Prediction of Flight Status of Logistics UAVs Based on an Information Entropy Radial Basis Function Neural Network
Abstract
:1. Introduction
2. Information Entropy
3. RBF Neural Network
4. D-RBF Neural Network
5. Proof of Convergence
Dynamic Adjustment of the Network Structure
6. Prediction of Regression Value of UAV Flight Status
6.1. Establishment and Analysis of UAV Model
6.2. Flow Field Analysis during Flight
6.3. Flight Test
6.4. Normalization of Experimental Data
6.5. Simulation Experiment
- (1)
- The D-RBF neural network does not depend on the initial structure of the network. It can dynamically adjust the number of neurons and disconnect the weakest connection, according to the connection strength of hidden layer neurons and output layer neurons. It can respond in real time.
- (2)
- The entropy of hidden layer neuron and output layer neuron is calculated. The output information of hidden layer neuron and the connection strength between hidden layer neuron and output layer neuron are measured, and the mathematical expression is given to realize the dynamic adjustment of network structure.
- (3)
- By experimentally comparing the performance differences among D-RBF, SORBF, RBF, and GGAP-RBF, the convergence speed of the average error is accelerated and D-RBF is proved to have good convergence performance.
- (4)
- The present D-RBF neural network solves the regression prediction of the flight state of the logistics UAV, providing guidance for the research of the flight stability performance of the logistics UAV under different loads and flight speeds.
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Conflicts of Interest
References
- Sheng, Z.; Yang, S.; Yu, Y.; Vasilakos, A.V.; McCann, J.A.; Leung, K.K. A Survey on the Ietf Protocol Suite for the Internet of Things: Standards, Challenges, and Opportunities. IEEE Wirel. Commun. 2013, 20, 91–98. [Google Scholar] [CrossRef]
- Bodner, D.A.; Wade, J.P.; Watson, W.R.; Kamberov, G.I. Designing an Experiential Learning Environment for Logistics and Systems Engineering. Procedia Computer Science 2013, 16, 1082–1091. [Google Scholar] [CrossRef] [Green Version]
- Salvo, G.; Caruso, L.; Scordo, A. Urban Traffic Analysis through an UAV. Procedia Soc. Behav. Sci. 2014, 111, 1083–1091. [Google Scholar] [CrossRef] [Green Version]
- Perez-Montenegro, C.; Scanavino, M.; Bloise, N.; Capello, E.; Guglieri, G.; Rizzo, A. A Mission Coordinator Approach for a Fleet of UAVs in Urban Scenarios. Transp. Res. Procedia 2018, 35, 110–119. [Google Scholar] [CrossRef]
- Lane, E.L.T.B.L. UAV market takes off for Intelligent Energy in inventory solutions. Fuel Cells Bull. 2017. [Google Scholar] [CrossRef]
- Porretta, M.; Dupuy, M.-D.; Schuster, W.; Majumdar, A.; Ochieng, W. Performance Evaluation of a Novel 4D Trajectory Prediction Model for Civil Aircraft. J. Navig. 2008, 61, 393. [Google Scholar] [CrossRef]
- Prandini, M.; Putta, V.; Hu, J. A probabilistic measure of air traffic complexity in 3-D airspace. Int. J. Adapt. Control Signal Proc. 2010, 24, 813–829. [Google Scholar] [CrossRef]
- Benavides, J.V.; Kaneshige, J.; Sharma, S.; Panda, R.; Steglinski, M. Implementation of a Trajectory Prediction Function for Trajectory Based Operations. In Proceedings of the Aiaa Atmospheric Flight Mechanics Conference, Atlanta, GA, USA, 16–20 June 2014. [Google Scholar]
- Chen, L.; Shan, Y.; Tian, W.; Li, B.; Cao, D. A Fast and Efficient Double-Tree RRT∗-Like Sampling-Based Planner Applying on Mobile Robotic Systems. IEEE/ASME Trans. Mech. 2018, 23, 2568–2578. [Google Scholar] [CrossRef]
- Wiest, J.; Hoffken, M.; Kresel, U.; Dietmayer, K. Probabilistic trajectory prediction with Gaussian mixture models. In Proceedings of the Intelligent Vehicles Symposium (IV), Madrid, Spain, 3–7 June 2012. [Google Scholar]
- Qiao, S.; Shen, D.; Wang, X.; Han, N.; Zhu, W. A Self-Adaptive Parameter Selection Trajectory Prediction Approach via Hidden Markov Models. IEEE Trans. Intell. Transp. Syst. 2015, 16, 284–296. [Google Scholar] [CrossRef]
- Sanchez-Garcia, J.; Reina, D.G.; Toral, S.L. A distributed PSO-based exploration algorithm for a UAV network assisting a disaster scenario. Future Gener. Comput. Syst. 2018, 90, 129–148. [Google Scholar] [CrossRef]
- Yuan, W.; Liu, Y.; Wang, H.; Cao, Y. A Geometric Structure-Based Particle Swarm Optimization Algorithm for Multiobjective Problems. IEEE Trans. Syst. Man Cybern. Syst. 2017, 47, 2516–2537. [Google Scholar] [CrossRef] [Green Version]
- Hafez, A.T.; Kamel, M.A. Cooperative Task Assignment and Trajectory Planning of Unmanned Systems Via HFLC and PSO. Unmanned Syst. 2019, 7, 65–81. [Google Scholar] [CrossRef]
- Song, B.; Wang, Z.; Zou, L. On Global Smooth Path Planning for Mobile Robots using a Novel Multimodal Delayed PSO Algorithm. Cogn. Comput. 2017, 9, 5–17. [Google Scholar] [CrossRef]
- Luengo, J.; García, S.; Herrera, F. A study on the use of imputation methods for experimentation with Radial Basis Function Network classifiers handling missing attribute values: The good synergy between RBFNs and EventCovering method. Neural Netw. 2009, 23, 406–418. [Google Scholar] [CrossRef] [PubMed]
- Stefano, F.; Francesco, B.; Vincenzo, P.; Alberto, B.N. A hierarchical RBF online learning algorithm for real-time 3-D scanner. IEEE Trans. Neural Netw. 2010, 21, 275–285. [Google Scholar]
- Bortman, M.; Aladjem, M. A growing and pruning method for radial basis function networks. IEEE Trans. Neural Networks 2009, 20. [Google Scholar] [CrossRef] [Green Version]
- Platt, J. A Resource-Allocating Network for Function Interpolation. Neural Comput. 1991, 3, 213–225. [Google Scholar] [CrossRef]
- Yingwei, L.; Sundararajan, N.; Saratchandran, P. A Sequential Learning Scheme for Function Approximation Using Minimal Radial Basis Function Neural Networks. Neural Comput. 1997, 9, 461–478. [Google Scholar] [CrossRef]
- Panchapakesan, C.; Palaniswami, M.; Ralph, D.; Manzie, C. Effects of moving the center’s in an RBF network. IEEE Trans. Neural Networks 2002, 13, 1299–1307. [Google Scholar] [CrossRef]
- Lian, J.; Lee, Y.; Sudhoff, S.D.; Zak, S.H. Self-organizing radial basis function network for real-time approximation of continuous-time dynamical systems. IEEE Trans. Neural Netw. 2008, 19, 460–474. [Google Scholar] [CrossRef]
- Huang, G.-B.; Saratchandran, P.; Sundararajan, N. A generalized growing and pruning RBF (GGAP-RBF) neural network for function approximation. IEEE Trans. Neural Netw. 2005, 16, 57–67. [Google Scholar] [CrossRef]
- Yu, H.; Reiner, P.D.; Xie, T.; Bartczak, T.; Wilamowski, B.M. An Incremental Design of Radial Basis Function Networks. IEEE Trans. Neural Networks Learn. Syst. 2014, 25, 1793–1803. [Google Scholar] [CrossRef]
- Hao, C.; Yu, G.; Xia, H. Online modeling with tunable RBF network. IEEE Trans. Cybern. 2013, 43, 935–947. [Google Scholar]
- Feng, H.-M. Self-generation RBFNs using evolutional PSO learning. Neurocomputing 2006, 70, 241–251. [Google Scholar] [CrossRef]
- Buzzi, C.; Grippo, L.; Sciandrone, M. Convergent Decomposition Techniques for Training RBF Neural Networks. Neural Comput. 2001, 13, 1891–1920. [Google Scholar] [CrossRef] [Green Version]
- Li, D.; Wang, Z.; Cao, C.; Liu, Y. Information entropy based sample reduction for support vector data description. Appl. Soft Comput. 2018, 71, 1153–1160. [Google Scholar] [CrossRef]
- Harte, J.; Newman, E.A. Maximum information entropy: A foundation for ecological theory. Trends Ecol. Evol. 2014, 29, 384–389. [Google Scholar] [CrossRef]
- Zheng, K.; Wang, X. Feature selection method with joint maximal information entropy between features and class. Pattern Recognit. 2018, 77, 20–29. [Google Scholar] [CrossRef]
- He, Q.P. Information Entropy Based Indices for Variable Selection Performance Assessment; Computer Aided Chemical Engineering: San Diego, CA, USA, 2018; pp. 2281–2286. [Google Scholar]
- Majumdar, S.; Mukherjee, N.; Roy, A.K. Information entropy and complexity measure in generalized Kratzer potential. Chem. Phys. Lett. 2018, 716, 257–264. [Google Scholar] [CrossRef] [Green Version]
- Santos, W.D.D. The entropic and symbolic components of information. BioSystems 2019, 182, 17–20. [Google Scholar] [CrossRef]
- Neves, G.; Cooke, S.F.; Bliss, T.V.P. Synaptic plasticity, memory and the hippocampus: A neural network approach to causality. Nat. Rev. Neurosci. 2008, 9, 65–75. [Google Scholar] [CrossRef] [PubMed]
- Broomhead, D.S.; Lowe, D. Multivariable Functional. Interpolation and Adaptative Networks. Complex Syst. 1988, 2, 321–355. [Google Scholar]
- Erdelj, M.; Król, M.; Natalizio, E. Wireless Sensor Networks and Multi-UAV systems for natural disaster management. Comput. Netw. 2017, 124, 72–86. [Google Scholar] [CrossRef]
Number | Pulling Force (N) | Output Pull (N) | Flight Resistance (N) | Fuselage Weight (kg) | Flight Speed (m/s) | Flight Time (min) | Flight Condition Weights (Excellent, Moderate, Poor, Noisy) |
---|---|---|---|---|---|---|---|
1 | 3.2 | 1.76 | 0.21 | 1.1 | 3.58 | 9.21 | (0.90, 0.05, 0.04, 0.01) |
2 | 3.2 | 1.92 | 0.23 | 1.2 | 3.9 | 8.46 | (0.87, 0.1, 0.03, 0) |
3 | 3.2 | 2.08 | 0.24 | 1.4 | 4.14 | 6.24 | (0.68, 0.1, 0.0.08, 0.1) |
4 | 3.2 | 2.24 | 0.26 | 1.0 | 4.72 | 7.18 | (0.88, 0.1, 0.02, 0) |
5 | 3.2 | 2.4 | 0.28 | 1.7 | 4.10 | 3.76 | (0.58, 0.08, 0.1, 0.24) |
6 | 3.2 | 2.88 | 0.34 | 1.4 | 4.65 | 5.28 | (0.74, 0.08, 0.06, 0.12) |
7 | 3.2 | 3.04 | 0.36 | 1.1 | 6.08 | 5.46 | (0.84, 0.13, 0.03, 0) |
8 | 3.2 | 3.2 | 0.38 | 2.0 | 5.32 | 2.19 | (0.21, 0.13, 0.17, 0.49) |
9 | 4.0 | 2.6 | 0.31 | 1.0 | 4.37 | 7.92 | (0.91, 0.06, 0.03, 0) |
10 | 4.0 | 2.8 | 0.33 | 1.2 | 5.01 | 6.31 | (0.89, 0.07, 0.04, 0) |
11 | 4.0 | 3.2 | 0.38 | 1.6 | 5.98 | 5.41 | (0.84, 0.09, 0.04, 0.03) |
12 | 4.0 | 3.4 | 0.40 | 1.8 | 6.08 | 3.85 | (0.71, 0.05, 0.03, 0.21) |
13 | 4.0 | 3.6 | 0.42 | 2.1 | 6.01 | 2.77 | (0.41, 0.15, 0.06, 0.38) |
14 | 4.0 | 3.8 | 0.45 | 1.6 | 6.88 | 4.08 | (0.69, 0.21, 0.06, 0.04) |
15 | 4.0 | 4.0 | 0.47 | 1.4 | 7.04 | 4.14 | (0.76, 0.16, 0.06, 0.02) |
16 | 4.8 | 3.12 | 0.37 | 1.7 | 5.12 | 7.36 | (0.91, 0.04, 0.05, 0) |
17 | 4.8 | 3.36 | 0.40 | 1.9 | 6.08 | 6.01 | (0.79, 0.09, 0.04, 0.08) |
18 | 4.8 | 3.6 | 0.42 | 2.8 | 4.75 | 3.56 | (0.40, 0.01, 0.08, 0.51) |
19 | 4.8 | 4.08 | 0.48 | 2.1 | 5.63 | 4.28 | (0.51, 0.25, 0.1, 0.14) |
20 | 4.8 | 4.56 | 0.54 | 1.4 | 7.52 | 4.83 | (0.87, 0.08, 0.05, 0) |
21 | 4.8 | 4.8 | 0.56 | 1.9 | 7.49 | 4.27 | (0.71, 0.18, 0.07, 0.04) |
Normalized Processing Nodes | Maximum Pulling Force | Output Pull | Flight Resistance | Weight Capacity | Flight Speed | Endurance Time |
---|---|---|---|---|---|---|
1 | 0.32 | 0.176 | 0.021 | 0.11 | 0.358 | 0.921 |
2 | 0.32 | 0.192 | 0.023 | 0.12 | 0.39 | 0.846 |
3 | 0.32 | 0.208 | 0.024 | 0.14 | 0.414 | 0.624 |
4 | 0.32 | 0.224 | 0.026 | 0.10 | 0.472 | 0.718 |
5 | 0.32 | 0.240 | 0.028 | 0.17 | 0.410 | 0.376 |
Function | Algorithm | Expected Error | Detection Error | Final Network | Convergence Time |
---|---|---|---|---|---|
Regression fitting | D-RBF | 0.01 | 0.0138 | 24 | 79.40 |
RBF | 0.01 | 0.0326 | 6 | 236.72 | |
SORBF | 0.01 | 0.0163 | 20 | 84.21 | |
GGAP-RBF | 0.01 | 0.0142 | 19 | 99.12 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2021 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
Yang, Q.; Ye, Z.; Li, X.; Wei, D.; Chen, S.; Li, Z. Prediction of Flight Status of Logistics UAVs Based on an Information Entropy Radial Basis Function Neural Network. Sensors 2021, 21, 3651. https://doi.org/10.3390/s21113651
Yang Q, Ye Z, Li X, Wei D, Chen S, Li Z. Prediction of Flight Status of Logistics UAVs Based on an Information Entropy Radial Basis Function Neural Network. Sensors. 2021; 21(11):3651. https://doi.org/10.3390/s21113651
Chicago/Turabian StyleYang, Qin, Zhaofa Ye, Xuzheng Li, Daozhu Wei, Shunhua Chen, and Zhirui Li. 2021. "Prediction of Flight Status of Logistics UAVs Based on an Information Entropy Radial Basis Function Neural Network" Sensors 21, no. 11: 3651. https://doi.org/10.3390/s21113651