Design of an augmented telerobotic stereo vision system and associated security concerns

摘要

A telerobotic stereo vision system is used to extend operator's eye-hand motion co-ordination to a distance and has been used in many applications. One of the most critical problems of such system is communication delay that leads to teleoperation instability. This problem can be minimized by using Augmented Reality (AR) concepts of superimposing virtual objects onto the real video image of the workspace to create a simulation plan in the local machine. This facilitates trial and error increasing task safety with reduced network interactions. In this paper, a detailed model to augment a given stereo vision system with AR is presented. At first, a six DOF 3D graphical arm is designed and then superimposed onto the video image using camera calibration and registration methods. Motion activation algorithms are developed and interfaces are designed to facilitate task simulations. Data transmission security and integrity over the network are also discussed. The system is implemented using Microsoft .NET with Visual C# and DirectX 9 to augment a stereo vision system comprising of a PUMA-560 robot operating over a LAN.