Force/position control strategies provide an effective framework to deal with tasks involving interaction with the environment. One of these strategies proposed in the literature is external force feedback loop control. It fully employs the available sensor measurements by operating the control action in a full dimensional space without using selection matrices. The performance of this control strategy is affected by uncertainties in both the robot dynamic model and environment stiffness. ; The purpose of this paper is to improve controller robustness by applying a neural network technique in order to compensate the effect of uncertainties in the robot model. We show that this control strategy is robust with respect to payload uncertainties, position and environment stiffness, and dry and viscous friction. Simulation results for a three degrees-of-freedom manipulator and various types of environments and trajectories show the effectiveness of the suggested approach compared with classical external force feedback loop structures.
Zielona Góra: Uniwersytet Zielonogórski
AMCS, volume 19, number 1 (2009) ; kliknij tutaj, żeby przejść
Biblioteka Uniwersytetu Zielonogórskiego
2024-04-08
2024-04-08
9
https://zbc.uz.zgora.pl/publication/88530
Dulęba, Ignacy Korbicz, Józef - red. Uciński, Dariusz - red.
Bańka, Stanisław Dworak, Paweł Jaroszewski, Krzysztof Makowski, Ryszard - ed. Zarzycki, Jan - ed.
Dzieliński, Andrzej Korbicz, Józef - red. Uciński, Dariusz - red.
Ławryńczuk, Maciej Tatjewski, Piotr Korbicz, Józef - red. Uciński, Dariusz - red.
Denai, Azzedine Mouloud Attia, Sid Ahmed Korbicz, Józef - red. Uciński, Dariusz - red.
Ławryńczuk, Maciej Korbicz, Józef - ed.
Pedro, Jimoh Olarewaju Dahunsi, Olurotimi Akintunde Korbicz, Józef - red. Uciński, Dariusz - red.
Ahmida, Zahir Charef, Abdelfettah Becerra, Victor M. Korbicz, Józef - red. Uciński, Dariusz - red.