Michael Popp

Real-time framework for multimodal human-robot interaction

This paper presents a new framework for multimodal data processing in real-time. This framework comprises modules for different input and output signals and was designed for human-human or human-robot interaction scenarios. Single modules for the recording of selected channels like speech, gestures or mimics can be combined with different output options (i.e. robot reactions) in a highly flexible manner. Depending on the included modules, online as well as offline data processing is possible. This framework was used to analyze human-human interaction to gain insights on important factors and their dynamics. Recorded data comprises speech, facial expressions, gestures and physiological data. This naturally produced data was annotated and labeled in order to train recognition modules which will be integrated into the existing framework. The overall aim is to create a system that is able to recognize and react to those parameters that humans take into account during interaction. In this paper, the technical implementation and application in a human-human and a human-robot interaction scenario is presented.

The Influence of Telemanipulation-Systems on Fine Motor Performance

Extravehicular activities (EVAs) are a hazardous and expensive procedural method to operate in outer space. A possible support or alternative for manned missions in terms of on-orbit servicing are telemanipulation-systems. Whether or not such systems can actually achieve the efficiency of suited astronauts remains a central issue in telemanipulation research. Both scenarios, extravehicular activities as well as telemanipulation-systems, are restricted by different environmental factors, especially in terms of tasks that require fine motor skills. For suited astronauts, different factors, such as restricted mobility and reduced tactile feedback through the gloves, as well as a restricted field of view, impair fine motor skills. On the other hand, time delay, limited degrees of freedom and restricted haptic and visual feedback are amongst the factors, which may cause impairment of performance during the work with telemanipulation-systems. In order to compare the efficiency of both scenarios, a testbed equipped with typical mounting tasks was developed. An experimental study showed that the testbed is a valid measure of fine motor skills. In two follow-up studies, the influence of some factors debilitating fine motor performance in telemanipulation-systems and simulated extra-vehicular activities was analysed and compared.

Emotion induction during human-robot interaction

Summary form only given. The aim of the presented study was to measure physiological correlates of emotions that are of particular interest in the field of human-robot interaction (HRI). Therefore, we did not focus on self-induced basic emotions but rather evoked states that might occur naturally in this context. Our video shows how such states (namely stress, boredom, surprise, and perplexity) were elicited during a joint construction task with an industrial robot (see figure 1). Participants were asked to build different LEGO objects, while the robot arm was passing the bricks with predetermined velocity. States of stress and boredom were generated by varying the handover interval from 3 seconds (stress) to 5 seconds (normal working condition) up to 35 seconds (boredom). Surprise was induced by passing an unexpected component. At the end of the experiment, we additionally wanted to know how people react if the robot seems to tease them by repeatedly changing the handover position. This experiment was realized by the support of researchers from the MMK and the IWB of the Technical University Munich who provided the technical facilities and know-how. The underlying project is supported within the DFG excellence initiative research cluster “Cognition for Technical Systems - CoTeSys”, see also www.cotesys.org.

Physiology and HRI: Recognition of over- and underchallenge

Contrary to common emotion recognition techniques by face or speech analysis, physiological data are involuntary and continuously available. Thus, they allow for emotion detection even in situations without spoken words or in case of non-extreme emotions, which are more likely to occur in human-robot interaction (HRI). In this paper, we describe the results of an experiment investigating non-extreme emotional states relevant for HRI scenarios (over- and underchallenge). Those states occurred naturally during the course of a LEGO construction task by manipulating working speed. Data collected from 28 subjects were analyzed and the results of different types of discriminant analysis and nearest neighbour methods were compared. Based on two physiological modalities (HR, SCR), correct classification rates of up to 76% for seven features and 74% for only two features were achieved. Overchallenge could be discriminated very well from the other two conditions (96.4 - 85.7%), whereas underchallenge is often confused with the intermediate condition with normal working speed.

High-fidelity telepresence and teleaction

The collaborative research center SFB453 (www.sfb453.de) aims to realize high-fidelity telepresence and teleaction systems.