Title: Adaptive neural super twisting controller based on terminal sliding mode and time delay estimation method for robotic manipulator
Authors: Amar Rezoug; Mustapha Hamerlain
Addresses: Division Robotique et Productique, Centre de Développement des Technologies Avancées, Cité 20 aout 1956, baba hassen, Algiers, Algeria ' Division Robotique et Productique, Centre de Développement des Technologies Avancées, Cité 20 aout 1956, baba hassen, Algiers, Algeria
Abstract: In this paper a robust control approach based on sliding mode theory and artificial neural network technique was proposed for trajectory tracking mode of n-DOF robot manipulator. Time delay estimation method and non-singular terminal sliding mode control were used to design the nominal control part of the sliding mode control without any knowledge about the robot model. Super twisting algorithm was designed using radial based function neural networks to replace the discontinuous control part. In order to test the effectiveness of the proposed approach, it was applied to 2-DOF robotic manipulator and compared with a classical approach.
Keywords: non-singular terminal sliding mode; robotic manipulator; radial based function neural networks; RBFNN; super twisting; time delay estimation.
International Journal of Digital Signals and Smart Systems, 2017 Vol.1 No.4, pp.348 - 364
Received: 02 Jan 2017
Accepted: 11 Aug 2017
Published online: 09 Feb 2018 *