Скачать или смотреть Perception-Aware Image-Based Visual Servoing of Aggressive Quadrotor UAVs (AIM 2023)

The maintenance of visual features within the sensor field-of-view (FOV) during aggressive maneuvers poses a significant challenge for underactuated aerial vehicles like quadrotors. However, existing IBVS methods rely on strict assumptions about target visibility or impose excessive constraints on the quadrotor's agility to meet this requirement. Additionally, the effectiveness of the visibility constraint formulated in prior works remains unverified in aggressive flight simulations or experiments. To address these issues, we present a robust image-based visual servo control (IBVS) scheme for quadrotors to perform aggressive maneuvers while ensuring target visibility. Based on the nonlinear model predictive control (NMPC) framework, we first propose a novel underactuation compensation scheme to eliminate the need for a virtual camera frame and allow us to formulate the true target visibility constraint. We then introduce a quaternion-based representation of spherical visual features to remove manifold constraints and derive its corresponding image kinematics. We validate the proposed method through three challenging visual servoing tasks: fast autonomous landing, aggressive long-distance flight, and fast-moving-object tracking. Extensive simulation and experimental results demonstrate the effectiveness of our visibility constraint under large robot rotations and underscore its importance in supporting robust and aggressive flights.

#FSCLab #UTIAS #UofT