Original scientific paper
Performance Comparison of Different Control Algorithms for Robot Manipulators
Tihomir ŽILIĆ
orcid.org/0000-0001-7951-444X
; Faculty of Mechanical Engineering and Naval Architecture University of Zagreb, Croatia
Josip KASAĆ
; Faculty of Mechanical Engineering and Naval Architecture University of Zagreb, Croatia
Mario ESSERT
; Faculty of Mechanical Engineering and Naval Architecture University of Zagreb, Croatia
Branko NOVAKOVIĆ
; Faculty of Mechanical Engineering and Naval Architecture University of Zagreb, Croatia
Željko ŠITUM
; Faculty of Mechanical Engineering and Naval Architecture University of Zagreb, Croatia
Abstract
In this work are compared performances of five different robot control algorithms. The following controllers under consideration are: PD controller, PID controller, analytical fuzzy controller, classical adaptive controller and adaptive controller based on neural network. The mentioned controllers are used to control two different robot configurations with two rotational degrees of freedom (in horizontal and vertical plane). The basic performances for control algorithms comparisons are: tracking error, rate of convergence, robustness on structural changes of control object, complexity of stability criterion and complexity of implementation.
Keywords
Adaptive control; Control performances; Fuzzy control; Neural networks; Robot control
Hrčak ID:
97950
URI
Publication date:
31.10.2012.
Visits: 1.547 *