Last week, we finished interfacing Crazyflie 2.0 with ROS, meaning we are able to get sensor data from Crazyflie 2.0 to our laptop by monitoring ROS topic IMU for Crazyflie. What we can get from the sensor is the linear acceleration, angular acceleration and orientation. Furthermore, the pressure sensor on board can be used for estimating z position. However, we need all x,y and z position data if we want to implement the controller for the case of quadcopter losing Propellers. I took a look at another research from Wolfgang Honig, the author of the Crazyflie ROS package. The link can be found here. They use motion capturing system data along with onboard IMU data to generate an accurate state estimation. The problem we are facing now is that we do not want to use any off-board sensor, but the error of state estimation solely using IMU is relatively large. Thus, next week I will focus on how to do more accurate state estimation or implement more sensor on Crazyflie.