Real time monitoring and avoidance using FG based 3D vision system

A smart 3D vision sensor system has been developed. This system uses Fiber grating (FG) based 3D vision sensor. The sensor system encodes the space coarsely by using limited number of geometrically interrelated 2D laser spots generated through two perpendicularly overlaid layers of FGs and projected on a targeted scene. The reflected spots and specially the disturbed due to the presence of an object or a person play an essential role in detecting targets. The author developed zone-based strategy to enable person detection within any of the zones, track the transitional movement between zones and use inter-zones relation to conclude human behavioral status. This approach has the advantage of reducing significantly the computation time of 3D information that is required for real time tracking and monitoring of human movement. The system has been implemented and this paper presents its operational characteristics. The developed tracking strategy is introduced with its application to monitor the behavior of a person and to help a mobile robot to detect obstacles.