Vision-based multi-point sensing for corridor navigation of Autonomous Indoor Vehicle

Autonomous Indoor Vehicle (AIV) is used for various purposes so it can reduce human workload. This study aims to develop a corridor navigation system for AIV that utilizes vision-based multi point sensing. Multi-point sensing techniques on floor and corridor areas are used to detect obstacle-free areas and estimate the direction of AIV. Area detection and direction estimation are used as input information of navigation algorithm based on a fuzzy inference system to control AIV movement. The results of the experiments show that AIV can run along the corridor with the largest mean error of 6.49% with respect the centerline of the corridor.