Sailboat navigation control system based on spiking neural networks

Nelson Santiago Giraldo; Sebastián Isaza; Ricardo Andrés Velásquez

doi:10.1007/s11768-023-00150-1

Sailboat navigation control system based on spiking neural networks

Título traducido de la contribución: Sistema de control de navegación de veleros basado en redes neuronales spiking

Nelson Santiago Giraldo, Sebastián Isaza, Ricardo Andrés Velásquez

Producción científica: Contribución a una revista › Artículo › revisión exhaustiva

Resumen

En este artículo, presentamos el desarrollo de un sistema de control de navegación para un velero basado en redes neuronales con espigas (SNN). Nuestra inspiración para esta elección de redes radica en su potencial para lograr una computación rápida y de bajo consumo energético en hardware especializado. Para entrenar nuestro sistema, utilizamos la regla de aprendizaje por refuerzo de plasticidad dependiente del tiempo de picos modulados y un entorno de simulación basado en la biblioteca BindsNET y el simulador USVSim. Nuestro objetivo fue desarrollar un sistema de control basado en redes neuronales de espigas que pueda aprender políticas que permitan a los veleros navegar entre dos puntos siguiendo una línea recta o realizando estrategias de virada y trasluchada, dependiendo de las condiciones del escenario de navegación. Presentamos la definición matemática del problema, el esquema de funcionamiento del entorno de simulación, los controladores de red neuronal spiking y la estrategia de control utilizada. Como resultado, obtuvimos 425 controladores basados en SNN que completaron la tarea de navegación propuesta, lo que indica que el entorno de simulación y la estrategia de control implementada funcionan eficazmente. Por último, comparamos el comportamiento de nuestro mejor controlador con el de otros algoritmos y presentamos algunas posibles estrategias para mejorar su rendimiento.

Título traducido de la contribución	Sistema de control de navegación de veleros basado en redes neuronales spiking
Idioma original	Inglés estadounidense
Páginas (desde-hasta)	489–504
Número de páginas	16
Publicación	Control Theory and Technology
Volumen	21
DOI	https://doi.org/10.1007/s11768-023-00150-1
Estado	Publicación electrónica previa a su impresión - ago. 29 2023

Áreas temáticas de ASJC Scopus

Ingeniería de control y sistemas
Procesamiento de senales
Sistemas de información
Modelización y simulación
Ingeniería aeroespacial
Control y optimización
Ingeniería eléctrica y electrónica

ODS de las Naciones Unidas

Este resultado contribuye a los siguientes Objetivos de Desarrollo Sostenible

Acceder al documento

10.1007/s11768-023-00150-1

Otros archivos y enlaces

Citar esto

@article{262feefaf9f143999b9eba6d8687d1ca,

title = "Sailboat navigation control system based on spiking neural networks",

abstract = "In this paper, we presented the development of a navigation control system for a sailboat based on spiking neural networks (SNN). Our inspiration for this choice of network lies in their potential to achieve fast and low-energy computing on specialized hardware. To train our system, we use the modulated spike time-dependent plasticity reinforcement learning rule and a simulation environment based on the BindsNET library and USVSim simulator. Our objective was to develop a spiking neural network-based control systems that can learn policies allowing sailboats to navigate between two points by following a straight line or performing tacking and gybing strategies, depending on the sailing scenario conditions. We presented the mathematical definition of the problem, the operation scheme of the simulation environment, the spiking neural network controllers, and the control strategy used. As a result, we obtained 425 SNN-based controllers that completed the proposed navigation task, indicating that the simulation environment and the implemented control strategy work effectively. Finally, we compare the behavior of our best controller with other algorithms and present some possible strategies to improve its performance.",

author = "Giraldo, {Nelson Santiago} and Sebasti{\'a}n Isaza and Vel{\'a}squez, {Ricardo Andr{\'e}s}",

note = "Funding Information: The Authors declare that this work was supported by the University of Antioquia with project PRG2017-16182 and by the Colombia Scientific Program within the framework of the call Ecosistema Cient{\'i}fico (Contract No. FP44842-218-2018). Publisher Copyright: {\textcopyright} 2023, The Author(s).",

year = "2023",

month = aug,

day = "29",

doi = "10.1007/s11768-023-00150-1",

language = "English (US)",

volume = "21",

pages = "489–504",

journal = "Control Theory and Technology",

issn = "2095-6983",

publisher = "Springer Science + Business Media",

}

TY - JOUR

T1 - Sailboat navigation control system based on spiking neural networks

AU - Giraldo, Nelson Santiago

AU - Isaza, Sebastián

AU - Velásquez, Ricardo Andrés

N1 - Funding Information: The Authors declare that this work was supported by the University of Antioquia with project PRG2017-16182 and by the Colombia Scientific Program within the framework of the call Ecosistema Científico (Contract No. FP44842-218-2018). Publisher Copyright: © 2023, The Author(s).

PY - 2023/8/29

Y1 - 2023/8/29

N2 - In this paper, we presented the development of a navigation control system for a sailboat based on spiking neural networks (SNN). Our inspiration for this choice of network lies in their potential to achieve fast and low-energy computing on specialized hardware. To train our system, we use the modulated spike time-dependent plasticity reinforcement learning rule and a simulation environment based on the BindsNET library and USVSim simulator. Our objective was to develop a spiking neural network-based control systems that can learn policies allowing sailboats to navigate between two points by following a straight line or performing tacking and gybing strategies, depending on the sailing scenario conditions. We presented the mathematical definition of the problem, the operation scheme of the simulation environment, the spiking neural network controllers, and the control strategy used. As a result, we obtained 425 SNN-based controllers that completed the proposed navigation task, indicating that the simulation environment and the implemented control strategy work effectively. Finally, we compare the behavior of our best controller with other algorithms and present some possible strategies to improve its performance.

AB - In this paper, we presented the development of a navigation control system for a sailboat based on spiking neural networks (SNN). Our inspiration for this choice of network lies in their potential to achieve fast and low-energy computing on specialized hardware. To train our system, we use the modulated spike time-dependent plasticity reinforcement learning rule and a simulation environment based on the BindsNET library and USVSim simulator. Our objective was to develop a spiking neural network-based control systems that can learn policies allowing sailboats to navigate between two points by following a straight line or performing tacking and gybing strategies, depending on the sailing scenario conditions. We presented the mathematical definition of the problem, the operation scheme of the simulation environment, the spiking neural network controllers, and the control strategy used. As a result, we obtained 425 SNN-based controllers that completed the proposed navigation task, indicating that the simulation environment and the implemented control strategy work effectively. Finally, we compare the behavior of our best controller with other algorithms and present some possible strategies to improve its performance.

UR - http://www.scopus.com/inward/record.url?scp=85169014415&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85169014415&partnerID=8YFLogxK

U2 - 10.1007/s11768-023-00150-1

DO - 10.1007/s11768-023-00150-1

M3 - Article

AN - SCOPUS:85169014415

SN - 2095-6983

VL - 21

SP - 489

EP - 504

JO - Control Theory and Technology

JF - Control Theory and Technology

ER -

Sailboat navigation control system based on spiking neural networks

Resumen

Áreas temáticas de ASJC Scopus

ODS de las Naciones Unidas

Acceder al documento

Otros archivos y enlaces

Huella

Citar esto