Target tracking for quadrotors based on deep reinforcement learning

Gao, Yan ORCID: https://orcid.org/0000-0001-5890-9717, Lin, Feiqiang, Wei, Changyun, Grech, Raphael and Ji, Ze ORCID: https://orcid.org/0000-0002-8968-9902 2024. Target tracking for quadrotors based on deep reinforcement learning. Presented at: 30th IEEE International Conference on Mechatronics and Machine Vision in Practice, Leeds, UK, 3-5 October 2024. 30th International Conference on Mechatronics and Machine Vision in Practice (M2VIP). IEEE, 10.1109/M2VIP62491.2024.10746058

PDF - Accepted Post-Print Version Download (947kB)

Abstract

In this paper, we propose a deep reinforcement learning-based method for quadrotors to learn depth-based tracking policies autonomously. To this end, we present a novel reward function that guides the quadrotor to follow the target, avoid collisions, and keep the target close to the centre of the onboard camera’s field of view without occlusions. In addition, to improve learning efficiency, we suggest using a teacher-student learning strategy. Specifically, we first train a state-based teacher policy encoding low-dimensional obstacle information, which then guides the vision-based student policy during training. Moreover, we introduce a variant of the Proximal Policy Optimisation algorithm based on the importance sampling algorithm. It facilitates the teacher-student learning process and enables the vision-based agent to escape local minima. The experimental results have demonstrated the satisfactory performance of our proposed method.

Item Type:	Conference or Workshop Item (Paper)
Date Type:	Published Online
Status:	Published
Schools:	Engineering
Publisher:	IEEE
ISBN:	979-8-3503-9191-6
Date of First Compliant Deposit:	16 September 2024
Date of Acceptance:	30 July 2024
Last Modified:	22 Nov 2024 12:15
URI:	https://orca.cardiff.ac.uk/id/eprint/172149

Actions (repository staff only)

Edit Item