Handling time delay of sensor data processing in a model based remote operation environment
Many of today's inspection systems for underground sewer pipes consist of a single TV-Camera on an operator con trolled vehicle. The live-video of this camera is the pri mary tool for teleoperation and damage assessment. Recent developments in the field have lead to systems with more advanced sensors as well as a greater degree of autonomy. While human involvement in teleoperation is still neces sary, the operator interface has to meet different challenges compared to conventional systems. Here we present an augmented virtual environment that can be used by the operator for supervisory control of our semi-autonomous inspection system for large under ground sewer pipes under operating conditions. At runtime the virtual environment is constructed using results from a number of data-processing modules that are unable to de liver their output in time for live-display. To overcome this problem we introduce the use of spatially arranged time do mains. We then show in detail how this technique is applied to achieve an overall consistent visualization.