Maria Vanessa aus der Wieschen

Teleoperation for learning by demonstration: Data glove versus object manipulation for intuitive robot control

Learning by demonstration is a useful technique to augment a robots behavioral inventory, and teleoperation allows lay users to demonstrate novel behaviors intuitively to the robot. In this paper, we compare two modes of teleoperation of an industrial robot, the demonstration by means of a data glove and by means of a control object (peg). Experiments with 16 lay users, performing assembly task on the Cranfield benchmark objects, show that the control peg leads to more success, more efficient demonstration and fewer errors.

A Comparison of Types of Robot Control for Programming by Demonstration

Programming by Demonstration (PbD) is an efficient way for non-experts to teach new skills to a robot. PbD can be carried out in different ways, for instance, by kinesthetic guidance, teleoperation or by using external controls. In this paper, we compare these three ways of controlling a robot in terms of efficiency, effectiveness (success and error rate) and usability. In an industrial assembly scenario, 51 participants carried out peg-in-hole tasks using one of the three control modalities. The results show that kinesthetic guidance produces the best results. In order to test whether the problems during teleoperation are due to the fact that users cannot, like in kinesthetic guidance, switch between control points using traditional teleoperation devices, we designed a new device that allows users to switch between controls for large and small movements. A user study with 15 participants shows that the novel teleoperation device yields almost as good results as kinesthetic guidance.

Intuitive error resolution strategies during robot demonstration

While robot learning from demonstration comes with great benefits [5], the intuitive interaction between naïve users and robots also poses challenges. For instance, users need to be prevented from causing damage and to be enabled to recover from errors. We studied the error resolution strategies of 28 lay users performing simple assembly tasks via teleoperation of a robotic arm in order to gain insight into the strategies users take. The two most common problems are too much pressure and singularity. Even though users were provided with instructions on how to undo singularity in an instruction video, they did not always recover successfully. In contrast, too much pressure, if noticed, was resolved mostly correctly by lifting the peg or by letting it drop into the hole rather than inserting it. Finally, users were quite clueless about how to resolve self-collision and over-rotation.

Divergent Language Choices and Maintenance of Intersubjectivity: The Case of Danish EFL Young Learners

ABSTRACT The role of students’ first language(s) in foreign language classrooms has been hotly debated in the last decades. Although this line of research has advanced our understanding of language choice in the L2 classroom, it has mostly dealt with adolescent and adult learners. From a contextual perspective, then, more micro-analytic research that focuses on language choice at the primary school level is needed. Against this background, this paper presents a case study of a Danish third-grade English as a foreign language classroom, in which a pattern of divergent language choices has been observed: the teacher consistently uses English, whereas the learners almost exclusively speak Danish, which might entail trouble in maintaining intersubjectivity and a joint pedagogical focus. Using Conversation Analysis methodology, we found two sequential formats that help ensure student understanding and thus maintain intersubjectivity: (1) learner translations and reformulations for peer support in expansion sequences, and (2) expansions initiated by students requesting information or clarification that display partial or no understanding. We argue that the sense-making practices co-constructed in this classroom context are possible because the teacher encourages shared multilingual meaning-making practices. This research has implications for teaching EFL to young learners, and classroom language policies.

Error Feedback for Robust Learning from Demonstration

The present study applies a user-centered approach to investigating feedback modalities for robot teleoperation by naïve users. It identifies the reasons why novice users need feedback and evaluates feedback modalities by employing participatory design. Moreover, drawing on document design theory, it studies which design guidelines need to be followed in the creation of legible error feedback screens.

A novel tele-operation device allowing for dynamic switching between control points during learning from demonstration

In an industrial setup it is of great importance to facilitate easy and fast robot teaching in order to meet the increasing demands from industry for high flexibility and efficiency. This paper presents a novel tele-operation device for teaching assembly tasks to robots.