TY - GEN
T1 - A spiking neural network with dynamic memory for a real autonomous mobile robot in dynamic environment
AU - Alnajjar, Fady
AU - Zin, Indra Bin Mohd
AU - Murase, Kazuyuki
PY - 2008
Y1 - 2008
N2 - This work concerns practical issues surrounding the application of learning and memory in a real mobile robot towards optimal navigation in dynamic environments. A novel control system that contains two-level units (low-level and high-level) is developed and trained in a physical mobile robot "e-Puck". In the low-level unit, the robot's task is to navigate in a various local environments, by training N numbers of Spiking Neural Networks (SNN) that have the property of spike time-dependent plasticity. All the trained SNNs are stored in a tree-type memory structure, which is located in the high-level unit These stored networks are used as experiences for the robot to enhance its navigation ability in new and previously trained environments. The memory is designed to hold memories of various lengths and has a simple searching mechanism. For controlling the memory size, forgetting and on-line dynamic clustering techniques are used. Experimental results have proved that the proposed model can provide a robot with learning and memorizing capabilities enable it to survive in complex and dynamic environments.
AB - This work concerns practical issues surrounding the application of learning and memory in a real mobile robot towards optimal navigation in dynamic environments. A novel control system that contains two-level units (low-level and high-level) is developed and trained in a physical mobile robot "e-Puck". In the low-level unit, the robot's task is to navigate in a various local environments, by training N numbers of Spiking Neural Networks (SNN) that have the property of spike time-dependent plasticity. All the trained SNNs are stored in a tree-type memory structure, which is located in the high-level unit These stored networks are used as experiences for the robot to enhance its navigation ability in new and previously trained environments. The memory is designed to hold memories of various lengths and has a simple searching mechanism. For controlling the memory size, forgetting and on-line dynamic clustering techniques are used. Experimental results have proved that the proposed model can provide a robot with learning and memorizing capabilities enable it to survive in complex and dynamic environments.
UR - http://www.scopus.com/inward/record.url?scp=56349158772&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=56349158772&partnerID=8YFLogxK
U2 - 10.1109/IJCNN.2008.4634103
DO - 10.1109/IJCNN.2008.4634103
M3 - Conference contribution
AN - SCOPUS:56349158772
SN - 9781424418213
T3 - Proceedings of the International Joint Conference on Neural Networks
SP - 2207
EP - 2213
BT - 2008 International Joint Conference on Neural Networks, IJCNN 2008
T2 - 2008 International Joint Conference on Neural Networks, IJCNN 2008
Y2 - 1 June 2008 through 8 June 2008
ER -