Dynamic Image-Based Visual Servoing for Quadrotor to Track a Planar Target with Unknown Motion

This paper presents an image-based visual servoing (IBVS) technique for a quadrotor to track a moving planar target having unknown linear and angular velocities. Image moment-based features are derived after transforming the optical measurements to a virtual image plane parallel to the planar target. The kinematic controller is designed incorporating the linear and angular target velocity estimator blocks. The theoretical analysis demonstrates that the errors converge exponentially to an ultimate bound for time-varying target motion and exponentially to zero if the target motion is purely translational with constant velocity. Unlike past literature, the proposed IBVS scheme can handle unknown arbitrary translational and angular (along the z-axis) motions of the target in a unified and model-free manner. Simulation results using circular and Liassajous target motion are provided to validate the efficacy of the proposed scheme.