Autonomous Drone for Human/Object Following using PID Control and Real Time Image Recognition

This article outlines the design and implementation of a vision-based autonomous human tracking quadrotor using an embedded computing platform. The proposed system combines real-time object detection and image-based error computation to provide high-level motion commands to track the target. The Raspberry Pi 5 acted as a companion computer to process computer vision algorithms, and the MicoAir F405 Mini flight controller provided low-level stabilization using internal PID controllers.   To execute this design, the system architecture was designed to separate the perception and control aspects of the system to enable efficient execution of tasks on resource-constrained platforms. The MAVLink protocol provided efficient communication between the onboard computer and flight controllers to execute tasks seamlessly.   To verify the effectiveness of the proposed system, simulations using a Betaflight-based FPV simulation environment showed that the proposed system is capable of sustaining continuous tracking of the target object, which in this case is a human. The results of this experiment show that the proposed system is capable of executing tasks in real-time to detect and respond to the target in dynamic tracking scenarios.   In conclusion, the proposed system provides a cost-effective and efficient solution to implement a vision-based tracking system using a UAV, which has numerous applications in various fields.

[1] J. Redmon and A. Farhadi, “YOLOv3: An incremental improvement,” arXiv preprint arXiv:1804.02767, 2018. [2] D. Mellinger and V. Kumar, “Minimum snap trajectory generation and control for quadrotors,” in Proc. IEEE Int. Conf. Robotics and Automation (ICRA), 2011, pp. 2520–2525. [3] M. J. Mahmood, S. A. Wani, and H. N. Sheikh, “Design and implementation of a PID controller for quadcopter trajectory tracking,” in Proc. IEEE Int. Conf. Intelligent Computing, Instrumentation and Control Technologies, 2019, pp. 1491–1496. [4] P. Corke, R. Mahony, and V. Kumar, “Multirotor aerial vehicles: Modeling, estimation, and control of quadrotor,” IEEE Robotics & Automation Magazine, vol. 19, no. 3, pp. 20–32, 2012. [5] S. Bouabdallah, A. Noth, and R. Siegwart, “PID vs LQ control techniques applied to an indoor micro quadrotor,” in Proc. IEEE/RSJ Int. Conf. Intelligent Robots and Systems (IROS), 2004, pp. 2451–2456. [6] T. Bresciani, “Modelling, identification and control of a quadrotor helicopter,” M.S. thesis, Lund University, 2008. [7] A. Kamath and P. Thomas, “Vision-based autonomous human tracking and following robot using Raspberry Pi,” in Proc. IEEE Int. Conf. Communication and Signal Processing, 2018, pp. 1337–1341. [8] M. Mueller, M. Hehn, and R. D’Andrea, “A computationally efficient motion primitive for quadrocopter trajectory generation,” IEEE Transactions on Robotics, vol. 31, no. 6, pp. 1294–1310, 2015. [9] S. Shah et al., “AirSim: High-fidelity visual and physical simulation for autonomous vehicles,” in Field and Service Robotics, 2018. [10] N. Koenig and A. Howard, “Design and use paradigms for Gazebo, an open-source multi-robot simulator,” in Proc. IEEE/RSJ IROS, 2004, pp. 2149–2154.