Evolving spiking neural network controllers for autonomous robots

被引：60
作者：
Hagras, H [1 ]
Pounds-Cornish, A [1 ]
Colley, M [1 ]
Callaghan, V [1 ]
Clarke, G [1 ]
机构：
[1] Univ Essex, Dept Comp Sci, Colchester CO4 3SQ, Essex, England
来源：
2004 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION, VOLS 1- 5, PROCEEDINGS | 2004年
关键词：
spiking neural network; genetic algorithms; autonmous mobile robots; robot navigation; fuzzy controllers;
D O I：
10.1109/ROBOT.2004.1302446
中图分类号：
TP [自动化技术、计算机技术];
学科分类号：
0812 ;
摘要：
In this paper we introduce a novel mechanism for controlling autonomous mobile robots that is based on using Spiking Neural Networks (SNNs). SNNs are inspired by biological neurons that communicate using pulses or spikes. As SNNs have shown to be excellent control systems for biological organisms, they have the potential to produce good control systems for autonomous robots. In this paper we present the use and benefits of SNNs for mobile robot control. We also present an adaptive Genetic Algorithm (GA) to evolve the weights of the SNNs online using real robots. The adaptive GA using adaptive crossover and mutation converge in a small number of generations to solutions that allow the robots to complete the desired tasks. We have performed many experiments using real mobile robots to test the evolved SNNs in which the SNNs provided a good response.
引用
收藏
页码：4620 / 4626
页数：7
相关论文
共 19 条
[11]   On the relevance of time in neural computation and learning [J].
Maass, W .
THEORETICAL COMPUTER SCIENCE, 2001, 261 (01) :157-178
[12]  
MAASS W, 1996, AUSTR C NEUR NETW
[13]  
Maass W, 1999, Pulsed neural networks
[14]  
MAASS W, 1995, ADV NEURAL INFORMATI, V7, P183
[15]   VISUAL FEATURE INTEGRATION AND THE TEMPORAL CORRELATION HYPOTHESIS [J].
SINGER, W ;
GRAY, CM .
ANNUAL REVIEW OF NEUROSCIENCE, 1995, 18 :555-586
[16]  
SINGER W, 1990, CONCEPTS NEUROSCI, P1
[17]  
SRINIVAS M, 1996, ADAPTATION GENETIC A
[18]  
THORPE S, 1997, RAPID VISUAL PROCESS
[19]  
TONKES B, 1997, P 8 AUSTR C NEUR NET
12