A Humanoid Robot’s Effortful Adaptation Boosts Partners’ Commitment to an Interactive Teaching Task

This work presents the preliminary results of an ongoing project which aims to use humanoid robots as sign language tutors. The study mainly focuses on children who have some problems when they are communicating with other individuals such as hearing-impaired or autistic children. In this paper, an interactive game, which is based on sign language, between a humanoid robot and a human participant is introduced. The game consists of an imitation based learning phase where the signs are taught in the first step and they are tested in the second step within the frame of an interaction game. The goal of the interactive game is to reinforce the semantic meaning of the signs in a motivating and engaging way, as well as to test the learning performance of the participants. We aim to design a comfortable learning environment by using the humanoid robot as an educational medium. The game also improves the participants' imitation and turn-taking skills and teaches the semantic meanings of the signs.

Short et al. [10] found that in a game between a human participant and a humanoid robot, the participant will perceive the robot as being more agentic and as having more intentionality if it cheats than if it plays without cheating. However, in that design, the robot that actively cheated also generated more motion than the other conditions. In this paper, we investigate whether the additional movement of the cheating gesture is responsible for the increased agency and intentionality or whether the act of cheating itself triggers this response. In a between-participant design with 83 participants, we disambiguate between these causes by testing (1) the cases of the robot cheating to win, (2) cheating to lose, (3) cheating to tie from a winning position, and (4) cheating to tie from a losing position. Despite the fact that the robot changes its gesture to cheat in all four conditions, we find that participants are more likely to report the gesture change when the robot cheated to win from a losing position, compared with the other conditions. Participants in that same condition are also far more likely to protest in the form of an utterance following the cheat and report that the robot is less fair and honest. It is therefore the adversarial cheat itself that causes the effect and not the change in gesture, providing evidence for a cheating detector that can be triggered by robots.

Gestures and speech modalities play potent roles in social learning, especially in educational settings. Enabling artificial learning companions (i.e., humanoid robots) to perform human-like gestures and speech will facilitate interactive social learning in classrooms. In this paper, we present the implementation of human-generated gestures and speech on the Pepper robot to build a robotic teacher. To this end, we transferred a human teacher gesture to a humanoid robot using a web and a kinect cameras and applied a video-based markerless motion capture technology and an observation-based motion mirroring method. To evaluate the retargeting methods, we presented different types of a humanoid robotic teacher to six teachers and collect their impressions on the practical usage of a robotic teacher in the classroom. Our results show that the presented AI-based open-source gesture retargeting technology was found attractive, as it gives the teachers an agency to design and employ the Pepper robot in their classes. Future work entails the evaluation of our solution to the stakeholders (i.e. teachers) for its usability.KeywordsMotion retargetingTeacher robotNon-verbal behaviorMotivationPepper robot

This work presents an enriched human-machine interface for performing effective tele-manipulation using a humanoid robot. To provide the slave with the ability to dexterously interact with the remote environment, we implement an autonomous impedance controller that regulates the slave's joints stiffness and damping according to the manipulation loading conditions. In the proposed strategy, free-space operations are performed with compliant limbs to ensure safe interactions during unforeseen collisions on the whole-arm together with a soft behaviour at the initial phase of contact. During the manipulation task, the designed controller accounts for the arm configuration and the direction/amplitude of the external force sensed at the end-effector to stiffen the humanoid arm joints, permitting to handle the task loads. Experiments performed on the compliant humanoid robot COMAN demonstrate the effectiveness of this human-inspired impedance controller during a teleoperated pick-and-place task. The interaction forces during manipulation can destabilize the floating base humanoid robot. To address this concern we also investigate the operators ability to rely on a cutaneous feedback of the slave's balance state to adjust their teleoperation strategy when required so as to complete the task while maintaining the slave's balance. A comparative study shows that the proposed vibrotactile feedback allows for a proper recognition of the slave's drop of stability during interaction tasks and that the tactile guidance leads to enhanced teleoperation performances characterized by a significantly lower number of falls.

Applications based on artificial intelligence and machine learning are becoming more popular in teaching learning. Advanced technologies have facilitated robots to carry out various human-like functions, which have navigated the interest of educators to discover the role of robots as potential teachers, instructors, or teaching assistants in education. An extensive search for articles for humanoid robots and education either in the title or keywords was done utilizing PubMed, Google Scholar and Web of Science data sets. The tracking terms were artificial intelligence, education, medical education, anatomy, robots, humanoid robots, teaching, teaching assistant and tutor. The usage of artificial intelligence in the form of humanoid robots is quite common. However, literature citing its usage in medical education is rare. Humanoid robots as a teacher or teaching assistants are predominantly used in learning foreign languages. Primarily, a humanoid robot can discharge five functions as a potential teacher. Humanoid robots can effectively fulfil numerous educational goals in medicine since they can replicate human responses, work relentlessly regardless of students' repeated mistakes, be loaded with innovative teaching methodologies, and be upgraded with more current information. As a subject of medicine, anatomy is highly visual; therefore, constant endeavors have been initiated to develop technology-enhanced learning over the decades. Although artificial intelligence in humanoid robots has been successfully used in primary education and in learning a foreign language, its scope as an anatomy teacher or teaching assistant is a new and unique idea that needs exploration.

(also presented at ICRA'19)

Aiming at the problem of calibration in humanoid robot, this paper presents a calibration method combining online and offline process. According to the constraint condition of the double support phase in the bipedal walking, we define the calibration problem as a nonlinear least squares problem and use the Levenberg-Marquardt method to solve the problem. We use NAO humanoid robot to collect data online and then the calibration process is performed offline on the local computer, so it does not affect the normal movement of the robot and improves the adaptability of the robot calibration.

To facilitate natural interactions between humans and embodied conversational agents (ECAs), we need to endow the latter with the same nonverbal cues that humans use to communicate. Gaze cues in particular are integral in mechanisms for communication and management of attention in social interactions, which can trigger important social and cognitive processes, such as establishment of affiliation between people or learning new information. The fundamental building blocks of gaze behaviors are gaze shifts : coordinated movements of the eyes, head, and body toward objects and information in the environment. In this article, we present a novel computational model for gaze shift synthesis for ECAs that supports parametric control over coordinated eye, head, and upper body movements. We employed the model in three studies with human participants. In the first study, we validated the model by showing that participants are able to interpret the agent’s gaze direction accurately. In the second and third studies, we showed that by adjusting the participation of the head and upper body in gaze shifts, we can control the strength of the attention signals conveyed, thereby strengthening or weakening their social and cognitive effects. The second study shows that manipulation of eye--head coordination in gaze enables an agent to convey more information or establish stronger affiliation with participants in a teaching task, while the third study demonstrates how manipulation of upper body coordination enables the agent to communicate increased interest in objects in the environment.

To better prepare future generations, knowledge about computers and programming are one of the many skills that are part of almost all Science, Technology, Engineering, and Mathematic programs; however, teaching and learning programming is a complex task that is generally considered difficult by students and teachers alike. One approach to engage and inspire students from a variety of backgrounds is the use of educational robots. Unfortunately, previous research presents mixed results on the effectiveness of educational robots on student learning. One possibility for this lack of clarity may be because students have a wide variety of styles of learning. It is possible that the use of kinesthetic feedback, in addition to the normally used visual feedback, may improve learning with educational robots by providing a richer, multi-modal experience that may appeal to a larger number of students with different learning styles. It is also possible, however, that the addition of kinesthetic feedback, and how it may interfere with the visual feedback, may decrease a student’s ability to interpret the program commands being executed by a robot, which is critical for program debugging. In this work, we investigated whether human participants were able to accurately determine a sequence of program commands performed by a robot when both kinesthetic and visual feedback were being used together. Command recall and end point location determination were compared to the typically used visual-only method, as well as a narrative description. Results from 10 sighted participants indicated that individuals were able to accurately determine a sequence of movement commands and their magnitude when using combined kinesthetic + visual feedback. Participants’ recall accuracy of program commands was actually better with kinesthetic + visual feedback than just visual feedback. Although the recall accuracy was even better with the narrative description, this was primarily due to participants confusing an absolute rotation command with a relative rotation command with the kinesthetic + visual feedback. Participants’ zone location accuracy of the end point after a command was executed was significantly better for both the kinesthetic + visual feedback and narrative methods compared to the visual-only method. Together, these results suggest that the use of both kinesthetic + visual feedback improves an individual’s ability to interpret program commands, rather than decreases it.

Gaze plays an important role in everyday communication between humans. Eyes are not only used to perceive information during interaction, but also to control it. Humanoid robots on the other hand are not yet very proficient in understanding and using gaze. In our study we enabled two humanoid robots to perceive and exert gaze actions. We then performed a pilot experiment with the two android robots playing the “Wink Murder” game with human players. We demonstrate that the designed framework allows the robots to complete the game successfully, validating the efficacy of our gaze tracking system. Moreover, human participants exhibited a rich variety of natural behaviors in the game, suggesting that it could represent a valid scenario for a more in-depth investigation of human-humanoid interaction.

A phase I trial is the first research study of an experimental agent in human participants and includes the first-in-human study after successful preclinical testing. The primary objective of the phase I trials is to establish the safety and tolerability of the drug in a relevant disease population and to make a dosage recommendation in a subsequent phase II trial. For cytotoxic drugs, the recommended phase II dosage is almost synonymous with the maximum tolerated dose under the assumption that the higher the dose, the more efficacious the agent. This assumption has been challenged recently with the emergence of non-cytotoxic drugs, such as molecularly targeted agents and immunotherapies. In this chapter, an overview of phase I trial designs is presented in three categories: rule-based, model-based, and model-assisted designs. Dose optimization strategies will be discussed at the conclusion of the chapter.

In this paper we present a modular and distributed control architecture in order to achieve natural interaction and mobile manipulation task goals for a humanoid robot. We propose a hierarchically organized architecture with three levels and introduce the mapping of the functional features in this architecture into hardware and software modules. We also describe different functional features which have been realized and integrated into the whole control architecture.

The purpose of this study is to present the experiential learning in the development of a Dynamic Anthropomorphic Robot with Intelligence(DARwIn)-High Performance(HP) at Robotics and Mechanism Laboratory(RoMeLa). DARwIn-HP, an miniature-sized humanoid platform, has been developed by self-directed undergraduate students of RoMeLa for robotics in education. They, as the target consumer of educational robots, contributed to the design and manufacturing of the DARwIn-HP by sharing their thoughts on the necessary functionality of an educational humanoid robot. This hands-on experience allows them to understand the fundamental materials of a humanoid robot in theory as well as the usage of mechanical tools in practice. All undergraduates after experiencing the development process also recognize the importance as well as the interests on learning the subjects in the engineering curriculum. As of the final process and post-activity of robot development, encouraging all students to participate in extracurricular activities provides them with a chance of evaluating their work. There are positive effects to engineering students such as enriching their competitiveness for a range of challenges facing society.

Humanoid robots find application in human-robot interaction tasks. However, despite their capabilities, their sequential computing system limits the execution of computationally expensive algorithms such as convolutional neural networks, which have demonstrated good performance in recognition tasks. As an alternative to sequential computing units, Field-Programmable Gate Arrays and Graphics Processing Units have a high degree of parallelism and low power consumption. This study aims to improve the visual perception of a humanoid robot called NAO using these embedded systems running a convolutional neural network. The methodology adopted here is based on image acquisition and transmission using simulation software: Webots and Choreographe. In each embedded system, an object recognition stage is performed using commercial convolutional neural network acceleration frameworks. Xilinx® Ultra96™, Intel® Cyclone® V-SoC and NVIDIA® Jetson™ TX2 cards were used, and Tinier-YOLO, AlexNet, Inception-V1 and Inception V3 transfer-learning networks were executed. Real-time metrics were obtained when Inception V1, Inception V3 transfer-learning and AlexNet were run on the Ultra96 and Jetson TX2 cards, with frame rates between 28 and 30 frames per second. The results demonstrated that the use of these embedded systems and convolutional neural networks can provide humanoid robots such as NAO with greater visual recognition in tasks that require high accuracy and autonomy.

“Critical thinking, Communication, Collaboration, Creativity in kindergarten with Educational Robotics”: A scoping review (2012–2023)