Adaptive Air Density Estimation for Precise Tracking Control and Accurate External Wrench Observation for Flying Robots

Air density changes depending on the local atmosphere and affects the rotor thrust of flying robots. This effect has to be compensated by the flight controller in order to realize precise tracking of a desired trajectory. So far, the influence of the air density has been disregarded or only considered implicitly in the control of flying robots. In this work, a nonlinear adaptive control approach is presented. It explicitly considers the air density in the dynamical model and enables air density estimation and tracking control under changing atmospheric conditions and with added payload. Furthermore, the estimated air density is used to enhance the accuracy of a state-of-the-art external wrench estimator. The adaptive control approach is evaluated in simulations and experiments with a quadrocopter and a coaxial hexacopter.