United States

This work presents a method for recognition of walking activities and prediction of gait periods using wearable sensors. First, a Convolutional Neural Network (CNN) is used to recognise the walking activity and gait period. Second, the output of the CNN is used by a Predicted Information Gain (PIG) method to predict the next most probable gait period while walking. The output of these two processes are combined to adapt the recognition accuracy of the system. This adaptive combination allows us to achieve an optimal recognition accuracy over time.

IEEE RO-MAN 2021

Parents’ Views on Using Social Robots for Language Learning

wearable sensors

assistive robotics

activity recognition

RO-MAN 2021's theme is trustworthiness and cooperation. We chose this term well before the pandemic put the world on edge and added a new reason for us to question our own personal safety when in the company of others. But what about in the company of social robots? What would cause us to trust, or not trust, them? How do we as researchers create that feeling, that critical component of any relationship? With social robots, it is of course largely in our power to design the physical form, the control code and the interface software to maximize trustworthiness, facilitate cooperation and enhance the quality of human-robot interaction. This theme also touches upon people's expectations and perceptions of robots, as well as other social, cognitive and cultural factors that will influence human-robot interactions. We will have a special session/track on this topic, invite targeted submissions and include plenary speakers who are specialist in this area. We sincerely hope that this focus will help enlighten the design-thinking of all of us as we fabricate and program our own various systems.

To access RO-MAN Virtual 2021, [please register here.](https://ro-man2021.org/registration/)

RO-MAN 2021's theme is trustworthiness and cooperation. We chose this term well before the pandemic put the world on edge and added a new reason for us to question our own personal safety when in the company of others. But what about in the company of social robots? What would cause us to trust, or not trust, them? How do we as researchers create that feeling, that critical component of any relationship? With social robots, it is of course largely in our power to design the physical form, the control code and the interface software to maximize trustworthiness, facilitate cooperation and enhance the quality of human-robot interaction. 

panel

This paper proposes a street-style study method to conduct human-robot interaction studies in-the-wild where the robots conduct their own experiments by recruiting their audience, conducting the study and gathering data. This street-style study method was implemented using multi-robot comedy as the platform deployed at an arts and technology festival to validate the concept. Twelve robot comedy shows occurred over seven hours with two robots on stage, who queried the audience during and at the end of each show. The multi-robot aspect enabled the robots to act out interactions relative each other, oneself and the audience. The final street-style study method evolved from pilots at a local farmer’s market, with hardware designed for portability and easy replication. The robots conducted their own human experiments in that they queried the audience after displaying experimentally balanced episodes of relational humor, with permutations of who was the `butt of the joke.' Our study results explore the relational humor of the two robots and the audience, asking the audience to agree or disagree with particular perspectives. Delivered as part of the show, the robots invite the audience to vote via a show of hands. ANOVA analyses of the percent-agreement results find that (1) audiences were generally positive about all aspects of the show unless both robots were being negative, and (2) audiences were more ready to protect the robot comedian's ego than their own, strongly supporting the statement that the robot was doing a good job.

Augmented Reality interface to verify Robot Learning

We try to design a control law using MPC and predictive tracking of a person using LSTM for a socially adept person following robot. The human-robot shared autonomy is the motivation. Tracking a person when it goes out of sight is a challenging task, for which LSTMs are brought to the picture. This adds the natural human part to the work.

Towards Out-Of-Sight Predictive Tracking for Long-Term Indoor Navigation of Non-Holonomic Person Following Robot

When we humans interact with each other, we adapt our behaviors to the people we interact with. We do this by interpreting subtle cues and signals from them in each interaction. With this paper, we created a robot that uses speech and body movement characteristics to recognize and distinguish between users interacting with it through small duration interactions. We investigate how these characterics are influenced by changing the physical context.

A Minimalistic Approach to User Group Adaptation of Robot Behaviors Using Movement and Speech Analysis

This research has developed a new method for predicting natural human arm movements. It introduces the five-parameter logistic function into this application space and uses supervised learning to predict fingertip trajectories and elbow swivel angles. The method presents small error in an experimental study, improving upon previous methods and verifying its effectiveness.

Human Arm Motion Prediction in Reaching Movements

We present a 3D human activity simulator, ``HOIsim", focusing on Human-Object Interactions (HOIs). Using HOIsim we model two complex activities for ``lunch" and ``breakfast". Furthermore, we introduce methods to meaningfully randomize activity flows and the environment topology, and make a dataset of the 3D pose maps publicly available.

HOIsim: Synthesizing Realistic 3D Human-Object Interaction Data for Human Activity Recognition

Human-human collaboration is characterized by a
back-and-forth, where an action of one agent elicits the response
of the other. This interaction is inherently multimodal and
includes both high-level modalities such as language and lowlevel ones such as force exchanges. In this work, we investigate
human collaborative manipulation: we show distinct patterns
that can be identified in low-level physical data and that can
be interpreted as primitives used by humans to negotiate about
various aspects of the motion and to execute the motion. These
primitives provide a high-level interpretation of the interaction
and can be used to connect low-level behavior to language.
We describe the human study used to collect the data, the data
analysis process, and discuss how the identified primitives could
be used by a robot’s interaction manager to mediate physical
Human-Robot Interaction (pHRI).

Physical Action Primitives for Collaborative Decision Making in Human-Human Manipulation

In this research, we investigate the use of External Human Machine Interface for delivery robots for expressing their navigational intent. Through a survey, we explore the effectiveness of display and light-based external interfaces in expressing the delivery robots navigational intent.

Understanding robot acceptance/rejection: the SAR model

Social robots have strong potential in health and social care, but the embodiment of these devices is important to ensure acceptability and adoption. Based on a reflection over four studies, we provide insight into the impact of various morphology types. Our results suggest the combination of familiar, zoomorphic appearance with animacy, life-simulation and speech capabilities was an interesting area of research for future robots developed for health and social care.

Morphology of Socially Assistive Robots for Health and Social Care: A Reflection on 24 Months of Research with Anthropomorphic, Zoomorphic and Mechanomorphic Devices

Can a verbal message by a robot change your attitude? More importantly, can it worsen your attitude? Our research study highlights how speech-based appeals
by a robot can successfully influence individuals to trust it with a valuable asset, i.e. their mental health. However, while an emotional appeal by a robot is more effectual
on affective attitudes, it backfires significantly more than a logical appeal when used on cognitive attitudes, leading to trust dissolution. It follows that a robot can, in fact,
successfully arouse emotions in order to persuade. In theory, robots may have an upper hand in executing an effective persuasion attempt compared to humans, who inherently use emotions when they have the intention to persuade.

Would You Trust a Robot with Your Mental Health? The Interaction of Emotion and Logic in Persuasive Backfiring

In this paper, we conduct a full meta-analysis of the existing literature on personality similarity versus dissimilarity (matching/mismatching) and attempt to come to a consensus on if robots with matching personalities are accepted more readily than those with mismatching personalities. In addition, we provide several gaps in this literature worth pursuing.

Premium content

Parents’ Views on Using Social Robots for Language Learning

Downloads

Next from IEEE RO-MAN 2021

Augmented Reality interface to verify Robot Learning

Similar lecture

Effect of Robot Agents on Teaching against Pseudoscience

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES