Bitcraze Forums

Using the ROS robotic operating system and the OPTITRACK infrared motion capture system for autonomous hovering of the aircraft, it was found that the aircraft was significantly offset towards the X-axis or Y-axis side and hit the laboratory's protective net. We used a number of aircraft to perform repeated experiments and found that only one aircraft was able to hover stably. The only difference between this aircraft is that the position of the motion capture system is different. We have modified the PID parameters, and the actual impact is small, and the take-off time has been extended. At the same time, we also ruled out the influence of the motive of the positioning system. Our team considered whether it was the effect of different centroid positions at the location of the markers. Need your help. Thank you

My initial though is that it could be that the centroid position is different between the Crazyflies as you suggest. I guess the centroid position is calculated to be somewhere in the middle of the tracking markers? How inaccurate are we talking about?