Stable encoding of robot paths using normalised radial basis networks: application to an autonomous wheelchair

Abstract

A Neural Network (NN) using Normalised Radial Basis Functions (NRBF) is used for encoding the sequence of positions forming the path of an autonomous wheelchair. The network operates by continuously producing the next position for the wheelchair. As the path passes several times over the same point, additional phase information is added to the position information. This avoids the aliasing problem. The use of normalised RBFs creates an attraction field over the whole space and enables the wheelchair to recover from perturbations, e.g. due to the avoidance of people.