作者: Kunihiko Nakazono , Kouhei Ohnisihi , Hiroshi Kinjo
DOI: 10.1109/ICMECH.2007.4280043
关键词: Genetic algorithm 、 Artificial neural network 、 Control theory 、 Rotation (mathematics) 、 Neuro controller 、 Control engineering 、 Nonholonomic system 、 Scheme (programming language) 、 Swing 、 Engineering 、 Control theory
摘要: A neuro-controller for load swing suppression in a jib crane system involving only rotation about the vertical axis is proposed. The controller trained by genetic algorithm, substantially simplifying design of controller. As such nonholonomic, conventional control problem difficult to solve, requiring knowledge complex theory. Using simple three-layered neural network as algorithm-based training, it demonstrated that scheme with performance comparable methods can be obtained relatively approach.