RoboCAP: Robotic Classification and Precision Pouring of Diverse Liquids and Granular Media with Capacitive Sensing

2024-05-13

arXiv.org

AI Summary

Accurate classification and controlled pouring of liquids and granular media within opaque containers remain challenging due to the absence of visual feedback and the complex, variable rheological properties of such materials. Method: This paper proposes an end-effector integrating an embedded parallel gripper with a capacitive sensing array, enabling non-visual, real-time identification of material composition and estimation of rheological states (e.g., viscosity, particle size). A closed-loop perception–decision–execution system is established by combining dielectric property modeling with dynamics-aware adaptive model predictive control (MPC). Robustness is ensured via hardware-in-the-loop calibration and real-time signal processing. Contribution/Results: Evaluated on 12 representative media, the system achieves ≤1.2 g mean error in 50 g pouring tasks—state-of-the-art accuracy—and 98.7% classification accuracy under full occlusion. It significantly enhances robotic autonomy in complex fluid–granular manipulation scenarios.

Technology Category

• Intelligent Robots: Manipulation
• Computer Vision: Vision for Robotics & Autonomous Driving
• Planning, Routing, and Scheduling: Activity and Plan Recognition

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Liquids and granular media are pervasive throughout human environments, yet remain particularly challenging for robots to sense and manipulate precisely. In this work, we present a systematic approach at integrating capacitive sensing within robotic end effectors to enable robust sensing and precise manipulation of liquids and granular media. We introduce the parallel-jaw RoboCAP Gripper with embedded capacitive sensing arrays that enable a robot to directly sense the materials and dynamics of liquids inside of diverse containers, including some visually opaque. When coupled with model-based control, we demonstrate that the proposed system enables a robotic manipulator to achieve state-of-the-art precision pouring accuracy for a range of substances with varying dynamics properties. Code, designs, and build details are available on the project website.

Problem

• Robotic sensing and manipulation of diverse liquids and granular media.
• Integration of capacitive sensing in robotic end effectors for precision.
• Achieving high-precision pouring accuracy for substances with varying properties.

Innovation

• Capacitive sensing in robotic end effectors
• Parallel-jaw RoboCAP Gripper with sensing arrays
• Model-based control for precision pouring accuracy