An official website of the United States government
Learning Audio Feedback for Estimating Amount and Flow of Granular Material
Granular materials produce audio-frequency mechanical vibrations in air and structures when manipulated. These vibrations correlate with both the nature of the events and the intrinsic properties of the materials producing them. We therefore propose learning to use audio-frequency vibrations from contact events to estimate the flow and amount of granular materials during scooping and pouring tasks. We evaluated multiple deep and shallow learning frameworks on a dataset of 13,750 shaking and pouring samples across five different granular materials. Our results indicate that audio is an informative sensor modality for accurately estimating flow and amounts, with a mean RMSE of 2.8g across the five materials for pouring. We also demonstrate how the learned networks can be used to pour a desired amount of material.
more »
« less
- Award ID(s):
- 1717066
- PAR ID:
- 10156262
- Date Published:
- Journal Name:
- Proceedings of Machine Learning Research
- Volume:
- 87
- ISSN:
- 2640-3498
- Page Range / eLocation ID:
- 529-550
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
Rare event prediction involves identifying and forecasting events with a low probability using machine learning (ML) and data analysis. Due to the imbalanced data distributions, where the frequency of common events vastly outweighs that of rare events, it requires using specialized methods within each step of the ML pipeline, that is, from data processing to algorithms to evaluation protocols. Predicting the occurrences of rare events is important for real-world applications, such as Industry 4.0, and is an active research area in statistical and ML. This article comprehensively reviews the current approaches for rare event prediction along four dimensions: rare event data, data processing, algorithmic approaches, and evaluation approaches. Specifically, we consider 73 datasets from different modalities (i.e., numerical, image, text, and audio), four major categories of data processing, five major algorithmic groupings, and two broader evaluation approaches. This article aims to identify gaps in the current literature and highlight the challenges of predicting rare events. It also suggests potential research directions, which can help guide practitioners and researchers.more » « less
-
This study investigates the Berry phase, a key concept in classical and quantum physics, and its manifestation in a classical system. We achieve controlled accumulation of the Berry phase by manipulating the elastic bit (a classical analogue to a quantum bit) in an externally driven, homogeneous, spherical, nonlinear granular network. This is achieved through the classical counterpart of quantum coherent superposition of states. The elastic bit's state vectors are navigated on the Bloch sphere using external drivers' amplitude, phase, and frequency, yielding specific Berry phases. These phases distinguish between trivial and nontrivial topologies of the elastic bit, with the zero Berry phase indicating pure states of the linearized granular system and the nontrivial π phase representing equal superposed states. Other superposed states acquire different Berry phases. Crucially, these phases correlate with the structure's eigenmode vibrations: trivial phases align with distinct, in-phase, or out-of-phase eigenmodes, while nontrivial phases correspond to coupled vibrations where energy is shared among granules, alternating between oscillation and rest. Additionally, we explore Berry's phase generalizations for non-cyclic evolutions. This research paves the way for advanced quantum-inspired sensing and computation applications by utilizing and controlling the Berry phase.more » « less
-
The flow threshold in dense granular materials is typically modeled by local, stress-based criteria. However, grain-scale cooperativity leads to size effects that cannot be captured with local conditions. In a widely studied example, flows of thin layers of grains down an inclined surface exhibit a size effect whereby thinner layers require more tilt to flow. In this paper, we consider the question of whether the size-dependence of the flow threshold observed in inclined plane flow is configurationally general. Specifically, we consider three different examples of inhomogeneous flow – planar shear flow with gravity, annular shear flow, and vertical chute flow – using two-dimensional discrete-element method calculations and show that the flow threshold is indeed size-dependent in these flow configurations, displaying additional strengthening as the system size is reduced. We then show that the nonlocal granular fluidity model – a nonlocal continuum model for dense granular flow – is capable of quantitatively capturing the observed size-dependent strengthening in all three flow configurations.more » « less
-
Lift and drag forces on moving intruders in flowing granular materials are of fundamental interest but have not yet been fully characterized. Drag on an intruder in granular shear flow has been studied almost exclusively for the intruder moving across flow streamlines, and the few studies of the lift explore a relatively limited range of parameters. Here, we use discrete element method simulations to measure the lift force,$$F_{{L}}$$, and the drag force on a spherical intruder in a uniformly sheared bed of smaller spheres for a range of streamwise intruder slip velocities,$$u_{{s}}$$. The streamwise drag matches the previously characterized Stokes-like cross-flow drag. However,$$F_{{L}}$$in granular shear flow acts in the opposite direction to the Saffman lift in a sheared fluid at low$$u_{{s}}$$, reaches a maximum value and then decreases with increasing$$u_{{s}}$$, eventually reversing direction. This non-monotonic response holds over a range of flow conditions, and the$$F_{{L}}$$versus$$u_{{s}}$$data collapse when both quantities are scaled using the particle size, shear rate and overburden pressure. Analogous fluid simulations demonstrate that the flow around the intruder particle is similar in the granular and fluid cases. However, the shear stress on the granular intruder is notably less than that in a fluid shear flow. This difference, combined with a void behind the intruder in granular flow in which the stresses are zero, significantly changes the lift-force-inducing stresses acting on the intruder between the granular and fluid cases.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
