Robot learning by a mining tunnel inspection robot

Abstract

Learning from Demonstration (LfD) is a method of teaching an agent a task by a number of suitable demonstrations. The agent will then perform the task without any further supervision. In this paper, Discrete Hidden Markov Model (DHMM) is applied to train a robot for a mining inspection task. An initial training method based on the Gaussian Mixture Model (GMM) was developed and is compared to DHMM. Results show that the learning speed based on DHMM is faster than the one for GMM and DHMM may prove to be more suitable for the mining inspection task under consideration. The proposed method has already been implemented, and some important problems on implementation are discussed.