Metaclassifiers for predicting the robotic navigational performance

Abstract

Prediction of robot steps can be used in path exploring problems and application of data mining techniques, enhances navigational direction of robots. In this paper the proposed method is validated on the data sets using multi classification algorithms with four types of movement classes like Action-ahead, Sharp right turn, small in degree-right turn, and small in degree left turn in a separate layer. We obtain the results based on Meta classifiers’ accuracies tabulated. A layered approach is followed for obtaining the more accurate multi-classification.