Using the msis atmoshphere model in a model where the simulation is run in another function.

[loufla]

Hello,

I am currently trying to implement Basilisk in a satellite orbit simulation where the environment class is initialized in an init function, and the simulation in advanced in another function. To do this, most models have to be initialized as self.model to be able to access them later on. This works perfectly, except for the atmosphere calculation for the drag. I tried all atmosphere models (msis, tabular and exponential), and msis is the only one that doesn't work.

I used the atmosphere implementation from the DragDeorbit example to initialize the model.

For msis, the atmosphere density is always 0, I output it using the following function during the simulation: self.get_logger().info(f'ATMO : {str((self.atmo.envOutMsgs[0].read().neutralDensity))}'). For both tabular and exponential atmospheric models, the output is coherent and correct for the current atmospheric density. This is true for any and all orbit altitudes, so it's not a matter of too high an orbit.

Does anyone have any idea of why this would be occurring for only the msis atmosphere simulation?

My guess is that it has to do with the messaging component, but when I try to initialize them as self.swMsgData, the simulation crashes, and the satellite altitude is output as nan.

Would you also know of a method to work with the msis atmospheric model that bypasses the messaging components?

Thank you !

[loufla]

Nevermind, adding the self. to the swMsgList seems to have made the msis atmosphere useable in the environment !