Andrey Rudenko - Prediction, planning and interaction methods for a mobile robot in dynamic scenes

Abstract: Successful adoption of mobile service robots will strongly depend on their ability to safely and efficiently operate in human environments, engage in natural communication, understand their users, and express intentions intuitively while avoiding unnecessary distractions. To achieve this advanced level of human-robot Interaction, robots need to acquire and incorporate knowledge of their users’ tasks and environment, and adopt multimodal communication approaches with expressive cues that combine speech, movement, gazes, and other modalities. We approach this vision in three steps, building methods for the Robot, the Human and the Environment. In this talk, we will discuss how to learn the generalized patterns of motion in dynamic environments, predict long-term trajectories, goals and high-level cues such as engagement. We will also discuss how to build gamified user studies to collect human trajectories and human-robot spatial interaction data in lab conditions using motion capture and gaze tracking, and validate robot operation in terms of legibility and trust. Bio: Andrey Rudenko is a senior scientist at the Munich Institute for Robotics and Machine Intelligence, Technical University of Munich (TUM). He earned his Master degree at the University of Freiburg (Germany), and his PhD at the University of Örebro (Sweden), jointly with Robert Bosch Corporate Research, working on human motion prediction in complex dynamic environments. After completing his PhD, he worked at Robert Bosch Corporate Research as a research scientist in advanced autonomous systems, additionally contributing as a work package lead to the EU Horizon 2020 DARKO project (2021-2025). His research tackles various aspects of anticipating human motion and environment interaction, human-aware path planning, human-robot interaction, data collection and robot validation in large-scale user studies.