key: Conference

  • Diff-Control: A Stateful Diffusion-based Policy for Imitation Learning

    Abstract: While imitation learning provides a simple and effective framework for policy learning, acquiring consistent action during robot execution remains a challenging task. Existing approaches primarily focus on either modifying the action representation at data curation stage or altering the model itself, both of which do not fully address the scalability of consistent action generation.…

  • iRoCo: Intuitive Robot Control From Anywhere Using a Smartwatch

    Summary: This paper introduces a framework for ubiquitous human-robot collaboration using a single smartwatch and smartphone. By integrating probabilistic differentiable filters, we optimize a combination of precise robot control and unrestricted user movement from ubiquitous devices. The system is available as WearMocap. We demonstrate and evaluate the effectiveness of WearMoCap in practical teleoperation and drone…

  • Anytime, Anywhere: Human Arm Pose From Smartwatch Data for Ubiquitous Robot Control and Teleoperation

    Nominated for the RoboCup Best Paper Award! (Top 6% of IROS23) This work devises an optimized machine learning approach for human arm pose estimation from a single smartwatch. Our approach results in a distribution of possible wrist and elbow positions, which allows for a measure of uncertainty and the detection of multiple possible arm posture…

  • A New Pathway to Approximate Energy Expenditure and Recovery of an Athlete

    This work proposes to use evolutionary computation as a pathway to allow a new perspective on the modeling of energy expenditure and recovery of an individual athlete during exercise. We revisit a theoretical concept called the “three component hydraulic model” which is designed to simulate metabolic systems during exercise and which is able to address…