Regarding sensor failures

tbohne / msc

Master's Thesis at Osnabrueck University / DFKI

0 stars 0 forks source link

Regarding sensor failures #104

Closed tbohne closed 2 years ago

tbohne commented 2 years ago

for each case (total failure, inf values, etc.) - when do these issues occur in practice, which real case are they representing?
detecting issues that aren't occurring in practice is meaningless

add in text: it is a valid assumption to receive the scans via some topic (practical perspective)

tbohne commented 2 years ago

further information to be added in thesis:

RIEGL scanner produces scans of type PointCloud2
how to receive scans? -> configurable:
- entering ROS system via topic (scanner node publishes scans on topic); potentially reduced (e.g. every 2nd value); if these scans get too big, it gets impractical; frequency too low
- give directory path and checkmark -> indicate that scans should be written to file system (produces .ply files - the scan in some human readable ASCII format)

tbohne commented 2 years ago

Actually two different cases then, right? If it's published on some topic, it's a sensor monitoring case, if it's written to file, it's a data management problem.