Reaction Wheel Saturation Bug

[schaubh]

Dear All,

I noticed a (possible) bug in the reactionWheelSatateEffector.cpp: When reaching the speed saturation, there is no way to desaturate the wheel using torque commands. Using torque in the opposide direction can't be used therefore. I added the following code in the ConfigureRWRequests function:

...
// Speed saturation
if (std::abs(this->ReactionWheelData[RWIter]->Omega) >= this->ReactionWheelData[RWIter]->Omega_max
&& this->ReactionWheelData[RWIter]->Omega_max > 0.0 /* negative Omega_max turns of wheel saturation modeling */
) {
// check if torque would desaturate wheel
if (!((this->ReactionWheelData[RWIter]->Omega > 0.0 && CmdIt->u_cmd < 0.0) || (this->ReactionWheelData[RWIter]->Omega < 0.0 && CmdIt->u_cmd > 0.0))){
CmdIt->u_cmd = 0.0;
}
}
...

```This fixes the mentioned problem. If someone can direct me how to use bitbucket to create a merge request, I might be able to push it to the repo

Tom

[schaubh]

Howdy Tom, thanks for sharing. Let me double check out code. I thought I had that worked out, but maybe not.

[schaubh]

To allow breaking of saturated wheel speeds, I changed the code to read:

// Speed saturation
        if (std::abs(this->ReactionWheelData[RWIter]->Omega) >= this->ReactionWheelData[RWIter]->Omega_max
            && this->ReactionWheelData[RWIter]->Omega_max > 0.0 /* negative Omega_max turns of wheel saturation modeling */
            && this->ReactionWheelData[RWIter]->Omega * CmdIt->u_cmd >= 0.0 // check if torque would accelerate wheel speed beyond Omega_max
            ) {
            CmdIt->u_cmd = 0.0;
        }

The RW unit tests have been adjusted to reflect this behavior. I have pushed the changes to develop.

[schaubh]

Great, looks good to me. Thanks for the quick fix