Line-Crawling Robot Navigation: A Rough Neurocomputing Approach
James Francis Peters, Tae-Chon Ahn, Maciej Borkowski, V. Degtyaryov, Sheela Ramanna · Studies in fuzziness and soft computing · 2003
This chapter considers a rough neurocomputing approach to the design of the classify layer of a Brooks architecture for a robot control system. This paradigm for neurocomputing that has its roots in rough set theory, works well in cases where there is uncertainty about the values of measurements used to make decisions. In the case of the line-crawling robot (LCR) described in this chapter, rough neurocomputing works very well in classifying noisy signals from sensors. The LCR is a robot designed to crawl along high-voltage transmission lines where noisy sensor signals are common because of the electromagnetic field surrounding conductors. In rough neurocomputing, training a network of neurons is defined by algorithms for adjusting parameters in the approximation space of each neuron. Learning in a rough neural network is defined relative to local parameter adjustments. Input to a sensor signal classifier is in the form of clusters extracted from convex hulls that “enclose” similar sensor signal values. This chapter gives a fairly complete description of a LCR that has been developed over the past three years as part of a Manitoba Hydro research project. This robot is useful in solving maintenance problems in power systems. A description of the locomotion features of a line-crawling robot and the basic architecture of a rough neurocomputing system for robot navigation are given. A brief description of the fundamental features of rough set theory used in the design of a rough neural network is included in this chapter. A sample sensor signal classification experiment using a recent implementation of rough neural networks is also given. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.