HomePage
 




 


 

ISSN 2083-6473
ISSN 2083-6481 (electronic version)
 

 

 

Editor-in-Chief

Associate Editor
Prof. Tomasz Neumann
 

Published by
TransNav, Faculty of Navigation
Gdynia Maritime University
3, John Paul II Avenue
81-345 Gdynia, POLAND
www http://www.transnav.eu
e-mail transnav@umg.edu.pl
Differentiable Programming for the Autonomous Movement Planning of a Small Vessel
ABSTRACT: In this work we explore the use of differentiable programming to allow autonomous movement planning of a small vessel. We aim for an end to end architecture where the machine learning algorithm directly controls engine power and rudder movements of a simulated vessel to reach a defined goal. Differentiable programming is a novel machine learning paradigm, that allows to define a systems parameterized response to control commands in imperative computer code and to use automatic differentiation and analysis of the information flow from the controlling inputs and parameters to the resulting trajectory to compute derivatives to be used as search directions in an iterative algorithm to optimize a goal function. Initially the method does not know about any manoeuvring or the vessels response to control commands. The method autonomously learns the vessels behaviour from several simulation runs. Finally, we will show how the simulated vessel is able to fulfil some small missions, like crossing a flowing river while avoiding crossing traffic.
REFERENCES
Bottou, L., Curtis, F.E., Nocedal, J.: Optimization Methods for Large-Scale Machine Learning. (2018).
Hesselbarth, A., Medina, D., Ziebold, R., - doi:10.1109/MITS.2020.2994103
Hu, Y.: DiffTaichi: Differentiable Programming for Physical Simulation, https://github.com/yuanming-hu/difftaichi, last accessed 2021/04/27.
Hu, Y., Anderson, L., Li, T.-M., Sun, Q., Carr, N., Ragan-Kelley, J., Durand, F.: Differentiable Programming for Physical Simulation. Presented at the International Conference on Learning Representations (2020).
Kiefer, J., Wolfowitz, J.: Stochastic Es - doi:10.1214/aoms/1177729392
Kingma, D.P., Ba, J.: Adam: A Method for Stochastic Optimization. Presented at the 3rd International Conference for Learning Representations , San Diego (2015).
Lloyd’s Register Groupe: Design Code for Unmanned Marine Systems. (2017).
Medina, D., Vilà-Valls, J., Hessel - doi:10.3390/rs12121955
Robbins, H., Monro, S.: A Stochastic Approximation Method. The Annals of Mathematical Statistics. 22, 3, 400–407 (1951).
Rødseth, Ø., Nordahl, H.: Definition of au - doi:10.13140/RG.2.2.21069.08163
SAE International: J3016B: Taxonomy and Definitions for Terms Related to Driving Automation Systems for On-Road Motor Vehicles - SAE International, https://www.sae.org/standards/content/j3016_201806/, last accessed 2021/04/27.
Schubert, A.U., Kurowski, M., Damerius, R., Fisc - doi:10.1088/1742-6596/1357/1/012006
Schubert, A.U., Kurowski, M., Gluch, M., Simanski - doi:10.24868/issn.2631-8741.2018.020
Citation note:
Bahls C., Schubert A.: Differentiable Programming for the Autonomous Movement Planning of a Small Vessel. TransNav, the International Journal on Marine Navigation and Safety of Sea Transportation, Vol. 15, No. 3, doi:10.12716/1001.15.03.01, pp. 493-499, 2021

File downloaded 64 times








Important: TransNav.eu cookie usage
The TransNav.eu website uses certain cookies. A cookie is a text-only string of information that the TransNav.EU website transfers to the cookie file of the browser on your computer. Cookies allow the TransNav.eu website to perform properly and remember your browsing history. Cookies also help a website to arrange content to match your preferred interests more quickly. Cookies alone cannot be used to identify you.
Akceptuję pliki cookies z tej strony