It would be very helpful to create separate logfiles for each test during a flight, which would save a lot of time. This simplifies the analysis.
I propose a function that allows me to start and stop the logging via a specific RC channel. As a result, getting a for each flight a number of small logs. Each of those belongs to a test sequence.
Jakob
1 Like
I can see how that could be helpful. However, at the same time we usually want to make sure we have a log of the whole flight, so we don’t miss something in case a crash happens, etc.
As a workaround I suggest that you increase a parameter (e.g. some parameter that you are not using, or that you add yourself) every time you begin a new test. Parameter changes show up in the log, so that way you have a “time marker” of where a test begins.
Julian, thank you for your suggestion.
the advantage introducing a additional SDLOG_MODE as far I see would be:
- In the Phase of testing, I am interested to look at specific behavior of the UAV by changing certain parameters.
Adding this function would be mainly a time saver and make testing more efficient. This allows to share small log files with other colleagues and the corresponding description of what I have tried to achieve or test. - I do not need a second person which assist me during the tests. Right I need to go for and backward between the PC with QGC and the Radio control and the notepad if I am alone in the field.
- A alternative could be if we could just extract the logging data of a given time window.
jakob
Yes we can extend SDLOG_MODE to support that.
Are you in a position to do a pull request with these changes? That would be greatly appreciated.
@bkueng This is out fo my capacity. But I am pretty good in testing. Should this request moved to github under issues aa a request?
You can do that, but it will likely be me implementing it, so I already have it on my list.
Hi @Jakob_Strebel1, I implemented this in https://github.com/PX4/Firmware/pull/12100, can you give this a test?
1 Like