Robot learning for navigation and manipulation

Abstract:

Traditional approaches to robot navigation and manipulation rely heavily on 3D computer vision and geometric modeling. However, recent years have seen a surge of interest in learning-based methods that promise greater adaptability and robustness in complex environments. In this talk, I will outline the core challenges in navigation and manipulation, present our latest research findings, and highlight compelling ideas from other groups that are shaping the future of robot learning.

Bio:

Dr. Joni Kämäräinen is a Professor of Signal Processing at the Department of Computing Sciences, Tampere University, Finland, where he leads the Vision Group (http://research.tuni.fi/vision). His academic background is rooted in computer vision and machine learning, with a strong emphasis on visual perception. In recent years, his research has strategically shifted toward robot vision and robot learning, aligning with the evolving demands of intelligent autonomous systems.

Zoom: https://chalmers.zoom.us/j/62416120718?from=addon (Mötes-ID: 624 1612 0718, Lösenkod: 835453)