ICMA 2022 Conference
Plenary Talk 3
Doing Manipulation in Clutter
Henrik I Christensen, Ph.D., Dr. Techn. h.c.
Contextual Robotics Institute
University of California, San Diego
Robots are slowly being deployed outside of hard automation factory floors, for use-cases in logistics and in everyday settings. A key challenge for such applications is the need to interact with a physical world that is semi- or unstructured. In this presentation we will discuss the design of systems for manipulating objects in significant clutter. Clean-up of tables, assisting in meal preparation, and sorting return objects are all examples of scenarios where a service robot has to interact with objects in the presence of clutter. We will describe our strategy for recognizing / describing objects, how to plan a (multi-) grasp strategy for interacting with objects and finally how to ensure robust execution in the presence of uncertainty. Several new methods will be presented, and we will also show the performance of these methods for multi real-world applications.
Prof. Henrik I Christensen
is the Qualcomm Chancellor’s Chair of Robot Systems and the Director of the Contextual Robotics Institute at UC San Diego. He is also a distinguished Professor of Computer Science. Dr. Christensen uses a systems approach to study robotics and intelligent systems. For a given problem there is a need for a solid mathematical foundation, an effective algorithm, a real-world implementation and evaluation in a realistic setting.
Dr. Christensen has published close to 400 contributions across robotics, computer vision and artificial intelligence. He has served on numerous editorial boards and is the principal editor of the US National Robotics Roadmap (2009, 2013, 2016 and 2020). He was the founder of the EU Network of Excellence in Robotics and has founded multiple successful research centers prior to joining UC San Diego. He serves as an advisor to companies and government agencies across four continents. He has also co-founded four companies including Robust.AI and ROBO-Global.