Abstract
Unmanned helicopter has broad application prospects, both in civil and military field. Helicopter is a strong coupled with many phenomena, inherently unstable, high-order, time-varying nonlinear complex system. It is a great challenge to investigate the system identification of helicopter, particularly when the non-stationary flight regimes are considered. In this paper, we address the system identification as dynamic regression. Inspired by the feature expression ability of deep learning, we use Deep convolutional neural networks (CNN) to represent dynamics model of helicopter. The parameters of the network are directly learned from the real flight data collected from helicopter. We provide model initialization method and optimization details for training. Since it captures the hidden states in aerobatic maneuvers without a-priori, the proposed identifier manifests strong robustness and high accuracy, even for untrained aerobatic maneuvers. The effectiveness of the proposed method is verified by various experiments on the real-world flight data from Stanford Autonomous Helicopter Project. Specifically, the Deep CNN identifier improves 71.60% overall in RMS acceleration prediction over the lasted Deep Rectified Linear Unit (ReLU) Network Model.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Starke, E.A., Staley, J.T.: Application of modern aluminum alloys to aircraft. Prog. Aerosp. Sci. 32(2–3), 131–172 (1996)
Johnson, W.: Helicopter Theory. Courier Corporation, New York (2012)
Leishman, G.J.: Principles of Helicopter Aerodynamics with CD Extra. Cambridge University Press, Cambridge (2006)
Abbeel, P., Ganapathi, V., Ng, A.Y.: Learning vehicular dynamics, with application to modeling helicopters. In: Advances in Neural Information Processing Systems, pp. 1–8 (2006)
Padfield, G.D.: The Theory and Application of Flying Qualities and Simulation Modeling. Helicopter Flight Dynamics. AIAA: Education Series, Washington, DC (1996)
LeCun, Y., Bengio, Y., Hinton, G.: Deep learning. Nature 521(7553), 436–444 (2015)
Schmidhuber, J.: Deep learning in neural networks: an overview. Neural Netw. 61, 85–117 (2015)
Punjani, A., Abbeel, P.: Deep learning helicopter dynamics models. In: 2015 IEEE International Conference on Robotics and Automation (ICRA), pp. 3223–3230. IEEE (2015)
Takens, F.: Detecting strange attractors in turbulence. In: Rand, D., Young, L.-S. (eds.) Dynamical Systems and Turbulence, Warwick 1980. LNM, vol. 898, pp. 366–381. Springer, Heidelberg (1981). doi:10.1007/BFb0091924
Robinson, J.C.: A topological delay embedding theorem for infinite-dimensional dynamical systems. Nonlinearity 18(5), 2135 (2005)
Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. In: Advances in Neural Information Processing Systems, pp. 1097–1105 (2012)
Bouvrie, J.: Notes on Convolutional Neural Networks (2006)
Abbeel, P., Coates, A., Ng, A.Y.: Autonomous helicopter aerobatics through apprenticeship learning. Int. J. Robot. Res. 29(13), 1608–1639 (2010)
Riedmiller, M., Braun, H.: A direct adaptive method for faster backpropagation learning: the RPROP algorithm. In: 1993 IEEE International Conference on Neural Networks, pp. 586–591. IEEE (1993)
Jia, Y., Shelhamer, E., Donahue, J., Karayev, S., Long, J., Girshick, R., Darrell, T.: Caffe: convolutional architecture for fast feature embedding. In: Proceedings of the 22nd ACM International Conference on Multimedia, pp. 675–678. ACM (2014)
Guo, Z.H., Wu, J., Lu, H.Y., Wang, J.Z.: A case study on a hybrid wind speed forecasting method using BP neural network. Knowl.-Based Syst. 24(7), 1048–1056 (2011)
Acknowledgments
This work was supported in part by the National Natural Science Foundation of China (61422307 and 61673361), the Scientific Research Staring Foundation for the Returned Overseas Chinese Scholars and Ministry of Education of China. Authors also gratefully acknowledge supports from the Youth Top-notch Talent Support Program and the Youth Yangtze River Scholar.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this paper
Cite this paper
Chen, S., Cao, Y., Kang, Y., Zhu, R., Li, P. (2017). Deep CNN Identifier for Dynamic Modelling of Unmanned Helicopter. In: Liu, D., Xie, S., Li, Y., Zhao, D., El-Alfy, ES. (eds) Neural Information Processing. ICONIP 2017. Lecture Notes in Computer Science(), vol 10639. Springer, Cham. https://doi.org/10.1007/978-3-319-70136-3_6
Download citation
DOI: https://doi.org/10.1007/978-3-319-70136-3_6
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-70135-6
Online ISBN: 978-3-319-70136-3
eBook Packages: Computer ScienceComputer Science (R0)