Tracking of Mobile Objects with an UAV and a DNN Controller

Abstract

The success of the Ingenuity mission on Mars underscores the growing need for autonomous navigation technologies in UAVs. However, detecting and tracking moving objects with unknown dynamics remains a challenge in planetary exploration. Current optimal control algorithms outperform classical controllers but struggle to generate control signals within the required operating time, leading to high computational costs. We propose a Deep Neural Network (DNN) architecture pre-trained with a Model Predictive Control (MPC) for horizontal motion control, coupled with a Proportional-Integral-Derivative (PID) controller for altitude and orientation of a UAV for mobile target tracking. This approach reduces computational costs, significantly improves the speed of control signal generation, and maintains performance similar to MPC. Control commands are computed from the estimated trajectory using visual information and UAV states. Testing is conducted in Gazebo with the Parrot Bebop 2.0 and Husky Robot.