Litcius/Paper detail

Discrete-Time Circadian Rhythms Neural Network for Perturbed Redundant Robot Manipulators Tracking Problem With Periodic Noises

Zhijun Zhang, Siyuan Chen, Junjie Liang

2021IEEE Transactions on Industrial Informatics22 citationsDOI

Abstract

Via the Euler forward-difference rule, an Euler-type discrete-time circadian rhythms neural network model (E-DTCRNN) is proposed, developed, and investigated for motion planning of the redundant robot manipulator affected by periodic noises. In this article, an Euler-type discrete-time zeroing neural network model (E-DTZNN) is presented as comparison. The E-DTCRNN model is 0-stable, consistent, and convergent. In addition, through a hybrid torque and velocity optimization scheme synthesized by the proposed E-DTCRNN and the traditional E-DTZNN, a tracking trajectory is designed and applied to the motion planning of the redundant robot manipulator. Finally, groups of simulations and physical experiments verify the efficacy and noise suppression ability of the proposed E-DTCRNN model for motion planning of the manipulator.

Topics & Concepts

Control theory (sociology)Artificial neural networkNoise (video)Computer scienceRobotTrajectoryEuler's formulaEuler anglesTorqueTracking (education)MathematicsArtificial intelligencePhysicsControl (management)Mathematical analysisThermodynamicsGeometryImage (mathematics)PedagogyAstronomyPsychologyNeural Networks and ApplicationsIndustrial Technology and Control SystemsAdvanced Data Processing Techniques