Event-triggered finite-time control for a constrained robotic manipulator with flexible joints

In the article, an event-triggered control policy is developed for a robotic manipulator with flexible joint (RMFJ) in the presence of finite-time convergence under the requirement of output constraints. Due to the vibration existing in the flexible joint, compared with the rigid manipulator, it is exceedingly difficult for RMFJ to design an adaptive method for realizing highly accurate control. In order to reduce and restrain the vibration of the flexible joint, a finite-time convergence policy is designed by introducing the fractional order term. Moreover, the angular displacements of the link are constrained in the predefined region by using the Barrier function, which further enhances the operation safety. Neural networks learning is used to approximate unknown model parameters. Event-trigger design is used to reduce the communication burden.