Osman, A. A. E., R. A. El-Khoribi, M. E. Shoman, and W. M. A. Shalaby, "Trajectory Learning Using Principal Component Analysis", World Conference on Information Systems and Technologies, 2017. Abstract

Robots are increasingly used in numerous life applications. Therefore, humans are looking forward to create productive robots. Robot learning is the process of obtaining additional information to accomplish an objective configuration. Moreover, robot learning from demonstration is to guide the robot the way to perform a particular task derived from human directions. Traditionally, modeling the demonstrated data was applied on discrete data which would result in learning outcome distortions. So as to overcome such distortion, preprocessing of the raw data is necessary. In this paper, trajectory learning from demonstration scheme is proposed. In our proposed scheme, the raw data are initially preprocessed by employing the principal component analysis algorithm. We experimentally compare our proposed scheme with the most recent proposed schemes. It is found that the proposed scheme is capable of increasing the efficiency by minimizing the error in comparison to the other recent work with significant reduced computational cost.

Osman, A. A. E., R. A. El-Khoribi, M. E. Shoman, and W. M. A. Shalaby, "Trajectory learning using posterior hidden Markov model state distribution", Egyptian Informatics Journal, 2017. AbstractWebsite

Many life applications are extremely depending on using the robots, thus the human are seeking to develop efficient robots. Robot learning is to acquire extra knowledge in order to achieve objective configuration. In addition, robot learning from demonstration is about teaching the robot how to do specific task by the guidance of the human. Till now, learning from demonstration depends on discrete data which may cause distortion in the learning outcome. So that, preprocessing phase for the data is necessarily to handle this distortion. In this paper, we propose a new scheme for generating a generalized trajectory by employing set of demonstrated trajectories. Such that preprocessed data is used initially instead of the raw data, the preprocessing is done using posterior hidden Markov model state distribution. The rest of the model is based on set of key points identified for each demonstration. Our proposed scheme is experimentally compared to the previous works. The results show that our proposed scheme is able to reduce the error in comparison to other recent schemes with insignificant added computational cost.