Attitude Control of Quad-copter using Deterministic Policy Gradient Algorithms (DPGA)

No Thumbnail Available

Date

2019

Authors

Journal Title

Journal ISSN

Volume Title

Publisher

Ieee

Research Projects

Organizational Units

Thumbnail Image
Organizational Unit
Mechatronics Engineering
(2002)
The Atılım University Department of Mechatronics Engineering started its operation in 2002 as the Education Program in Mechatronics Engineering holding a “department” status in Turkey. In addition, it is the first and the only institution for mechatronic engineering education to obtain a MÜDEK (Association for Evaluation and Accreditation of Engineering Programs) accreditation for a duration of 5 years. Mechatronics engineering is a discipline of engineering that combines mechanical, electrical and electronic engineering and software technologies on a machine or a product. These features place the field on a pedestal in today’s industry. The education at our department is also backed by substantial laboratory opportunities. Our students create interesting products of their skills and creativity for their dissertation projects. Should they wish to do so, our students may also proceed with a double-major program in the fields of Computer Engineering, Electrical - Electronics Engineering, Industrial Engineering, or Mechanical, Automotive or Software Engineering. Upon their demands, the Department of Mechatronic Engineering also offers a “Cooperative Education” program implemented in coordination with industrial institutions. Students receiving a portion of their training at industrial institutions and prepare for professional life under this program

Journal Issue

Abstract

In aerial robotics, intelligent control has been a buzz for the past few years. Extensive research efforts can be witnessed to produce control algorithms for stable flight operation of aerial robots using machine learning. Supervised learning has the tendency but training an agent using supervised learning can be a tedious task. Moreover, the data gathering could be expensive and always prone to inaccuracies due to parametric variations and system dynamics. An alternative approach is to ensure the stability of the aerial robots with the help of Deep Re-inforcement Learning (DRL). This paper deals with the intelligent control of quad-copter using deterministic policy gradient algorithms. In this research, state of the art Deep Deterministic Policy Gradient (DDPG) and Distributed Distributional Deep Deterministic Policy Gradient (D4PG) algorithms are employed for attitude control of quad-copter. An open source simulation environment GymFC is used for training of quad-copter. The results for comparative analysis of DDPG & D4PG algorithms are also presented, highlighting the attitude control performance.

Description

Khan, Muhammad/0000-0002-9195-3477;

Keywords

Deep reinforcement learning, DDPG, D4PG, Quad-copter control, GymFC

Turkish CoHE Thesis Center URL

Citation

5

WoS Q

Scopus Q

Source

2nd International Conference on Communication, Computing and Digital Systems (C-CODE) -- MAR 06-07, 2019 -- Bahria Univ, Islamabad, PAKISTAN

Volume

Issue

Start Page

149

End Page

153

URI

https://doi.org/10.1109/c-code.2019.8681003
 https://hdl.handle.net/20.500.14411/2823

Collections

WOS