Trajectory tracking control of a self-balancing robot via adaptive neural networks
Isaac Gandarilla, Jorge Montoya–Cháirez, Víctor Santibáñez, Carlos Aguilar-Avelar, Javier Moreno–Valenzuela
Abstract
In order to ensure trajectory tracking on a two degrees-of-freedom self-balancing robot (SBR) a control scheme, based on the combination of adaptive neural networks and input–output linearization, is presented in this paper. Both external and internal dynamics are analyzed and proof of uniform ultimate boundedness of the position errors is given. The controller performance is assessed via real-time experiments and compared with respect to other control schemes. Better tracking accuracy and disturbance rejection capability is produced by the introduced controller.
Topics & Concepts
Control theory (sociology)TrajectoryComputer scienceController (irrigation)Tracking (education)Artificial neural networkFeedback linearizationRobotAdaptive controlPosition (finance)LinearizationScheme (mathematics)Control engineeringControl (management)MathematicsArtificial intelligenceEngineeringNonlinear systemAgronomyPedagogyAstronomyEconomicsBiologyQuantum mechanicsFinanceMathematical analysisPsychologyPhysicsAdaptive Control of Nonlinear SystemsIterative Learning Control SystemsExtremum Seeking Control Systems