|Sept 09||13:30 - 14:00||Welcome and Introduction|
|14:00 - 15:30||S. Biundo-Stephan: Introduction to Companion-Technology
Companion-Technology aims to enable the implementation of arbitrary technical systems as Companion-Systems - systems that smartly adapt their functionality to a user’s individual requirements. They comply with his or her abilities, preferences, and current needs and adjust their behavior as soon as critical changes of the environment or changes of the user’s emotional state or disposition are observed. Companion-Systems show particular characteristics such as competence, individuality, adaptability, availability, cooperativeness, and trustworthiness, which are realized by integrating the technical functionality of systems with a combination of cognitive processes. The lecture introduces the essence of Companion-Technology. It shows how the adaptive functionality of technical system is implemented by exploiting AI planning technology and presents an example Companion-System.
|15:30 - 16:00||Coffee Break|
|16:00 - 18:00||S. Schwenker/S. Meudt/G. Palm: Data Analysis and Machine Learning for Multimodal Affect Recognition
In this seminar we introduce automatic affect recognition based on multimodal sensory data. In the first part we introduce learning paradigms, classifiers, and information fusion architectures for the challenging task of multimodal affect recognition, particularly the focus is on classifier fusion, semi-supervised learning, and the extraction of discriminative features from audio, video, pose and bio-signals. In this context we introduce our ATLAS framework, an efficient tool to analyze and annotate multimodal sensory data streams. The second part of this seminar is devoted to the presentation of case studies on different affective data sets (University Ulm Multimodal Affective Corpus (uulmMAC), SenseEmotion) that have been recently carried out in our lab.
|09:00 - 11:00||A. Schulz: Reversal Learning and strategy change in biological systems
Companion systems interact with users via flexible and goal-directed dialogues. During dialogues both user and companion system identify and communicate their goals iteratively. In that sense, user and companion system can be conceptualized as communication partners, equipped with a processing scheme producing actions as outputs in consequence of (1) inputs from the other communication partner and (2) internally represented goals. A quite general core competence of communication partners is the capability for strategy change, defined as the modification of action planning under the boundary condition of maintaining a constant goal. Interestingly, the biological mechanism for this capability are largely unknown. We describe a behavioural task with negative reinforcement that employs an animal model for learning and strategy change. We survey the theoretical concepts and the biological inspired neural models for learning and strategy change and consider the aspects of animal learning been useful for Companion systems.
|11:00 - 11:30||Coffee Break|
|11:30 - 13:00||A. Schulz: Reversal Learning and strategy change in biological systems
|13:00 - 14:00||Lunch|
|14:00 - 15:30||D. Wolf: Multi-Modal Interaction with Virtual Environments
Two important features of companion systems are adaptability and availability. When such a companion system is realized in a virtual or augmented reality environment, researchers and software developers are faced with two design questions: 1. How can users express their intention and manipulate the virtual environment in a natural and efficient way? 2. How can the system interact with users and convey information, apart from using the visual and auditive channels alone? These questions are driving my research to design novel interaction concepts for both, information input and output. Therefore, I am designing and evaluating software and hardware systems that bridge the gap between the virtual and the real world. I will demostrate how a multi-modal approach can help to convey information in a more subtle way and create a more natural interaction with a VR/AR environment.
|15:30 - 16:00||Coffee Break|
|16:00 - 17:00||D. Wolf: Multi-Modal Interaction with Virtual Environments
|18:00||Summer School Dinner "Drei Kannen - Historisches Brauhaus"|
|Monday,||09:00 - 11:00||Keynote: L.-P. Morency at ICCT|
|Sept 11||11:00 - 11:30||Coffee Break|
|11:30 - 13:00||at ICCT|
|13:00 - 14:00||Lunch (including Posters)|
|14:00 - 15:30||A. Platz: AI is the new UI
AI can help to raise interaction between humans and machines to another level, one of genuine cooperation. Designing user experiences with the companion metaphor as a leitmotif requires the elicitation of the experience qualities that resonate with the metaphor. And it requires the conceptualization of the situation as the cardinal point i.e. the triad of human, activity and context.
From recognition to understanding – intelligent interaction: Interfaces that are more than just operating devices rather multidimensional conversations and not just do what is said but understand what is meant.
From interaction to co-action - co-operation: More than just a reminder- or recommender system it is about creating a partnership between humans and computers based on an understanding which is at the heart of user experience design, taking the activity into the center rather than the technical artifact.
From work efficiency to cognitive quality - cognitive assistant / augmented intelligence: More than just automation of routine work it is about designing systems that enhance human expertise and augment human skills.
|15:30 - 16:00||Coffee Break|
|16:00 - 17:30||A. Platz: AI is the new UI|
|From 19:00||ICCT Conference Dinner|
|09:00 - 10:30||A. Stoica: Human-Robot Interaction
Companion technology extends its application to space exploration. In robotic exploration missions scientists and mission planners interact with the remote companion robot via increasingly richer interfaces, with a trend of creating for the scientist the perception it walks along the rover on a planetary surface and can seamlessly examine rocks fetched by the robot based on the gesture which he or she, as a virtual astronaut, indicates. In human flight mission, the on-board computer, increasingly intelligent, could act as a the on-board companion which interact with the astronaut which may descent on the planet surface. During human space flights, companion robots are imagined to assist astronauts in a range of tasks, from Extra-vehicular activities (EVA) to being their robotic surgeons. In surface exploration companion robots provide their superior perception and mobility to increase the value of the human-robot team and collaborate with astronauts in regular activities. The presentation explores these trends and emphasizes the importance of good understanding of the human intent, in which the ability to predict human actions and their outcome is critical.
|10:30 - 11:00||Coffee Break|
|11:00 - 12:30||A. Stoica: Human-Robot Interaction|
|12:30 - 13:30||Lunch (including Posters)|
|13:30 - 15:00||at ICCT|
|15:00 - 15:30||Coffee Break|
|15:30 - 18:15||ICCT and Excursion|
|From 19:00||City Tour|
|Wednesday,||09:00 - 10:30||Keynote: J. Hirschberg at ICCT|
|Sept 13||10:30 - 11:00||Coffee Break|
|11:00 - 12:45||P. Galambos: Human-centered Robotics and Cyber Physical Control
Today, the combination of various enabling technologies allows us to build and operate very complex mechatronic devices, intelligent production systems (system of systems), and autonomous machines never seen before. These advances necessarily involve the latest results of systems and control theory, nonlinear optimization, knowledge representations, and neurocomputing as well as the modern computational and telecommunication infrastructure (cloud technologies, large bandwidth, low-latency networks, etc.). Such systems over a certain level of complexity are often referred to as Cyber Physical Systems (CPS). Cyber Physical Control (CPC) as a multidisciplinary research field investigates the applications of existing theories and methodologies in the context of regulating CPS at different levels from low-level sensing and actuation to system-level orchestration. The lecture discusses various aspects of modern robotics in a CPC context with special focus on effective augmentation of human capabilities.
|12:45 - 13:45||Lunch (including Posters)|
|13:45 - 15:00||P. Galambos: Human-centered Robotics and Cyber Physical Control|
|15:00||End of Summer School
(Participants have the opportunity to join the remaining session of ICCT)
The program is subject to changes.