DocumentCode :
1849817
Title :
Mechanism for an Intelligent Neural network based Driving system
Author :
Srinivasan, T. ; Jonathan, J. B Siddharth ; Chandrasekhar, Arvind
Author_Institution :
Sri Venkateswara Coll. of Eng., Sriperumbudur, India
fYear :
2004
fDate :
38199
Firstpage :
53
Lastpage :
60
Abstract :
Over the past decade, the field of automated intelligent transport systems has been the focus of intensive research. This paper proposes a Mechanism for Intelligent Neural network based Driving system (MIND), an advanced automated transport system with considerable advantages over previous attempts in this field. The system uses a multilayer feedforward neural network with backpropagation learning. In addition, the design of MIND involves the convergence of a plethora of technologies like the Global Positioning System (GPS), a geographic information system (GIS), and laser ranging. MIND can guide a mobile agent through a hostile and unfamiliar domain after being trained by a human user with domain expertise. One of the many areas in which MIND scores against the competition is that the system is completely domain independent and incurs a lot less processor overhead. MIND thus provides more functionality even though it requires a lot less input as compared to other attempts in this field This reduction in the size of the input vector translates into more efficient and faster processing. Another of MIND´s hallmark features is its ability to negotiate turns and implement lane-changing maneuvers with a view to overtaking obstacles. It does this by employing a novel technique, selective net masking. A simulation of MIND´s neural network was performed on a variety of network topologies, and the best network selected.
Keywords :
Global Positioning System; automated highways; backpropagation; collision avoidance; feedforward neural nets; geographic information systems; laser ranging; mobile agents; multilayer perceptrons; GIS; GPS; Global Positioning System; MIND; Mechanism for Intelligent Neural network based Driving system; automated intelligent transport systems; backpropagation learning; frontal impact collision vectors; geographic information system; lane-changing maneuvers; laser ranging; mobile agent; multilayer feedforward neural network; network topologies; processor overhead; selective net masking; side impact collision vectors; turns; Backpropagation; Feedforward neural networks; Geographic Information Systems; Global Positioning System; Intelligent networks; Intelligent systems; Mobile agents; Multi-layer neural network; Neural networks; Optical design;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
E-Tech 2004
Print_ISBN :
0-7803-8655-8
Type :
conf
DOI :
10.1109/ETECH.2004.1353844
Filename :
1353844
Link To Document :
بازگشت