No abstract

Robotics and Artificial Intelligence (AI) have always been among the most popular topics in science fiction (sci-fi) movies. This paper endeavors to review popular movies containing Fictional Robots (FR) to extract the most common characteristics and interesting design ideas of robots portrayed in science fiction. To this end, 134 sci-fi films, including 108 unique FRs, were investigated regarding the robots’ different design aspects (e.g., appearance design, interactive design and artificial intelligence, and ethical and social design). Also, in each section of this paper, some characteristics of FRs are compared with real social robots. Since some researches point to the significant role of the cinema in forming the community’s expectations, it is very important to consider these characteristics and differences in choosing the future pathway of robotics. As some examples of findings, we have found that unlike the non-metallic skins/covers of real social robots, most FRs are still covered by highly detailed metal components. Moreover, the FR ability of interactions are generally (more than 90%) shown to be similar or even more advanced than normal Human–Human interactions, and this milestone was achieved by ignoring the AI challenges of real HRI. On the other hand, the ethical aspects of movies do inspire us to consider the potential ethical aspects of real robot design. All in all, according to popularity of movies, studying FR could be a step toward more appropriate development of robotics and AI entities to be accepted by general users in the real world. Highlights: We reviewed 134 sci-fi movies containing 108 unique fictional robots regarding different design aspects. Fictional Robot (FR) is an artificial entity acting as a result of a fictional technology and playing a role in a movie. Investigating fictional robots can shed light on the development of real robotics and AI entities.

This paper addresses the lack of proper Learning from Demonstration (LfD) architectures for Sign Language-based Human–Robot Interactions to make them more extensible. The paper proposes and implements a Learning from Demonstration structure for teaching new Iranian Sign Language signs to a teacher assistant social robot, RASA. This LfD architecture utilizes one-shot learning techniques and Convolutional Neural Network to learn to recognize and imitate a sign after seeing its demonstration (using a data glove) just once. Despite using a small, low diversity data set (~ 500 signs in 16 categories), the recognition module reached a promising 4-way accuracy of 70% on the test data and showed good potential for increasing the extensibility of sign vocabulary in sign language-based human–robot interactions. The expansibility and promising results of the one-shot Learning from Demonstration technique in this study are the main achievements of conducting such machine learning algorithms in social Human–Robot Interaction.

No abstract

No abstract