Loading...

Teaching to Point at different Objects as an Interactive Gesture to Robot by Learning from Demonstration

Razmjoofard, Amir Reza | 2020

541 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 53193 (08)
  4. University: Sharif University of Technology
  5. Department: Mechanical Engineering
  6. Advisor(s): Meghdari, Ali; Taheri, Alireza
  7. Abstract:
  8. The usage of robots as our friends has been proliferated these days. Knowing that they are going to be used in ordinary houses, we should develop methods and algorithms in order to provide a situation for end-users to program their own robots for their desired tasks. Learning from Demonstrations (LfD) can play a crucial role in this field. In this study, we had taught a non-verbal communication method (pointing) to a robot utilizing LfD. The learning method used was TP-GMM1. The rationale to use this method was that it models all the degrees of freedom together, and we thought it might be an essential parameter to make a movement more natural and understandable which could be two vital parameters for pointing. However, the conventional type of TP-GMM cannot utilize non-linear features that were essential in pointing the scenario. In this regard, we have developed a method to exploit non-linear feature among linear ones with TP-GMM. After testing the method in the simulation phase and getting promising results, it is tested on a real robot in the pointing scenario. Finally, to study the opinions of ordinary people about the naturalness and understandability of the movement, they are asked to fill a questionnaire which was attached with videos of 6 actions; three of the videos were reproduced by our method, while the other 3 were reproduced by a simple Inverse Kinematic (IK) method. The IK method is used only to be able to compare our method and evaluate how people fill the questionnaire. 72 people participated in this project, and the results show that people have found the actions which were reproduced by our method natural and understandable. Also, according to people’s opinions, the difference between our method and the simple method was significant. This comparison was not the goal of this project and is done to take our primary steps for future investigations
  9. Keywords:
  10. Human Robot Interaction (HRI) ; Social Robotics ; Understandability ; Nonlinear Feature ; Learning from Demonstration (LfD) ; Naturalness ; Task Parametrized Gaussian Mixture Models (TP-GMM) ; Non-Verbal Communication

 Digital Object List

 Bookmark

No TOC