Abstract
Two parallel installed CCDs are used to capture the stereo photo vision of a three-dimensional (3D) pneumatic arm. The parallel stereo vision can be adapted for the 3D pneumatic arm’s feedback control signals via the stereo triangulation and co-ordinate transformation. Since the imaging process restricts the sampling rate, a self-organizing sliding-mode fuzzy controller is implemented to simplify the fuzzy rules to reduce the computer load and its learning mechanism can optimize fuzzy rules online to improve the control performance. The objective of this paper is to compare the measuring accuracy and control performance of non-contacted vision-based and contacted encoder-based 3D pneumatic arms, and then justify the feasibility of this parallel stereo vision-based 3D pneumatic arm by a variety of trajectory tracking experiments.
Get full access to this article
View all access options for this article.
