An official website of the United States government
Abstract This paper focuses on the representation and synthesis of coupler curves of planar mechanisms using a deep neural network. While the path synthesis of planar mechanisms is not a new problem, the effective representation of coupler curves in the context of neural networks has not been fully explored. This study compares four commonly used features or representations of four-bar coupler curves: Fourier descriptors, wavelets, point coordinates, and images. The results demonstrate that these diverse representations can be unified using a generative AI framework called variational autoencoder (VAE). This study shows that a VAE can provide a standalone representation of a coupler curve, regardless of the input representation, and that the compact latent dimensions of the VAE can be used to describe coupler curves of four-bar linkages. Additionally, a new approach that utilizes a VAE in conjunction with a fully connected neural network to generate dimensional parameters of four-bar linkage mechanisms is proposed. This research presents a novel opportunity for the automated conceptual design of mechanisms for robots and machines.
more »
« less
Transforming Hand-Drawn Sketches of Linkage Mechanisms Into Their Digital Representation
Abstract This paper introduces a new method using deep neural networks for the interactive digital transformation and simulation of n-bar planar linkages, which consist of revolute and prismatic joints, based on hand-drawn sketches. Instead of relying solely on computer vision, our approach combines topological knowledge of linkage mechanisms with the outcomes of a convolutional deep neural network. This creates a framework for recognizing hand-drawn sketches. We generate a dataset of synthetic images that resemble hand-drawn sketches of linkage mechanisms. Next, we fine-tune a state-of-the-art deep neural network to detect discrete objects using building blocks that represent joints and links in various positions, sizes, and orientations within these sketches. We then conduct a topological analysis on the detected objects to construct a kinematic model of the sketched mechanisms. The results demonstrate the effectiveness of our algorithm in handling hand-drawn sketches and converting them into digital representations. This has practical implications for improving communication, analysis, organization, and classification of planar mechanisms.
more »
« less
- Award ID(s):
- 2126882
- PAR ID:
- 10488731
- Publisher / Repository:
- ASME Journal of Computing and Information Science in Engineering
- Date Published:
- Journal Name:
- Journal of Computing and Information Science in Engineering
- Volume:
- 24
- Issue:
- 1
- ISSN:
- 1530-9827
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
This paper presents a method to develop continuum/ compliant mechanisms based on planar bar-node linkage precursors. The method takes as inputs the initial node positions and connectivity data of a given bar-node linkage and converts it into a continuum/compliant mechanism having the same targeted motion. The line bars of the given bar-node linkage are thickened into trapezoidal planar members and the nodes are thickened by introducing fillets at each intersection of bars. The thicknesses of the bars and the shape parameters of the fillets in the continuum/compliant linkage are optimized to obtain the same targeted motion of the given bar-node linkage while keeping stresses below a maximum allowable value. Each design generated during the optimization process is evaluated using finite element analysis. The present method allows for the synthesis of mechanisms having the following advantages over conventional bar-node linkages: 1) They do not require complex ball or pin joints; 2) they can be readily 3-D printed and sizescaled, and 3) they can be optimized to decrease stresses below a maximum allowable value. Furthermore, the method uses a relatively small number of optimization variables (thicknesses of the members, shape-parameters of the fillets), making it an efficient alternative to more complex and computationally intensive methods for synthesizing compliant mechanisms such as those incorporating topology optimization.more » « less
-
Abstract This paper presents a novel real-time kinematic simulation algorithm for planar N-bar linkage mechanisms, both single- and multi-degrees-of-freedom, comprising revolute and/or prismatic joints and actuators. A key feature of this algorithm is a reinterpretation technique that transforms prismatic elements into a combination of revolute joint and links. This gives rise to a unified system of geometric constraints and a general-purpose solver which adapts to the complexity of the mechanism. The solver requires only two types of methods—fast dyadic decomposition and relatively slower optimization-based—to simulate all types of planar mechanisms. From an implementation point of view, this algorithm simplifies programming without requiring handling of different types of mechanisms. This versatile algorithm can handle serial, parallel, and hybrid planar mechanisms with varying degrees-of-freedom and joint types. Additionally, this paper presents an estimation of simulation time and structural complexity, shedding light on computational demands. Demonstrative examples showcase the practicality of this method.more » « less
-
null (Ed.)Abstract Cognate linkages are mechanisms that share the same motion, a property that can be useful in mechanical design. This article treats planar curve cognates, that is, planar mechanisms with rotational joints whose coupler points draw the same curve, as well as coupler cognates and timed curve cognates. The purpose of this article is to develop a straightforward method based solely on kinematic equations to construct cognates. The approach computes cognates that arise from permuting link rotations and is shown to reproduce all of the known results for cognates of four-bar and six-bar linkages. This approach is then used to construct a cognate of an eight-bar and a ten-bar linkage.more » « less
-
Abstract Deep neural networks such as GoogLeNet, ResNet, and BERT have achieved impressive performance in tasks such as image and text classification. To understand how such performance is achieved, we probe a trained deep neural network by studying neuron activations, i.e.combinations of neuron firings, at various layers of the network in response to a particular input. With a large number of inputs, we aim to obtain a global view of what neurons detect by studying their activations. In particular, we develop visualizations that show the shape of the activation space, the organizational principle behind neuron activations, and the relationships of these activations within a layer. Applying tools from topological data analysis, we presentTopoAct, a visual exploration system to study topological summaries of activation vectors. We present exploration scenarios usingTopoActthat provide valuable insights into learned representations of neural networks. We expectTopoActto give a topological perspective that enriches the current toolbox of neural network analysis, and to provide a basis for network architecture diagnosis and data anomaly detection.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1115/1.4064037
