Zarudzki, Mateusz (Cranfield University), Shin, Hyo-Sang (Cranfield University), Lee, Chang-Hun (Cranfield University)

An Image Based Visual Servoing Approach for Multi-Target Tracking Using an Quad-Tilt Rotor UAV

Scheduled for presentation during the "UAS Applications - IV" (ThA4), Thursday, June 15, 2017, 10:40−11:00, Lummus Island

2017 International Conference on Unmanned Aircraft Systems, June 13-16, 2017, Miami Marriott Biscayne Bay, Miami, FL,

This information is tentative and subject to change. Compiled on April 20, 2024

Abstract

This paper proposes a guidance algorithm of a UAV for multiple targets tracking, considering physical constraints of the platform and its sensor. The on-board sensor used for target tracking is a camera. To relax the need of a gimbal system and provide flexibility in utilisation of the vision sensor, a new rotorcraft type of UAV is developed. The main focus in designing tracking guidance is to develop an image based visual servoing approach, appropriate to the newly developed platform. This paper shows that the tracking guidance can be significantly simplified with the new type of UAV developed. A complementary to the guidance system a control system for the unconventional UAV is also presented. Both control and guidance algorithms are based on the PID control techniques. The performance of the proposed tracking guidance along with the controller designed is validated by numerical simulations and flight tests.