[ANSWERED] crazyflie 2 loco positioning

[johnny]

Hi,

I would like to know whether the loco positioning is capable of knowning whether the crazyflie 2 is rotating or not? Do you think it can replace expensive localisation systems such as Vicon cameras for estimating its pose? Of course the accuracy will not be the same...

john

[arnaud]

Hi,

The Loco Positioning UWB radio system is not capable of measuring any angle of the crazyflie pose.

Crazyflie 2.0 is using a Kalman filter to estimate its pose, including all angles, using the UWB radio system and the dynamic of the system using accelerometer and gyroscope measurements. This means that the kalman filter is able to know and output the Crazyflie rotation by tracking its modements. One condition for that to work well is that the Crazyflie has to know its initial orientation when starting up (currently we start it up facing X).

As for replacing Vicon cameras, it all depends of your needs. The LPS is going to be less accurate and currently we still consider flying more than one Crazyflie experimental. On the plus side the LPS allows to have fully autonomous crazyflies: the crazyflie estimate its position using direct measurement so there is no need to send the position to each Crazyflie in real time.

[johnny]

Hi Arnaud,

I understand and thank you for this information.

I will go for an external localisation system then, incorporating high speed cameras to get accurate orientations and positions.

Regards,
John