Published at : 06 Oct 2021
Volume : IJtech
Vol 12, No 4 (2021)
DOI : https://doi.org/10.14716/ijtech.v12i4.3928
M Ary Heryanto | Department of Electrical Engineering, Faculty of Engineering, Universitas Dian Nuswantoro, Semarang 50131, Indonesia |
Benyamin Kusumoputro | Department of Electrical Engineering, Faculty of Engineering, Universitas Indonesia, Kampus UI Depok, Depok 16424, Indonesia |
In
this study, a neural network–based direct inverse control (DIC) approach was
developed and simulated to control various maneuvers of an unmanned aerial
vehicle (UAV) quadcopter. The aim was to propose an inner loop control
algorithm for UAV quadcopter maneuvers using a neural network–based DIC system.
The appropriate connection weights of neurons in the controller were determined
through a backpropagation learning algorithm using real quadcopter maneuver
flight data. The neural network–based DIC was trained and then tested using a
trajectory dataset different from the training dataset. The experimental
results showed that the neural network–based DIC could follow the maneuvers of
the testing trajectory dataset with excellent performance, as indicated by an
overall mean squared error (MSE) of 1.461 and attitude MSEs of 3.104 for roll,
0.889 for pitch, 1.834 for yaw and 0.018 for altitude. These results indicate
that the proposed artificial neural network–based DIC can be used to control
the attitude and altitude of the quadcopter during maneuvers.
Backpropagation; Direct inverse control; Maneuver; Neural Network; Quadcopter
Unmanned aerial vehicle (UAV) research has grown rapidly in recent years
in both the civilian and military sectors (Duan
et al., 2010; Gandhi and Ghosal, 2018;
Krishnan et al., 2018; Nenni et al., 2020). As multirotor
drones, UAV quadcopters have several advantages, including the ability to
perform vertical takeoff and landing, high maneuverability, and simple
mechanical structures. Quadcopters are under-actuated, with strong coupling and
highly nonlinear systems (Gheorghita
et al., 2015; Wang et al., 2016); therefore,
controlling them is a considerable challenge.
UAV quadcopters are increasingly attracting the attention of
researchers, and developing an autonomous quadcopter control method is of
central importance. Numerous studies have been conducted to address the issue
of quadcopter control. Proportional integral
derivative (PID), LQR, backstepping, and sliding mode control are the most widely
used methods (Argentim et
al., 2013;
Tripathi et al., 2015; Wang et al., 2016; Najm and Ibraheem, 2019; Nguyen et al., 2019).
Moreover,
neural network (NN)-based control systems have been developed with decent
performance (Anuradha
et al., 2009; Xianglei et al., 2011;
Suprapto et al., 2017; Muliadi &
Kusumoputro, 2018; Yuning et al., 2019; Mahadika et al., 2020).
Like other aerial vehicles, quadcopters have abilities of six degrees of
freedom—pitch, roll, yaw, and x, y, z—but only four basic
movements are generated directly—namely, roll, pitch, yaw, and thrust (Wang et
al., 2016). Roll and pitch are related to changes in
the quadcopter’s position on the x and y axes, and yaw is related
to rotational motion on the z axis. Roll, pitch, and yaw produce
attitude movement, while thrust is related to changes in altitude movement.
Since the quadcopter movement depends entirely on these four basic motions, it
is extremely important to maintain stable attitude (pitch, roll, and yaw) and
altitude control conditions, especially during maneuvers.
Our research is based on a cross configuration of a quadcopter constructed in the Computational Intelligence and Intelligent Systems Laboratory, Universitas Indonesia. We previously also developed an NN-DIC scheme as a control system and optimized it to easily control the hovering state of the quadcopter (Heryanto et al., 2015). Our results showed that the optimized NN-DIC system improved the controller’s performance, as indicated by a faster settling time compared with that of a non-optimized NN-DIC method. We also confirmed in subsequent experiments (Heryanto et al., 2017) that the NN-DIC system can maintain the quadcopter’s attitude and altitude in a simple UAV simulation flight with reasonable errors. Based on these results, in this study, we aimed to investigate the response of attitude and altitude control using our developed NN-DIC method on a maneuvering trajectory flight of a quadcopter.
The rest of this paper is organized as follows. Section 2 explains the structure and dynamics of the quadcopter. Section 3 describes the NN-DIC strategy and the data acquisition method. Section 4 presents the simulation results of the experiment. Section 5 concludes the paper.
The
purpose of this study was to continue our research on exploring the
capabilities of NN-DIC applied to a quadcopter under maneuver conditions. The
NN-DIC system succeeded in maintaining the quadcopter’s trajectory. The
experimental results also showed that the quadcopter could follow the
trajectory altitude with a very small MSE (0.018), as well as some inaccuracy
in attitude. These results indicate that the NN-DIC system can control a
quadcopter under maneuver conditions, such as clockwise/counterclockwise
up-and-down helix movements. A full quadcopter control system is being
developed and will be presented in the near future.
The
authors would like to express their gratitude to the Ministry of Research,
Technology, and Higher Education of Indonesia for supporting this work through
a Universitas Indonesia Scholarship Research Grant number 1251.38/E4.4/2012. We
also thank Universitas Dian Nuswantoro.
Alkamachi, A., Erçelebi, E., 2017.
Modelling and Genetic Algorithm Based-PID Control of H-Shaped Racing Quadcopter.
Arabian Journal for Science and Engineering, Volume 42(7), pp. 2777–2786
Anuradha, D.B., Reddy, G.P., Murthy,
J.S.N., 2009. Direct Inverse Neural Network Control of a Continuous Stirred
Tank Reactor (CSTR). In: Proceedings of the International
MultiConference of Engineers and Computer Scientists, Volume 3
Argentim, L.M. Rezende, W.C., Santos,
P.E., Aguiar, R.A., 2013. PID, LQR and LQR-PID on a Quadcopter Platform. In:
2013 International Conference on Informatics, Electronics and Vision (ICIEV),
pp. 1–6
Bresciani, T., 2008. Modelling,
Identification and Control of a Quadrotor Helicopter. Master's Thesis, Graduate
Program, Lund University, Sweden
Duan, H., Shao, S., Su, B., Zhang, L.,
2010. New Development Thoughts on the Bio-inspired Intelligence Based Control
for Unmanned Combat Aerial Vehicle. Science China Technological Sciences,
Volume 53(8), pp. 2025–2031
Gandhi, D.A., Ghosal, M., 2018. Novel
Low Cost Quadcopter for Surveillance Application. In: 2018 International
Conference on Inventive Research in Computing Applications (ICIRCA), pp.
412–414
Gheorghita, D., Vîntu, I., Mirea, L.,
Br?escu, C., 2015. Quadcopter Control System. In: 2015 19th
International Conference on System Theory, Control and Computing, ICSTCC 2015,
pp. 421–426
Heryanto, M.A., Suprijono, H.,
Suprapto, B.Y., Kusumoputro, B., 2017. Attitude and Altitude Control of a
Quadcopter using Neural Network Based Direct Inverse Control Scheme. Advanced
Science Letters, Volume 23(5), pp. 4060–4064
Heryanto, M.A., Wahab, W.,
Kusumoputro, B., 2015. Optimization of a Neural Network Based Direct Inverse
Control for Controlling a Quadrotor Unmanned Aerial Vehicle. MATEC Web of
Conferences, Volume 34, pp. 4–7
Krishnan, R.A., Jisha, V.R.,
Gokulnath, K., 2018. Path Planning of an Autonomous Quadcopter Based Delivery
System. In: 2018 International Conference on Emerging Trends and
Innovations in Engineering and Technological Research (ICETIETR), pp. 1–5
Mahadika, P., Subiantoro, A.,
Kusumoputro, B., 2020. Neural Network Predictive Control Approach Design for Adaptive
Cruise Control. International Journal of Technology, Volume 11(7), pp.
1451–1462
Muliadi, J., Kusumoputro, B., 2018.
Neural Network Control System of UAV Altitude Dynamics and its Comparison with
the PID Control System. Journal of Advanced Transportation, Volume 2018,
Article ID 3823201
Najm, A.A., Ibraheem K.I., Ahmad T.A.,
Amjad J.H., 2020. Genetic Optimization-Based Consensus Control of Multi-Agent
6-DoF UAV System. Sensors, Volume 20(12), pp. 1–31
Najm, A.A., Ibraheem, I.K., 2019.
Nonlinear PID Controller Design for a 6-DOF UAV Quadrotor System. Engineering
Science and Technology, an International Journal, Volume 22(4), pp.
1087–1097
Nenni, M.E., Di Pasquale, V., Miranda,
S., Riemma, S., 2020. Development of a Drone-Supported Emergency Medical
Service. International Journal of Technology, Volume 11(4), pp. 656–666
Nguyen, A.T., Xuan-Mung, N., Hong,
S.-K., 2019. Quadcopter Adaptive Trajectory Tracking Control: A New Approach
via Backstepping Technique. Applied Sciences, Volume 9(18), pp. 2–17
Suprapto, B.Y., Mustaqim, A., Wahab,
W., Kusumoputro, B., 2017. Modified Elman Recurrent Neural Network for
Attitude and Altitude Control of Heavy-Lift Hexacopter. In: 2017 15th
International Conference on Quality in Research (QiR): International Symposium
on Electrical and Computer Engineering, pp. 309–314
Tripathi, V.K., Behera, L., Verma, N.,
2015. Design of Sliding Mode and Backstepping Controllers for a Quadcopter. In:
2015 39th National Systems Conference, pp. 1–6
Wang, P., Man, Z., Cao, Z., Zheng, J.,
and Zhao, Y., 2016. Dynamics Modelling and Linear Control of Quadcopter. In:
2016 International Conference on Advanced Mechatronic Systems (ICAMechS),
pp. 498–503
Xianglei, D., Shuguang, Z., Lvchang,
H., and Rong, H., 2011. The Neural Network Direct Inverse Control of Four-Wheel
Steering System. In: 2011 Third International Conference on Measuring
Technology and Mechatronics Automation, pp. 865–869
Yuning,
J., Rasool, M.A.U., Bo, Q., Farid, G., and Chaudary, S.T., 2019. An Adaptive
Neural Network State Estimator for Quadrotor Unmanned Air Vehicle. In: (IJACSA)
International Journal of Advanced Computer Science and Applications, Volume
10(2), pp. 316–321