Dynamic neural networks for parallel stewart platforms

Abstract

Redundancy resolution is a critical problem in the control of parallel Stewart platform. The redundancy endows us with extra design degree to improve system performance. In this thesis, the kinematic control problem of Stewart platforms is formulated to a constrained quadratic programming. The KKT conditions of the problem is obtained by considering the problem in its dual space, and then a dynamic neural network is designed to solve the optimization problem recurrently. Theoretical analysis reveals the global convergence of the proposed neural network to the optimal solution in terms of the defned criteria. Simulation results verifes the effectiveness in the tracking control of the Stewart platform for dynamic motions. Redundancy resolution of parallel manipulators is widely studied and have brought many challenges in the control of robotic manipulators. The dual neural network, which is categorized under the recurrent neural networks inherits parallel processing capabilities, are widely investigated for the control of serial manipulators in past decades and has been extended to the control of parallel Stewart platforms in our previous works. However, conventional dual neural network solutions for redundancy resolution requires prior knowledge of the robot, which may not be accessible accurately in real time applications. In this thesis, we establish a model-free dual neural network to control the end-effector of a Stewart platform for the tracking of a desired spacial trajectory, at the same time as learning the unknown time-varying parameters. The proposed model is purely data driven. It does not rely on the system parameters as apriori and provides a new solution for stabilization of the self motion of Stewart platforms. Theoretical analysis and results show that we can achieve a globally convergent neural model in this thesis. It is also shown to be optimal under the model free criterion. In this thesis, we carried out numerical simulations which highlight and illustrate relateable performance capability in terms of model-free optimization. Simulation results provided, verify the tracking control of the end effector while controlling the dynamic motion of the Stewart platform.