Sensor Counter Approach for a Mobile Robot to Navigate a Path Using Programming by Demonstration

Abstract This paper presents a sensor counter approach for a mobile robot to navigate a path using programming by demonstration. In this paper a hybrid method which uses sensor values and counter values as path variables has been proposed, in order to avoid the dynamic obstacle in the environment we propose an obstacle avoidance algorithm (OAA) which is merged with hybrid method. Proposed method has been implemented and tested in a mobile robot platform AAMoR-1. Experiments has been done in an real time test environment to find the accuracy of the robot, the robot has been taught a particular trajectory by a human operator and later changing the robot to autonomous mode where in it moves in the same trajectory as taught, obstacle avoidance algorithm has also been tested by adding an obstacle in the taught path. Good test results have been obtained.

[1]  Jean-Claude Latombe,et al.  An Approach to Automatic Robot Programming Based on Inductive Learning , 1984 .

[2]  Stephen A. Billings,et al.  Task identification and characterisation in mobile robotics through non-linear modelling , 2007, Robotics Auton. Syst..

[3]  Stephen A. Billings,et al.  Robot learning through task identification , 2006, Robotics Auton. Syst..

[4]  Daniel C. Halbert,et al.  Programming by Example , 2010, Encyclopedia of Machine Learning.

[5]  Akihisa Ohya Human robot interaction in mobile robot applications , 2002, Proceedings. 11th IEEE International Workshop on Robot and Human Interactive Communication.

[6]  Stephen A. Billings,et al.  Programming Mobile Robots by Demonstration through System Identification , 2007, EMCR.

[7]  T. S. B. Sudarshan,et al.  Mobile Robot Programming by Demonstration , 2011, 2011 Fourth International Conference on Emerging Trends in Engineering & Technology.

[8]  Dermot Kerr,et al.  Towards Automated Code Generation for Autonomous Mobile Robots , 2010, AGI 2010.