Controlling Crazyflie through RPMs (PWM)

[piratax007]

Hello, I have successfully trained a policy for setpoint tracking using RPM as action space. Now I'm trying to deploy it to a real Crazyflie using a ROS2 package (Crazywarm and Vicon system). As you refer to in #118, exists an empirical study of the relation between PWM and RPMs for Crazyflie (https://www.bitcraze.io/documentation/repository/crazyflie-firmware/master/functional-areas/pwm-to-thrust/). My question, is that using the data from that study and comparing with the transformation used by GymPybullet I found high discrepancies as is shown in the plot. The black line corresponds with the linear transformation used by GymPybyllet.

Could you help me to understand this and how can affect the transferability of the trained policy to a real drone?

Thank you so much

[JacopoPan]

The relation between RPMs and thrust in gym-pybullet-drones is modeled as quadratic:

gym-pybullet-drones/gym_pybullet_drones/envs/BaseAviary.py

Lines 693 to 694 in 3d7b12e

	forces = np.array(rpm**2)*self.KF
	torques = np.array(rpm**2)*self.KM

There assumption of linearly between RPMs and PWM (that don't fully hold away from a hover range) in DSLPIDControl.py and CFAviary.py but those shouldn't come into play if you purely trained RL for RPMs in a class derived from BaseRLAviary.py

[piratax007]

I understand that training using RPMs as actions space the conversion to PWM doesn't play a role. Nevertheless, transferring a policy trained to produce RPMs to a real Crazyflie, requires translating the RPM to PWM because the lowest control signal received by the Crazyflie firmware is PWM.

BaseRLAviary translates the policy outputs to RPM through

target = action[k, :]
if self.ACT_TYPE == ActionType.RPM:
    rpm[k,:] = np.array(self.HOVER_RPM * (1+0.05*target))

Then, to obtain the PWM should be applied:

$PWM = \dfrac{RPM - 4070.3}{0.2685}$

Is it correct?

[JacopoPan]

In BaseAviary, the relationship between RPMs and thrust is a simple quadratic one using coefficient kf="3.16e-10" which is taken from the system identification resources lined in the .urdf

<!-- links>
      <carlos url="https://arxiv.org/pdf/1608.05786.pdf" />
      <julian url="https://www.research-collection.ethz.ch/handle/20.500.11850/214143" />
      <mit url="http://groups.csail.mit.edu/robotics-center/public_papers/Landry15.pdf" />
</links -->

If you want to go by the system identification in https://www.bitcraze.io/documentation/repository/crazyflie-firmware/master/functional-areas/pwm-to-thrust/ you should modify

gym-pybullet-drones/gym_pybullet_drones/envs/BaseAviary.py

Line 693 in 3d7b12e

forces = np.array(rpm**2)*self.KF

so that thrust = 1.0942e-07rpm^2 - 2.1059e-04rpm + 1.5417e-01 instead
and then consider that thrust = 0.409e-03rpm^2 - 140.5e-03rpm - 0.099 when you want to derive the corresponding PWM command