, Neurocomputing, 1997, Vol. 16, Issue 4, pp. 333-349.
14. Lee Hyung Sang, Myung-Chul Han The estimation for forward kinematic solution of Stewart platform using the neural network, Proceedings of IEEE/RSJ International Conference. Intel-ligent Robots and Systems, 1999, Vol. 1, pp. 501-506.
15. Parikh P.J., Lam S.Y. A Hybrid Strategy to Solve the Forward Kinematics Problem in Parallel Manipulators, IEEE Transactions on Robotics, 2005, Vol. 21, No. 1, pp. 18-25.
16. Rybak L.A., Mamaev Yu.A., Virabyan L.G. Sintez algoritma korrektsii traektorii dvizheniya vykhodnogo zvena robota-geksapoda na osnove teorii iskusstvennykh neyronnykh setey [Cor-rection algorithms synthesis for the motion path of the hexapod robot output link based on the theory of artificial neural networks], Vestnik Belgorodskogo gosudarstvennogo tekhnologicheskogo universiteta im. V.G. SHukhova [Bulletin of Belgorod State Technological University named after. V.G. Shukhov], 2016, No. 12. pp. 142-151.
17. Xinxin Guo, Guixi Ke, Fengwu Zheng, Lijie Zhang Forward Kinematics Analysis of the Stewart Parallel Platform Based on the Elman Recurrent Network, Proceedings of 5th International Con-ference on Intelligent Human-Machine Systems and Cybernetics, 2013, Vol. 2, pp. 175-177.
18. Mohammed A., Li S. Dynamic Neural Networks for Kinematic Redundancy Resolution of Parallel Stewart Platforms, IEEE Transactions on cybernetics, 2016, Vol. 46, Issue 7, pp. 1538-1550.
19. Ramesh Kumar P., Bandyopadhyay B. The forward kinematic modeling of a Stewart platform using NLARX model with wavelet network, Proceedings of 11th IEEE International Confer-ence on Industrial Informatics, 2013, pp. 343-348.
20. Vstraivaemye sistemy [Embedded Systems], Official web site of NVIDIA Corporation, 2018. Available at: https://www.nvidia.ru/autonomous-machines/embedded-systems/ (accessed 05 November 2018).
21. Fu K.S., Gonzalez R.C., Lee C.S. Robotics. Control, Sensing, Vision, and Intelligence. New-York: McGraw-Hill, 1987, 580 p.
22. Dzhukich D.Y., Zhukov Yu.A., Korotkov E.B., Moroz A.V., Slobodzyan N.S. TSifrovoe upravlenie geksapodom na osnove obratnoy modeli dinamiki s realizatsiey na radiatsionno stoykom ARM-mikrokontrollere [Hexapod digital control using the inverse dynamics and it implementation on the radiation-resistant ARM-microcontroller], Voprosy radioelektroniki [Questions of radio electronics], 2018, No. 7, pp. 103-110.
23. Neural Network Toolbox пакет расширения Matlab, Official web site of MathWorks distribu-tor in Russia and the CIS, 2018. 05 ноября. Available at: https://matlab.ru/products/neural-network-toolbox/ (accessed 05 November 2018).
