JST-CREST / IEEE-RAS Spring School on "Social and Artificial Intelligence for User-Friendly Robots"

17-24 March 2019, Shonan Village, Japan

Invited Speakers

  • Mohsen Kaboli, Bavarian Motor Works (BMW), Germany
  • Mohsen Kaboli Tutorial: RoboTac: The Sens of Touch in Robotics, from Perception to Learning
    Abstract : Humans rely on sense of touch for perception and control of the body, grasping, manipulating, and identifying objects via their physical properties such as texture, shape, and stiffness. For robotic systems interact with humans, dynamic environments, recognizing object properties is a crucial but difficult task for advanced vision techniques due to occlusion, poor lighting situations, and a lack of precision. Tactile sensing instead can provide a rich and direct feedback with the robotic systems from multiple contact points and a large tactile sensing area. In this talk I will cover the new methods I developed to tackle key challenges for active tactile object perception and learning in robotics. The new methods propose novel active pre-touch and touch-based exploration strategies for unknown workspaces. It introduces robust tactile feature descriptors to perceive the textural properties of the objects. I will present the first tactile-based approach to explore and determine the center of mass of rigid objects. Moreover, I propose a novel probabilistic active touch learning method to efficiently learn about objects as well as a new active tactile object discrimination to strategically discriminate among objects via their physical properties. For the first time in the tactile learning domain, this newly developed method proposes tactile transfer learning techniques which enable the robotic systems to re-use their past tactile experience (prior tactile knowledge) to learn about new objects with a low number of training samples. Furthermore, it introduces a novel tactile-based framework to enable the robotic systems to safely manipulate deformable objects with a dynamic center of mass. I will also describe a novel approach for touch modality identification during the tactile human-robot communication

    Bio: Dr. Mohsen Kaboli is a research scientist in robotics, tactile intelligence, and machine learning at the BMW research, innovation, and technology department. Prior to that, he was a postdoctoral researcher at the institute for advanced study (IAS) at the Technical University of Munich. He was awarded a Ph.D. degree in tactile perception and learning in robotics with the highest distinction (summa cum laude) from the Technical University of Munich (TUM) in 2017. He was a finalist for the 2018 Georges Giralt Ph.D. Award for the best robotics Ph.D. thesis in Europe. He received his Master’s degree in signal processing and machine learning under the supervision of Prof. Danica Kragic from the Royal Institute of Technology (KTH), Sweden in 2011. In April 2013 he was awarded a three-year Marie Curie scholarship in order to pursue his Ph.D. at the Institute for Cognitive Systems (ICS). In March 2012, he received an internship scholarship from the Swiss National Foundation for 18 months in order to continue his research as a research assistant at the Idiap lab, Ecole Polytechnique Federale de Lausanne (EPFL), Switzerland. From September 2015 till January 2016, Mohsen spent 5 months as a visiting research scholar at the Intelligent Systems and Informatics lab (ISI) directed by Prof. Yasuo Kuniyoshi at the University of Tokyo, Japan. He has also been a visiting researcher at the Human Robotics lab, the department of Bioengineering at the Imperial College London supervised by Prof. Etienne Burdet from February till April 2014. In November 2013, he visited the Shadow Robot Company for two months.