we should add the ability to log sensory data (and other information we find useful) to a file so that we can go back and understand exactly why the robot behaved the way it did in a previous run. This will allow us to better understand what is happening and thereby to create more robust solutions based on what we are observing. For example, we can only guess, at this point, why the block was dumped prematurely during teleop this past weekend.
There are existing frameworks that allow for this. We should either plan to leverage one of them or implement a similar framework for ourselves. Key abilities include:
-ability to log any event at a specific logging level (severities)
-ability to turn on logging to a specific level (so that only events at that severity or higher are logged)
-ability to limit the size of the log file
This is a critical component to allow us to understand how our software, and hardware, are functioning.
we should add the ability to log sensory data (and other information we find useful) to a file so that we can go back and understand exactly why the robot behaved the way it did in a previous run. This will allow us to better understand what is happening and thereby to create more robust solutions based on what we are observing. For example, we can only guess, at this point, why the block was dumped prematurely during teleop this past weekend.
There are existing frameworks that allow for this. We should either plan to leverage one of them or implement a similar framework for ourselves. Key abilities include:
-ability to log any event at a specific logging level (severities) -ability to turn on logging to a specific level (so that only events at that severity or higher are logged) -ability to limit the size of the log file
This is a critical component to allow us to understand how our software, and hardware, are functioning.