Frequently Asked Questions - Carleton-SRCL/SPOT Wiki

I'm having problems with the Raspberry Pi, what can I do?

_* Your best bet is to try connecting to the Pi remotely. If you can achieve a connection, then the Pi is still working and the problem lies elsewhere. Check that all the pre-compiled functions are still on the Pi (such as "Float.elf"). If that all checks out, there may be a hardware problem. _

How do I record high quality videos of my experiments?

* We have four security cameras in the lab, all of which record the last 48 hours of footage. You can use the application called QC Viewer on the laboratory ground station computer to export the footage per the instructions in the wiki.

I compiled my experiment and it worked just fine the first time. However, I made some changes and recompiled, and it seems that the changes are not present?

* This is a MATLAB bug (see issue #2, which is now closed), but the most recent version of the main repository should have fixed this problem. As a workaround, you can try and manually copy the .elf file created in your experiment folder on the ground station PC to the following directory on the Raspberry Pi: /home/pi/ExperimentSoftware/MATLAB_ws/R2020a/. You should be able to run the file using the runModel function.

Help! My experiment worked yesterday, but today one of the platforms is not stable and I did not change anything.

* This happens occasionally, I've seen many experiments where it works one day and then fails the next day. The most likely culprits are 1) the motion tracking system calibration has drifted - refer to the calibration guide - and 2) there has been a change in the mass or inertia. Unless you have added signficant weight to a platform, try calibrating the motion capture system. If that fails, change the gains for the PID controller in the Run_Initializer.m script.

I just finished an experiment, but clicking on “Save Data” does not extract any data… Furthermore, when I manually check on the Raspbbery pi, there is no saved data! What gives?

* The most likely culprit is that you did not properly terminate the experiment before you tried to save the data. While the experiment is running, data is temporarily saved in a buffer. Once the experiment has been stopped (by clicking “Stop Experiment” in the GUI, or by running the stopModel function), the data is saved to a .mat file. If you try to save the data before terminating the experiment, the data will be corrupted and you won’t be able to open it, or it will simply cease to exist.

I just compared the experiment and simulation data, and there seems to be a significant error. What could be going on?

* There are a few possible causes. First, note that the experiment will not start until the LEDs have successfully connected. Once the LEDs are connected, this becomes t = 0 seconds in the experiment. Thus, if it take 5 seconds to connect, there will be a 5 second delay between simulated and experiment data. To fix this, you can post-process the data so that it is time aligned properly (either using the timeseries MATLAB function or by simply adding samples). Another culprit could be the mass properties of the platform are wrong, which would cause errors between the simulation and experiment. Check that the mass properties are correct. Lastly, check that the positions of the LEDs are correct relative to the center of mass, and check that the PhaseSpace calibration has been done recently.

I grabbed the latest code from the GitHub but it's not working correctly. What do I do?

* Your best bet is to raise the issue in GitHub itself, as it will notify me and I will address the problem. When opening an issue, please include details on the steps required to replicate the problem. Note: Due to the ongoing COVID-19 pandemic, code validation is challenging so do not expect a fast turnaround. In the interim, please revert back to a previous software version.

I have a suggestion for the software!

* Great! I am always looking to add new features. Please contact me through GitHub by adding your idea to the "Issues" tab and tagging it with either "Software Enhancement" or "Hardware Enhancement".

When I try to float a platform/run thruster check, nothing happens. What did I do wrong?

* When starting from a new PC, sometimes the default programs must be reinstalled. Please follow the setup procedure for deploying basic SPOT functions.