Indonesian Journal of Electrical Engineering and Informatics (IJEEI) Vol. 10, No. 4, December 2022, pp. 983~995 ISSN: 2089-3272, DOI: 10.52549/ijeei.v10i4.4216  983 Journal homepage: http://section.iaesonline.com/index.php/IJEEI/index ANFIS multi-tasking algorithm implementation scheme for ball-on-plate system stabilization Oussama Hadoune 1 , Mohamed Benouaret 2 1,2 Department of Electronics, LASA Laboratory, Faculty of Technology, University of Badji Mokhtar Annaba, Algeria Article Info ABSTRACT Article history: Received Oct 10, 2022 Revised Dec 22, 2022 Accepted Jan 2, 2023 This paper presents the design and realization of a ball-on-plate system using a 3-degree-of-freedom parallel robot controlled by an adaptive neuro-fuzzy in- ference system. The ball-on-plate system is nonlinear, multivariable, with an under-actuated feature. Initially, the parallel robot is designed using SolidWorks and mechanized using a computer numerical control machine. Followed by the presentation of the ball-on-plate system mathematical model and the simplified model obtained. Afterwards, the inverse kinematics are performed to derive the appropriate angle for each servomotor. Eventually, the controller is designed and implemented in a double loop feedback scheme. A comparison between the proposed controller and a conventional proportional– integral–derivative controller in terms of time response, overshoot, and steady- state error is carried out. Furthermore, a comparison between sequential and asynchronous parallel processing is conducted for two different scenarios. The first scenario is when moving the ball to the origin while the second is for disturbance rejection. Simulation and experimental results show that the adaptive neuro-fuzzy inference system implemented using asynchronous parallel processing improves the real-time system stability by considerably decreasing oscillations as well as enhancing the ball movement smoothness with a small stead-state error. Keywords: Parallel robot ANFIS Multitasking Ball-on-plate Neural-network Copyright © 2022 Institute of Advanced Engineering and Science. All rights reserved. Corresponding Author: Oussama HADOUNE Department of Electronics, Faculty of Technology University of Badji Mokhtar Annaba Bp 12, 23000 Annaba, Algeria Email: oussama.hadoune@univ-annaba.org 1. INTRODUCTION Parallel robots are often used in different areas such as industrial practices, flight simulators, and medical science due to their high stiffness (the ability to lift a heavy weight that is shared by multiple legs), high-speed task execution (due to low moment of inertia), and their ability to solve high-precision problems. The ball-on-plate system (BPS) is generally used on 2-Degrees of Freedom (DOF), which is used as an educational platform where controllers can be tested. This type of ball-on-plate system is mainly used by schools and research institutes to help improve students’ learning process. In this regard, the ball -on-plate system can be an excellent educational kit to understand nonlinear control, uncertainties, and controller design. A rotary pendulum system is a benchmark problem as well; however, it can be dangerous if we lose control over it. Therefore, the ball-on-plate system is the most suitable platform on which controllers can be tested safely. Several research papers on the ball-on-plate system have already been published. Tudic et al. [1] conducted a comparative study between PID and a PD controllers in which they were applied to a 2-DOF ball balancer that was designed and manufactured using a 3D printer. Note that both controllers were implemented on an Arduino platform. The experimental results demonstrate that the PD algorithm prove to be more successful than the PID controller. Furthermore, Chevalier et al. [2] used a ball-on plate system on a 6-DOF Stewart platform controlled by an automatic calibration with a robust PD controller to compensate for the characterized dynamic model changes by the use of balls with different surfaces, masses, and diameters without