Emilio Corigliano authored 8 months ago and Nicolò Caruso committed 7 months ago

Method to convert quaternions to euler angles assuming that the roll stays close to 0. This applies to systems with 2 DoFs (yaw and pitch).

[SkyQuaternion] Fixed warning on type comparison

Federico Lolli authored 8 months ago and Nicolò Caruso committed 7 months ago

[Follower] Add minimize rotationa as suggested in previous obsw pull req.

See [this](avn/swd/obsw!34)
for more context

[logger] Updated logger

[ARP Follower] Yaw gain now 0.1

The yaw gain gain now is no more 1 but fixed to 0.1

[ARP] Added timestamp to AntennaAngles for log

Added timestamp to the AntennaAngles logged. Changed accordingly the constructors.

[Fix] Follower algorithm limitation issue due to proportional on yaw

The 0.1* product to have a proportional on the yaw was creating issues with the angle limiting function.

[ARP] Added nrPropagations for log in AntennaAngle now logged by SMA

Added nrPropagations to AntennaAngles
Moved the log of it from the Follower to SMA

[ARP] Follower initRocketCoordinates now is NASOrigin

Changed semantic of initialRocketCoordinates to NASOrigin since now it is set always as we have a NASOrigin.

[ARP] WIP fix for online initial coordinates support

Add support for initial coordinates online setting

[ARP] Follower Mutexes for antenna and NAS origin's coordinates

Now getter states to protect such variables with mutexes

Federico Lolli authored 8 months ago and Nicolò Caruso committed 7 months ago

Created the algorithm of a basic propagator which linearly propagates the last available rocket position by means of the rocket NAS velocity.
A catch test has been implemented to test the algorithm in a static environment, a dynamic one and with different update frequencies.

[Propagator] Now the constructor parameter must be passed in ms

The previous setting of 6 Mhz was not enough to fix the SD logging issues Lyra boards were experiencing.

Before some sigmas were actually variances, now they are all used as standard deviations

Co-Authored-By: Davide Basso <davide.basso@skywarder.eu>

using R_acc instead of R_mag

Co-Authored-By: Davide Basso <davide.basso@skywarder.eu>

Co-Authored-By: Davide Basso <davide.basso@skywarder.eu>

Co-Authored-By: Davide Basso <davide.basso@skywarder.eu>

Missing a ',' before the secondary type

[AnalogSensors] Improved Honeywell driver legibility, made sensor offset atomic and added offset getter

Emilio Corigliano authored 8 months ago and Davide Mor committed 8 months ago

- Doing computeForce before correctBaro
- Checking if S matrix (scalar) is invertible

Emilio Corigliano authored 9 months ago and Davide Mor committed 8 months ago

- Initialized mass parameter
- Fixed apogee estimation massive formula
- Simplified apogee estimation formula splitting it
- Added some documentation

[HIL] Moved HIL framework from OBSW to boardcore, enhanced with HILSensor and hillificator, turned all sensor drivers HIL compliant

- HIL is now the single point of contact of the whole framework (HILTransceiver and HILPhasesManager); Its run method implements the periodic set of the actuators status to be sent back to the simulator.
- HILPhasesManager is in charge of changing HIL flags, registering callbacks for specific flags, registering outcomes for important flight phases.
- HILTransceiver is in charge of handling the HIL communication through the serial port
- Implemented a move constructor for Sensor;
- To Mock a sensor and use in the HIL framework it's not needed a specific HILSensors for each data type but it is mocked by means of the move constructor thanks to the hillificator helper function. An HILSensor behaves just like the real sensor (real sensor sampling can be enabled to use also the driver's code) but exposing the samples coming from the simulator.
- A test has been implemented to give a usage example of the framework;
- All sensors have been checked so that they aren't final and they have sampleImpl protected;