Human-Machine Interaction Modeling for Continuous Improvement of Robot Behavior

Human-Machine Interaction Modeling for Continuous Improvement of Robot Behavior

The current developments in digitalization, AI-based data processing, and powerful hardware lay the groundwork for future embodied AI assistants. These intelligent robots must be versatile, adaptive and flexible with respect to changes in the environment or requirements to become optimized for their purpose for assistance in every day life and work. Furthermore, they must be tailored to the needs of the user in order to achieve a high level of acceptance and to fulfill the desire for individuality. The user wants to decide how a system should look and behave. This is not only true for everyday personal assistance but also for collaboration between robots and humans in production, logistics or care. In addition, individually tailored robots should be easily designed by the user and, if possible, automatically adapted during use to changing needs or observed new requirements.

The X-Rock series addresses precisely these aspects. X-Rock enables users to design their own personal assistant without the need for any expert knowledge but will also help domain experts in identifying possibilities for improvement of a system. Modularization and modeling developed in D-Rock enables efficient reuse of components and describes how components can be used in a given context. Q-Rock automatically maps the structural hardware and software complexity of current robotic systems to behaviors. M-Rock will directly build on the results of D-Rock and Q-Rock. Its main goal is to enable the usage of explicit and implicit user feedback to not only optimize behavior on the software side with respect to the individual requirements of the user, as shown in Q-Rock’s outlook, but also to enable subsequent optimization of the Q-Rock software flow including hardware selection. M-Rock will enable an automatic adaption of a robot to the individual requirements and preferences of the interacting human. To this end, we combine explicit feed- back (i.e., rating scale for performance evaluation) with implicit feedback. As a source for implicit human feedback M-Rock will make use of the users’ EEGs.

With the help of two evaluation scenarios we will show how the developments in M-Rock can be used to optimize software and hardware of a robot by enabling the usage of explicit and implicit human feedback within the Q-Rock cycle and how it can be used by laymen as well as by domain experts alike.


Federal Ministry of Education and Research (BMBF)


Federal Ministry of Education and Research (BMBF)

Publications about the project

Durgesh Haribhau Salunkhe, Guillaume Michel, Elise Olivier, Marcello Sanguineti, Damien Chablat, Shivesh Kumar

In: 2022 Workshop: "New frontiers of parallel robotics" (second edition). IEEE International Conference on Robotics and Automation (ICRA-2022) May 23-27 Philadelphia PA United States IEEE 5/2022.

To the publication
Ibrahim Tijjani

In: Andreas Müller, Mathias Brandstötter (editor). Advances in Service and Industrial Robotics. International Conference on Robotics in Alpe-Adria-Danube Region (RAAD-2022) Advances in Service and Industrial Robotics June 8-10 Klagenfurt am Wörthersee Austria Pages 536-544 Mechanisms and Machine Science 120 ISBN 978-3-031-04870-8 Springer International Publishing 4/2022.

To the publication

German Research Center for Artificial Intelligence
Deutsches Forschungszentrum für Künstliche Intelligenz