RW attitude control for Cubesat like missions

[Pietro De Marchi]

Hi there,

I am trying a simulation in Basilisk framework about attitude control with RWs.

In particular I am testing that with a sufficiently light satellite (20kg more or less).

The model I am using in this case is 'Honeywell_HR16', testing a wide range of OmegaV and maxMom for the three RW. I am reporting here instead the settings for the control:

------------------------------------------------------------------------------------

sNavObject = simpleNav.SimpleNav()

sNavObject.ModelTag = "SimpleNavigation"

scSim.AddModelToTask(simTaskName, sNavObject)

# add RW power modules

rwPowerList = []

for c in range(numRW):

powerRW = ReactionWheelPower.ReactionWheelPower()

powerRW.ModelTag = scObject.ModelTag + "RWPower" + str(c)

powerRW.basePowerNeed = 5. # baseline power draw, Watts

powerRW.rwStateInMsg.subscribeTo(rwStateEffector.rwOutMsgs[c])

powerRW.mechToElecEfficiency = 0.5

scSim.AddModelToTask(simTaskName, powerRW)

rwPowerList.append(powerRW)

#

# setup the FSW algorithm tasks

#

# Create a simpleBattery and attach the sources/sinks to it

powerMonitor = simpleBattery.SimpleBattery()

powerMonitor.ModelTag = "powerMonitor"

dd = 0 # try to set battery charge to 0 !!!!!!!!!!!!!!!!!!

powerMonitor.storageCapacity = 300000 # W-s

powerMonitor.storedCharge_Init = powerMonitor.storageCapacity * 0.8 # 20% depletion

scSim.AddModelToTask(simTaskName, powerMonitor)

# connect RW power to the battery module

for c in range(numRW):

powerMonitor.addPowerNodeToModel(rwPowerList[c].nodePowerOutMsg)

# setup guidance module

attGuidanceConfig = hillPoint.hillPointConfig()

attGuidanceWrap = scSim.setModelDataWrap(attGuidanceConfig)

attGuidanceWrap.ModelTag = "hillPoint"

scSim.AddModelToTask(simTaskName, attGuidanceWrap, attGuidanceConfig)

# setup the attitude tracking error evaluation module

attErrorConfig = attTrackingError.attTrackingErrorConfig()

attErrorWrap = scSim.setModelDataWrap(attErrorConfig)

attErrorWrap.ModelTag = "attErrorInertial3D"

scSim.AddModelToTask(simTaskName, attErrorWrap, attErrorConfig)

# setup the MRP Feedback control module

mrpControlConfig = mrpFeedback.mrpFeedbackConfig()

mrpControlWrap = scSim.setModelDataWrap(mrpControlConfig)

mrpControlWrap.ModelTag = "MRP_Feedback"

scSim.AddModelToTask(simTaskName, mrpControlWrap, mrpControlConfig)

------------------------------------------------------------------------------------

So nothing to do, control errors are very high and apparently there is no chance to set up a control with such small mass.

Using higher mass (as reported in all Basilisk examples, around 300kg...) it perfectly works, just tuning RW characteristics.

Anyone else has experienced such a problem and known a possible solution?

I can provide more details about the simulation if needed.

Regards,

PDM

[Hanspeter Schaub]

Did you adjust the MRP feedback control gains to be sensible for this small inertia?  If not, you'll have a crazy fast response that will be unstable with the discrete time steps.