Optical Flow and IMU Fusion for Drone Horizontal Velocity Control
paper Menu
This paper presents a vision-aided, horizontal velocity control system for unmanned aerial vehicles (UAVs). Angular velocity data provided by an Inertial Measurement Unit (IMU) is combined with pixel displacement data given by an optical flow sensor to produce a measure of horizontal velocity in two dimensions. Robust velocity control is achieved without reliance on the Global Positioning System (GPS). To validate the proposed approach, a series of dynamic simulations and hardware experiments were conducted. The hardware tests were designed to characterize the output of the optical flow sensor as a function of horizontal velocity and altitude. This data was then used to formulate an optical flow sensor mathematical model that, to our knowledge, is new and has not appeared in any prior optical flow sensor related literature. This model is shown to produce output data that is consistent with the experimentally measured sensor data gathered under equivalent conditions. In addition, this model has been used to create the optical flow sensor model employed when simulation testing a UAV’s controlled dynamic motion. Simulation results demonstrate that this method successfully achieves accurate and reliable control of a UAV’s velocity in the horizontal plane.
