Scenario-Based Engineering of Cognitive Robots and Systems

The wider adoption of autonomous, cognitive robots and systems by society depends on the perceived trust of various stakeholders (e.g., engineers, users, regulators) in the dependable and ethical nature of the systems. However, developers face unprecedented engineering challenges caused by the intrinsic characteristics of cognitive robots, such as their enormous design space, challenging and uncertain operating conditions, and their demand to recover and learn quickly from failures. The construction and responsible deployment of cognitive robots remains a key challenge, calling for novel concepts, methods, and tools that consider the complete life cycle of cognitive robots in both the real and virtual worlds.

In this talk, I will present my group’s contributions to automating the engineering of robot software systems by combining foundational and applied research, often in collaboration with the industry. To this end, I present my past, present, and future research based on the concept of scenarios, enabling developers to capture and express knowledge and assumptions regarding robot tasks, platforms, and environments explicitly and formally in the form of domain models. Previously, these models were considered as a way to support humans during the software design process, and I will show how scenarios can be exploited for various testing activities (e.g., simulation-based testing, standard conformance test, and adversarial testing). I argue that scenarios should be used throughout the complete lifecycle of autonomous robots to ensure their safety, robustness, and transparency and present preliminary findings on how scenarios can be exploited by robots themselves for the sake of autonomously adapting their software to the various and changing run-time requirements induced by  the robot’s task or environment.

Short Bio:

Nico Hochgeschwender is Professor for Software Engineering of Cognitive Robots and Cognitive Systems at University of Bremen. Before, he was a Professor for Autonomous Systems at Hochschule Bonn-Rhein-Sieg (H-BRS) in Sankt Augustin, Germany, where he founded the Institute of AI and Autonomous Systems. He worked in the avionics industry as a system engineer, Postdoctoral Fellow at the University of Luxembourg, and Research Group Leader at DLR. His research interests lie at the intersection of AI-enabled Robotics and Software Engineering, with a focus on ensuring dependability, transparency, and explainability of robotics and autonomous systems; benchmarking and performance evaluation of cognitive robots; and domain-specific modelling and languages for robotics. He holds a PhD degree from the University of Luxembourg and is currently a PI of three EU-funded research projects: SESAME (Safe and Secure Multi-Robot Systems), METRICS (Metrological Evaluation and Testing of Robots in International Competitions), and SOPRANO (Socially-Acceptable and Trustworthy Human-Robot Teaming for Agile Industries). In 2013, he co-founded the RoboCup@Work league as a scientific competition targeting industrial and work-related scenarios and ten years later, his team won the RoboCup@Work competition. He is a member of the GI, VDI, and IEEE and Co-Chair of the IEEE RAS Technical Committee on Software Engineering for Robotics and Automation.