AASS Seminar - Image and Time Series Processing in Mobile Robotics and Beyond

17 oktober 2024 13:00 Labbet

For more information about the AASS Seminar Series, please contact:
Alessandro Saffiotti

The research centre AASS arranges a seminar with Kenny Schlegel, TU Chemnitz, DE.

Remote attendance: https://oru-se.zoom.us/j/65096623293

Abstract

This talk will present our research in mobile robotics, focusing on essential subtasks such as navigation and perception. Through examples of robotic systems, including a shopping assistant robot, we will highlight the challenges involved in robotic perception, particularly in vision and signal processing. For example, visual place recognition plays a critical role in SLAM (Simultaneous Localization and Mapping), where efficiently encoding input images presents significant technical challenges, especially in environments with changing conditions like weather or light. Furthermore, we will discuss the importance of time series data analysis for autonomous systems, such as robots and vehicles, where fast and accurate data processing is crucial for decision-making. For example, in the context of autonomous driving, time series data from vehicles or drivers can reveal important system states and scenarios. By applying advancements in computing with neural representations to these tasks, we can improve the encoding and interpretation of perceptual data, enabling more efficient and scalable methods that extend beyond the field of mobile robotics into a wider range of applications.

Short bio

Kenny Schlegel holds a Bachelor's and Master's degree in Electrical Engineering, with a specialization in automation systems. He is currently a Ph.D. candidate at Chemnitz University of Technology, Germany, where his research focuses on mobile robotics, particularly in image and time series processing. His interests are in advancing Artificial Intelligence methods via contributions to the field of hyperdimensional computing, which enables efficient and compact data representation through high-dimensional neural representations. His works find applications in areas such as classification from spatiotemporal data. He has also been involved in applied projects such as the development of an autonomous shopping assistant robot.