scholarly journals Simply Coded Evolutionary Artificial Neural Networks on a Mobile Robot Control Problem

2010 ◽  
Vol 23 (1) ◽  
pp. 1-8
Author(s):  
Yoshiaki Katada ◽  
Takuya Hidaka
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Mobile Robot ◽  
Control Problem ◽  
Robot Control ◽  
Mobile Robot Control ◽  
Artificial Neural

PRUNING ARTIFICIAL NEURAL NETWORKS USING NEURAL COMPLEXITY MEASURES

2008 ◽  
Vol 18 (05) ◽  
pp. 389-403 ◽  
Author(s):  
THOMAS D. JORGENSEN ◽  
BARRY P. HAYNES ◽  
CHARLOTTE C. F. NORLUND
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Networks ◽  
Robot Control ◽  
Complexity Measures ◽  
Information Theoretic ◽  
Pruning Method ◽  
The Neural Network ◽  
Artificial Neural ◽  
Pruning Technique

This paper describes a new method for pruning artificial neural networks, using a measure of the neural complexity of the neural network. This measure is used to determine the connections that should be pruned. The measure computes the information-theoretic complexity of a neural network, which is similar to, yet different from previous research on pruning. The method proposed here shows how overly large and complex networks can be reduced in size, whilst retaining learnt behaviour and fitness. The technique proposed here helps to discover a network topology that matches the complexity of the problem it is meant to solve. This novel pruning technique is tested in a robot control domain, simulating a racecar. It is shown, that the proposed pruning method is a significant improvement over the most commonly used pruning method Magnitude Based Pruning. Furthermore, some of the pruned networks prove to be faster learners than the benchmark network that they originate from. This means that this pruning method can also help to unleash hidden potential in a network, because the learning time decreases substantially for a pruned a network, due to the reduction of dimensionality of the network.

scholarly journals PREDETERMINED MOVEMENT OF MOBILE ROBOT USING NEURAL NETWORKS

2014 ◽  
pp. 64-68
Author(s):  
Oleh Adamiv ◽  
Vasyl Koval ◽  
Iryna Turchenko
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Mobile Robot ◽  
Formation Process ◽  
Robot Control ◽  
Experimental Results ◽  
The Road ◽  
Mobile Robot Control ◽  
Different Parts ◽  
Training Sets

This paper describes the experimental results of neural networks application for mobile robot control on predetermined trajectory of the road. There is considered the formation process of training sets for neural network, their structure and simulating features. Researches have showed robust mobile robot movement on different parts of the road.

Mobile robot control by neural networks

1990 ◽  
Vol 21 (7) ◽  
pp. 96-106
Author(s):  
Shigemi Nagata Member ◽  
Nobuo Watanabe ◽  
Kazuo Asakawa Nonmembers
Keyword(s):  
Neural Networks ◽  
Mobile Robot ◽  
Robot Control ◽  
Mobile Robot Control

Real-Time Neural Network Based Camera Localization and its Extension to Mobile Robot Control

1997 ◽  
Vol 08 (03) ◽  
pp. 279-293 ◽  
Author(s):  
Doo-Hyun Choi ◽  
Se-Young Oh
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Mobile Robot ◽  
Real Time ◽  
Robot Control ◽  
Imaging System ◽  
System Modeling ◽  
Hybrid Approach ◽  
Camera Localization ◽  
Mobile Robot Control

The feasibility of using neural networks for camera localization and mobile robot control is investigated here. This approach has the advantages of eliminating the laborious and error-prone process of imaging system modeling and calibration procedures. Basically, two different approaches of using neural networks are introduced of which one is a hybrid approach combining neural networks and the pinhole-based analytic solution while the other is purely neural network based. These techniques have been tested and compared through both simulation and real-time experiments and are shown to yield more precise localization than analytic approaches. Furthermore, this neural localization method is also shown to be directly applicable to the navigation control of an experimental mobile robot along the hallway purely guided by a dark wall strip. It also facilitates multi-sensor fusion through the use of multiple sensors of different types for control due to the network's capability of learning without models.

scholarly journals Uso de la Inteligencia Artificial para Incrementar la Autonomía de un Robot Móvil Mindstorms NXT en Tareas de Clasificación de Patrones

2018 ◽  
pp. 50-58
Keyword(s):  
Artificial Intelligence ◽  
Neural Networks ◽  
Artificial Neural Networks ◽  
Mobile Robot ◽  
Programming Language ◽  
Optical Sensors ◽  
Visual Basic ◽  
Artificial Vision ◽  
Artificial Neural ◽  
External Computer

Uso de la Inteligencia Artificial para Incrementar la Autonomía de un Robot Móvil Mindstorms NXT en Tareas de Clasificación de Patrones Abraham Esteban Gamarra Moreno1, Juan Gamarra Moreno2, Job Daniel Gamarra Moreno3 1 Universidad Nacional del Centro del Perú, Av. Mariscal Castilla N° 3909, Huancayo, Perú 2 Universidad Nacional Mayor de San Marcos, Calle Germán Amézaga N° 375, Lima, Perú 3 Universidad Continental, Av. San Carlos 1980, Huancayo, Perú Recibido el 16 de junio del 2018. Aceptado el 5 de julio del 2018 Resumen La inteligencia artificial es un área que intenta dotar de inteligencia a las máquinas y entre los tópicos que desarrolla están los sistemas expertos, la lógica difusa, los sistemas de planificación, los algoritmos de búsqueda, la computación evolutiva, redes neuronales artificiales entre otros. Los tópicos de la inteligencia artificial que utiliza este artículo son la visión artificial y las redes neuronales artificiales; además utiliza el microbot o robot móvil Mindstorms NXT, que tiene una capacidad limitada en el procesamiento, así como en el almacenamiento de información. La limitación del robot móvil se da porque no tiene a bordo un computador potente para procesar los algoritmos de visión artificial y de las redes neuronales artificiales; por lo que se utiliza un computador externo para realizar su control a través de la tecnología bluetooth. El procesamiento de los algoritmos de visión artificial y de redes neuronales artificiales se realiza en el computador externo y las acciones que ejecuta el robot móvil son enviadas a este, a través de la comunicación bluetooth. El artículo considera que existe autonomía en un robot móvil, cuando este realiza sus acciones sin intervención humana y los indicadores seleccionados para medir esta autonomía son la localización autónoma de los patrones a reconocer y el reconocimiento autónomo o clasificación de estos patrones. La implementación de la localización autónoma de los patrones a reconocer utiliza sensores ópticos, sensores ultrasónicos y el lenguaje de programación C#; así como el reconocimiento autónomo de patrones utiliza una cámara inalámbrica ubicada en el robot móvil, algoritmos de visión artificial, redes neuronales artificiales y el lenguaje de programación visual basic .NET. Los resultados muestran que el promedio del indicador porcentaje de patrones localizados en forma correcta en el entorno por el robot móvil Mindstorms NXT es de 37.81% cuando no se usa la inteligencia artificial y es de 97.18% cuando se usa la inteligencia artificial. Además, el promedio del indicador porcentaje de patrones reconocidos en forma correcta en el entorno por el robot móvil Mindstorms NXT es de 46.25% cuando no se usa la inteligencia artificial y es de 96.87% cuando se usa la inteligencia artificial. Descriptores: inteligencia artificial, visión artificial, redes neuronales, clasificación de patrones, sensores ópticos, sensores de ultrasonido, microbots, Mindstorms NXT. Abstract Artificial intelligence is an area that tries to equip the machines with intelligence and among the topics developed are expert systems, fuzzy logic, planning systems, search algorithms, evolutionary computation, artificial neural networks among others. The topics of artificial intelligence used in this article are artificial vision and artificial neural networks; also uses the microbot or mobile robot Mindstorms NXT, which has a limited capacity in the processing, as well as in the storage of information. The limitation of the mobile robot is because it does not have a powerful computer on board to process artificial vision algorithms and artificial neural networks; so an external computer is used to perform its control through bluetooth technology. The processing of artificial vision algorithms and artificial neural networks is done on the external computer and the actions performed by the mobile robot are sent to it, through bluetooth communication. The article considers that there is autonomy in a mobile robot, when it performs its actions without human intervention and the indicators selected to measure this autonomy are the autonomous localization of the patterns to be recognized and the autonomous recognition or classification of these patterns. The implementation of the autonomous localization of the patterns to be recognized uses optical sensors, ultrasonic sensors and the C # programming language; as well as the autonomous recognition of patterns uses a wireless camera located in the mobile robot, artificial vision algorithms, artificial neural networks and the visual basic .NET programming language. The results show that the average of the indicator percentage of patterns correctly located in the environment by the Mindstorms NXT mobile robot is 37.81% when artificial intelligence is not used and it is 97.18% when artificial intelligence is used. In addition, the average of the indicator percentage of patterns correctly recognized in the environment by the Mindstorms NXT mobile robot is 46.25% when artificial intelligence is not used and is 96.87% when using artificial intelligence. Keywords: artificial intelligence, artificial vision, artificial neural networks, pattern classification, optical sensors, ultrasound sensors, microbots.

Sign in / Sign up

Export Citation Format

Share Document