# Design of Combined Neural Network and Fuzzy Logic Controller for Marine Rescue Drone Trajectory-Tracking

^{*}

## Abstract

**:**

## 1. Introduction

- Brings high efficiency in adjusting the Drone to follow the set trajectory.
- Suppress the effect of noise at a high level with high efficiency.
- Reduce the operator’s responsibility thanks to the high learning nature of the controller, making it easier to control.
- High practicality and is a premise for the development of research on actual operation models.

## 2. Drone Model

#### 2.1. Structure of Drone

#### 2.2. Dynamics Model of the Drone

#### 2.3. Overview Neural Network–Fuzzy Logic Controller

## 3. Drone Controller Scheme

#### 3.1. Position Controller and Altitude Controller of Marine Rescue Drone

- −
- 6 degrees of freedom: X, Y, Z, roll, pitch, yaw
- −
- 4 control inputs: ${U}_{1}$ for thrust, ${U}_{2}$, ${U}_{3}$, ${U}_{4}$ for torque on ${x}_{B},{y}_{B},{z}_{B}$ axis.

- −
- The equations describing the vertical motion of the Quadrotor:

#### 3.2. Disturbance Signal

_{0}, y

_{0}, z

_{0}, t)

_{0}(x

_{0}, y

_{0}, z

_{0}, t)

_{0}(x

_{0}, y

_{0}, z

_{0}, t)

_{0}(x

_{0}, y

_{0}, z

_{0}, t)

Algorithm 1: Wind setting algorithm |

% Disturbance Signal ts = 0.01; WindLevel = [10]; WindSpeed = [13244]; |

#### 3.3. Algorithm Neural Network–Fuzzy Controller

Algorithm 2: Neural-Fuzzy controller setting algorithm |

fis = anfis(Drone_NF_ControllerData); numObs = 3; opt = anfisOptions(‘InitialFIS’,6,’EpochNumber’,40); [trainFIS,trainFISError,~,validationFIS,validationFISError] = anfis(NFControllerData,opt); observationInfo = rlNumericSpec([numObs 1]); observationInfo.Name = ‘observations’; numAct = 1; actionInfo = rlNumericSpec([numAct 1]); actionInfo.LowerLimit = −1; actionInfo.UpperLimit = 1; actionInfo.Name = ‘Control’; actionInfo.Description = ‘Thurst’; mdl = ‘Drone_NF_Controller’; load_system(mdl); blk = [mdl,’/RL Data’]; |

#### 3.4. Model of Neural Network–Fuzzy Logic Controller in Matlab/Simulink

## 4. Results and Discussion

## 5. Conclusions

- The design of the Neural-Fuzzy controller will help improve the ability to control the trajectory to help search and rescue at sea become more accurate and faster. That is the main purpose of this article. When using a hybrid controller, it helps individual systems to complement and compensate for their shortcomings, and combine to promote their own strengths, helping to create an efficient system.
- Neural-Fuzzy controller is 67% more efficient, responds approx 85% faster and has better anti-interference ability with three times larger amplitude thrust than PID controller.
- This proposed hybrid controller can be used across different drone sizes, weights, and configurations without the need to re-adjust the PID gain.
- Along with the strong development of science and technology; Neural systems and fuzzy networks are increasingly being applied in many fields. Based on human logic, with the advantage of simplicity and accurate handling of uncertain information; The artificial neural network and the combined fuzzy network have brought many significant effects in the field of control and automation.

## Author Contributions

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Data Availability Statement

## Conflicts of Interest

## References

- Chamayou, G. Drone Theory; Penguin: London, UK, 2015. [Google Scholar]
- Park, S.; Her, J.; Kim, J.; Lee, D. Design, modeling and control of omni-directional aerial robot. In Proceedings of the 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Daejeon, Korea, 9–14 October 2016. [Google Scholar]
- Niu, J.; Wang, D.; Han, P.; Fang, J.; Ren, X.; Qin, Y.; Xu, Z. Image Enhancement of Low Light UAV via Global Illumination Self-aware feature Estimation. In Proceedings of the 2021 3rd International Conference on Natural Language Processing (ICNLP), Beijing, China, 26–28 March 2021. [Google Scholar]
- Vrontisa, D.; Christofi, M.; Pereira, V.; Tarba, S.; Makrides, A.; Trichina, E. Artificial intelligence, robotics, advanced technologies and human resource management: A systematic review. Int. J. Hum. Resour. Manag.
**2021**, 33, 1237–1266. [Google Scholar] [CrossRef] - Bennett, M.K.; Younes, N.; Joyce, K. Automating Drone Image Processing to Map Coral Reef Substrates Using Google Earth Engine. Drones
**2020**, 4, 50. [Google Scholar] [CrossRef] - Ling, G.; Draghic, N. Aerial drones for blood delivery. Transfusion
**2019**, 59, 1608–1611. [Google Scholar] [CrossRef] [PubMed] [Green Version] - Ahmed, S.; Qiu, B.; Kong, C.; Xin, H.; Ahmad, F.; Lin, J. A Data-Driven Dynamic Obstacle Avoidance Method for Liquid-Carrying Plant Protection UAVs, Francisco Manzano-Agugliaro. Agronomy
**2022**, 12, 873. [Google Scholar] [CrossRef] - Scott, J.; Scott, C. Drone Delivery Models for Healthcare. In Proceedings of the 50th Hawaii International Conference on System Sciences, Hilton Waikoloa Village, HI, USA, 4–7 January 2017. [Google Scholar]
- Kim, S.J.; Jeong, Y.; Park, S.; Ryu, K.; Oh, G. A Survey of Drone use for Entertainment and AVR (Augmented and Virtual Reality). In Augmented Reality and Virtual Reality; Springer: Cham, Switzerland, 2017; pp. 339–352. [Google Scholar]
- Packer, J.; Reeves, J. Romancing the Drone: Military Desire and Anthropophobia from SAGE to Swarm. Can. J. Commun.
**2013**, 38, 309–332. [Google Scholar] [CrossRef] [Green Version] - Gray, P.C.; Fleishman, A.B.; Klein, D.J.; McKown, M.W.; Bézy, V.S.; Lohmann, K.J.; Johnston, D.W. A convolutional neural network for detecting sea turtles in drone imagery. Methods Ecol. Evol.
**2018**, 10, 345–355. [Google Scholar] [CrossRef] - Simon, P. Too Big to Ignore: The Business Case for Big Data; John Wiley & Sons: Hoboken, NJ, USA, 2013; Volume 72. [Google Scholar]
- Tezza, D.; Andujar, M. The State-of-the-Art of Human–Drone Interaction: A Survey. IEEE Access
**2019**, 7, 167438–167454. [Google Scholar] [CrossRef] - Um, Y.-C.; Cho, H.-L. Integral γ-Sliding Mode Control for a Quadrotor with Uncertain Time-Varying Mass and External Disturbance. J. Electr. Eng. Technol.
**2022**, 17, 707–716. [Google Scholar] [CrossRef] - Wang, R.; An, A.; Wen, Y.; Song, H. Study on the Influence of Parallel Fuzzy PID Control on the Regulating System of a Bulb Tubular Turbine Generator Unit. J. Electr. Eng. Technol.
**2021**, 16, 1403–1414. [Google Scholar] [CrossRef] - Khan, S.A.; Liu, C.; Ansari, J.A. Centralized Fuzzy Logic Based Optimization of PI Controllers for VSC Control in MTDC Network. J. Electr. Eng. Technol.
**2020**, 15, 2577–2585. [Google Scholar] [CrossRef] - Abdelhay, S.; Zakriti, A. Modeling of a Quadcopter Trajectory Tracking System Using PID Controller, ScienceDirect. Procedia Manuf.
**2019**, 32, 564–571. [Google Scholar] [CrossRef] - Desa, H.; Ahmed, S.F.; Azfar, A.Z. Adaptive Hybrid Control Algorithm Design for Attitude Stabilization of Quadrotor (Drone); Archives Des Sciences, University Malaysia Perlis: Arau, Malaysia, 2013; Volume 66. [Google Scholar]
- Teppo, L. Modelling and Control of Quadcopter; Independent research project in applied mathematics; Aalto University: Espoo, Finland, 2011. [Google Scholar]
- Lin, C.T.; Lee, C.S.G. Neural Fuzzy Systems: A Neural Fuzzy Synergism to Intelligent Systems; Prentice Hall: Hoboken, NJ, USA, 1996; ISBN 0132351692. [Google Scholar]
- Zadeh, L.A. Fuzzy Logic, Neural Networks, and Soft Computing; ACM: New York, NY, USA, 1994. [Google Scholar]
- Tiep, D.K.; Lee, K.N.; Im, D.Y.; Kwak, B.; Ryoo, Y.J. Design of Fuzzy—PID Controller for Path Tracking of Mobile Robot with Diferential Drive. Int. J. Fuzzy Log. Intell. Syst.
**2018**, 18, 220–228. [Google Scholar] [CrossRef] - Lin, C.-T. A neural fuzzy control system with structure and parameter learning. Fuzzy Sets Syst.
**1995**, 70, 183–212. [Google Scholar] [CrossRef] - Lee, S.-J.; Ouyang, C.-S. A Neuro-Fuzzy System ModelingbWith Self-Constructing Rule Generation and Hybrid SVD-Based Learing. IEEE Trans. Fuzzy Syst.
**2003**, 11, 341–353. [Google Scholar] - Sorbelli, F.B.; Corò, F.; Das, S.K.; Pinotti, C.M. Energy-Constrained Delivery of Goods with Drones under Varying Wind Conditions. IEEE Trans. Intell. Transp. Syst.
**2020**, 22, 6048–6060. [Google Scholar] [CrossRef] - Jang, J.-S.R. ANFIS: Adaptive-network-based fuzzy inference system. IEEE Trans. Syst. Man Cybern.
**1993**, 23, 665–685. [Google Scholar] [CrossRef] - Seng, T.L.; Khalid, M.B.; Yusof, R. Tuning of a neural fuzzy controller by genetic algorithm. IEEE Trans. Syst. Man Cybern. Part B Cybern.
**1999**, 29, 226–236. [Google Scholar] [CrossRef] [PubMed] - Sandhu, G.S.; Rattan, K.S. Design of a neural fuzzy controller. In Proceedings of the IEEE International Conference on Systems, Man, and Cybernetics. Computational Cybernetics and Simulation, Orlando, FL, USA, 12–15 October 1997. [Google Scholar]
- Louzazni, M.; Aroudam, E. Control and Stabilization of Three-Phase Grid Connected Photovoltaics using PID-Fuzzy Logic. In Proceedings of the IEEE International Conference on Intelligent Energy and Power Systems (IEPS), Kyiv, Ukraine, 2–6 June 2014. [Google Scholar]
- Gao, W.; Selmic, R.R. Neural network control of a class of nonlinear systems with actuator saturation. IEEE Trans. Neural Net.
**2006**, 17, 147–156. [Google Scholar] [CrossRef] [PubMed] - Zadeh, L.A.; Aliev, R.A. Chapter 16: Adaptive Neuro-Fuzzy Inference Systems (ANFISs). In Fuzzy Logic Theory and Applications; World Scientific Book: Singapore, 2018; pp. 453–493. [Google Scholar]
- The MathWorks, Inc. Fuzzy Logic Toolbox™ User’s Guide; Version 2.2.24; (Release 2016); The MathWorks, Inc.: Natick, MA, USA, 2016. [Google Scholar]
- Demuth, H.; Hagan, M.B.V.M. Neural Network Toolbox™ User’s Guide; Version 9.1; (Release 2016); The Matth Works, Inc.: Natick, MA, USA, 2016. [Google Scholar]

**Figure 7.**Block diagram of the control system using the ANFIS integrated controller and observer of the marine rescue drone.

**Figure 16.**(

**a**) Adjust the height of the Drone to follow the set signal. (

**b**) The actual change of the drone system. (

**c**) Change of the rudder so that the error deviates from the trajectory to zero.

**Figure 17.**Position tracking error. (

**a**) Position tracking error. (

**b**) Drone velocity along coordinate axes.

**Figure 18.**The return phi angle relative to the value set in the simulation. (

**a**) The return phi angle relative to the value set in the simulation. (

**b**) The return theta angle relative to the value set in the simulation.

**Figure 20.**The results compare the Z position (attitude) of the Neural-Fuzzy controller and the classic PID controller (no wind).

**Figure 21.**The results compare the Z position (attitude) of the Neural-Fuzzy controller and the classic PID controller (wind action).

**Figure 22.**Comparison of thrust of unmanned sea rescue aircraft using PID controller and Neural-Fuzzy hybrid controller (no wind).

**Figure 23.**Comparison of thrust of unmanned sea rescue aircraft using PID controller and Neural-Fuzzy hybrid controller (wind action).

Nature | Neural Network | Fuzzy Logic Control |
---|---|---|

Show knowledge | Through weights expressed hidden in the network | Expressed right in the law of composition |

Source of knowledge | From learning patterns | From the experience of experts |

Handling uncertain information | Quantitative | Quantitative and qualitative |

Knowledge retention | In neurons and the weights of each neuron junction | In the law of composition and membership functions |

Ability to update and improve knowledge | Through the learning process | Not available |

Sensitivity to model changes | Low | High |

VN | NR | FR | VFR | |
---|---|---|---|---|

LW | S | S | F | F |

M | S | F | F | VF |

H | S | F | VF | VF |

Value Name | Variable Value | Unit of Variable |
---|---|---|

${U}_{1}$ | ≤30 | N |

${U}_{2}$ | ≤1.2 | Nm |

${U}_{3}$ | ≤1.2 | Nm |

${U}_{4}$ | ≤1.2 | Nm |

Mass (m) | 0.2 | Kg |

Gravity acceleration (g) | 9.8 | $m/{s}^{2}$ |

Moment of inertia (I) | $\left[\begin{array}{cccc}0& 1& 0& 0\\ 0& 0& 1& 0\\ 0& 0& 0& 0.08\end{array}\right]$ | |

Angles | ≤60 | Degree |

Part | P | I | D |
---|---|---|---|

${A}_{1}$ | 0.95 | X | X |

${A}_{2}$ | 0.21 | X | X |

${A}_{3}$ | 1.1 | X | 1.2 |

${A}_{4}$ | 1 | X | X |

${A}_{5}$ | 0.2 | X | X |

${A}_{6}$ | 1.1 | X | 0.45 |

${A}_{7}$ | 0.6 | X | 0.4 |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Pham, D.-A.; Han, S.-H.
Design of Combined Neural Network and Fuzzy Logic Controller for Marine Rescue Drone Trajectory-Tracking. *J. Mar. Sci. Eng.* **2022**, *10*, 1716.
https://doi.org/10.3390/jmse10111716

**AMA Style**

Pham D-A, Han S-H.
Design of Combined Neural Network and Fuzzy Logic Controller for Marine Rescue Drone Trajectory-Tracking. *Journal of Marine Science and Engineering*. 2022; 10(11):1716.
https://doi.org/10.3390/jmse10111716

**Chicago/Turabian Style**

Pham, Duc-Anh, and Seung-Hun Han.
2022. "Design of Combined Neural Network and Fuzzy Logic Controller for Marine Rescue Drone Trajectory-Tracking" *Journal of Marine Science and Engineering* 10, no. 11: 1716.
https://doi.org/10.3390/jmse10111716