Barrier Lyapunov Function Based Adaptive Neural Network Position Constraint Task Space Control of Kinematically Redundant Manipulators
Bildiriler > Tam Metin Bildiri · 2026
Yazarlar: Çağrı HİNDİSTAN, Bayram Melih Yilmaz, Enver TATLICIOĞLU, Erkan Zergeroglu, MM Michalek, D Pazderski, A Bartoszewicz, J Kacprzyk
22nd Polish Control Conference-PCC, Polonya, Poznan, Uluslararası, Sözlü Sunum, Tam Metin Bildiri, 2026-2026, cilt 2071, ss 301-313
DOI: 10.1007/978-3-032-32213-5_25
Özet
An adaptive control methodology is developed in this work to regulate task space motions of kinematically redundant manipulators while rigorously respecting user-defined prescribed constraints. To enforce task space constraints while handling model uncertainties, the control structure combines a barrier Lyapunov function based design with an adaptive neural network, enabling both uncertainty compensation and guaranteed constraint satisfaction under a unified stability analysis. This integration leads to a reliable and constraint aware control framework capable of ensuring stable tracking behavior for safety-critical robotic applications. In addition, a null space controller is designed to exploit the kinematic redundancy, allowing the manipulator to meet a secondary objective without interfering with the primary task space objective. A comprehensive Lyapunov based stability analysis is conducted to rigorously guarantee both the boundedness of the tracking error and the satisfaction of the user-imposed constraints. Comparative numerical simulations are performed to validate the feasibility and effectiveness of the proposed adaptive neural network based task space constraint control strategy.