An Image-Based Approach to Variational Path Synthesis of Linkages

Author(s):  
Shrinath Deshpande ◽  
Anurag Purwar

Abstract This paper brings together computer vision, mechanism synthesis, and machine learning to create an image-based variational path synthesis approach for linkage mechanisms. An image-based approach is particularly amenable to mechanism synthesis when the input from mechanism designers is deliberately imprecise or inherently uncertain due to the nature of the problem. In addition, it also lends itself naturally to the creation of a unified approach to mechanism synthesis since pixels do not care if they were generated from a four-bar or six-bar. Path synthesis problem has generally been solved for a set of precision points on the intended path such that the designed mechanism passes through those points. This approach usually leads to a small set of over-fitting solutions to the particular precision points. However, most kinematic synthesis problems are concept generation problem where a designer cares more about generating a large number of plausible solutions. This paper models the input curve as a probability distribution of image pixels and employs a probabilistic generative model to capture the inherent uncertainty in the input. In addition, it gives feedback on the input quality and provides corrections for a more conducive input. The image representation allows for capturing local spatial correlations, which plays an important role in finding a variety of solutions with similar semantics as the input curve.

Author(s):  
Shrinath Deshpande ◽  
Anurag Purwar

Abstract This paper brings together computer vision, mechanism synthesis, and machine learning to create an image-based variational path synthesis approach for linkage mechanisms. An image-based approach is particularly amenable to mechanism synthesis when the input from mechanism designers is deliberately imprecise or inherently uncertain due to the nature of the problem. In addition, it also lends itself naturally to the creation of a unified approach to mechanism synthesis for different types of mechanisms, since for example, images are formed from a collection of pixels, which themselves could be generated from a four-bar or six-bar. Path synthesis problems have generally been solved for a set of precision points on the intended path such that the designed mechanism passes through those points. This approach usually leads to a small set of over-fitted solutions to particular precision points. However, most kinematic synthesis problems are concept generation problems, where a designer cares more about generating a large number of plausible solutions, which could reach given precision points only approximately. This paper models the input curve as a probability distribution of image pixels and employs a probabilistic generative model to capture the inherent uncertainty in the input. In addition, it gives feedback on the input quality and provides corrections for a more conducive input. The image representation allows for capturing local spatial correlations, which plays an important role in finding a variety of solutions with similar semantics as the input curve. This approach is also conducive to implementation for pressure-sensitive touch-based design interfaces, where the input is not a zero-thickness curve, but the sweep of a small patch on the finger.


2013 ◽  
Vol 135 (9) ◽  
Author(s):  
Punit Bandi ◽  
James P. Schmiedeler ◽  
Andrés Tovar

This work presents a novel method for designing crashworthy structures with controlled energy absorption based on the use of compliant mechanisms. This method helps in introducing flexibility at desired locations within the structure, which in turn reduces the peak force at the expense of a reasonable increase in intrusion. For this purpose, the given design domain is divided into two subdomains: flexible (FSD) and stiff (SSD) subdomains. The design in the flexible subdomain is governed by the compliant mechanism synthesis approach for which output ports are defined at the interface between the two subdomains. These output ports aid in defining potential load paths and help the user make better use of a given design space. The design in the stiff subdomain is governed by the principle of a fully stressed design for which material is distributed to achieve uniform energy distribution within the design space. Together, FSD and SSD provide for a combination of flexibility and stiffness in the structure, which is desirable for most crash applications.


2021 ◽  
Author(s):  
Shrinath Deshpande ◽  
Zhijie Lyu ◽  
Anurag Purwar

Abstract This paper brings together rigid body kinematics and machine learning to create a novel approach to path synthesis of linkage mechanisms under practical constraints, such as location of pivots. We model the coupler curve and constraints as probability distributions of image pixels and employ a Convolutional Neural Network (CNN) based Variational AutoEncoder (VAE) architecture to capture and predict the features of the mechanism. Plausible solutions are found by performing informed latent space exploration so as to minimize the changes to the input coupler curve while seeking to find user-defined pivot locations. Traditionally, kinematic synthesis problems are solved using precision point approach, wherein the input path is represented as a set of points and a set of equations in terms of design parameters are formulated. Generally, this problem is solved via optimization, wherein a measure of error between the given path and the coupler curve is minimized. A limitation of this approach is that the existing formulations depend on the type of mechanism, do not admit practical constraints in a unified way, and provide a limited number of solutions. However, in the machine design pipeline, kinematic synthesis problems are concept generation problems, where designers care more about a large number of plausible and practical solutions rather than the precision of input or the solutions. The image-based approach proposed in this paper alleviates the difficulty associated with inherently uncertain inputs and constraints.


Author(s):  
Punit Bandi ◽  
James P. Schmiedeler ◽  
Andrés Tovar

This work presents a novel method for designing crashworthy structures with controlled energy absorption based on the use of compliant mechanisms. This method helps in introducing flexibility at desired locations within the structure, which in turn reduces the peak force at the expense of a reasonable increase in intrusion. For this purpose, the given design domain is divided into two subdomains: flexible (FSD) and stiff (SSD) subdomains. The design in the flexible subdomain is governed by the compliant mechanism synthesis approach for which output ports are defined at the interface between the two subdomains. These output ports aid in defining potential load paths and help the user make better use of a given design space. The design in the stiff subdomain is governed by the principle of a fully-stressed design for which material is distributed to achieve uniform energy distribution within the design space. Together, FSD and SSD provide for a combination of flexibility and stiffness in the structure, which is desirable for most crash applications.


2014 ◽  
Vol 19 (3) ◽  
pp. 513-522 ◽  
Author(s):  
J. Buśkiewicz

Abstract A technique for path synthesis is employed to design a feeder for carrying products between two points. The feeder is assumed to be a one degree of freedom system of six links connected by means of revolute joints. The mathematical basis of the concept is presented. An exemplary solution is presented and discussed


2019 ◽  
Vol 142 (1) ◽  
Author(s):  
Wenrui Liu ◽  
Jianwei Sun ◽  
Jinkui Chu

Abstract An open path synthesis method for a spatial revolute-revolute-spherical-spherical (RRSS) mechanism is presented in this paper. The mathematical model for the trajectory curve is established. The characteristics of an RRSS mechanism in a standard installation position are revealed: the projection points of the coupler curve on the Oxy plane rotate by the corresponding input angles around the z-axis, and the generated points lie on an ellipse. Based on this finding, a 17-dimensional path generation problem can be translated into two lower-dimensional matching recognition problems and one actual size and installation position calculation problem. The path generation can be achieved by three steps. First, a database of four dimensional rotation angle parameters is established. By comparing the similarities between the mechanism feature curve of the prescribed open curve and its corresponding mechanism feature ellipse (MFE), the angles of installation, the initial angle of the input link, and the elliptic feature parameters of the desired RRSS mechanism can be approximately determined. Then, a 13-dimensional dynamic self-adapting numerical atlas database is established, which contains six basic dimensional types (BDTs) and seven wavelet feature parameters, and the BDTs of the desired RRSS mechanism are obtained. Finally, based on the relationship between the MFE of the prescribed curve and the BDTs of the desired RRSS mechanism, the calculation models for the actual link lengths and installation positions of the desired RRSS mechanism were established. Three examples are presented in this paper.


2020 ◽  
Vol 143 (1) ◽  
Author(s):  
Minho Lee ◽  
Dongwook Hwang ◽  
Yoonjin Lee ◽  
Byunghyun Choi ◽  
Woojin Park

Abstract This study presents a novel ideation aid termed technologically related products from other domains (TeRPODs) for addressing technology-push product concept generation problems. A technology-push product concept generation problem is specified in terms of a new technology and a target product/industry/business domain for which the technology is to be utilized to create novel products/applications. TeRPODs are example applications of the technology collected from domains other than the target domain and serve as brainstorming stimuli during the new product concept generation. An experiment was conducted to empirically evaluate the utility of TeRPODs. The technology-push product concept generation problem considered was that of developing new applications of the augmented reality head-up display (AR HUD) technology for the automotive domain. Two groups of participants, one utilizing a relevant set of TeRPODs (the TeRPODs group) for ideation and the other (the control group) without any ideation aids, performed individual brainstorming to generate new automotive AR HUD application concepts. The TeRPODs group produced a significantly better ideation outcome than the control group in the quantity and diversity of ideation outcome. Also, the use of TeRPODs significantly improved usefulness, novelty, preference, and profitability ratings of generated ideas. TeRPODs may serve as a useful ideation tool for technology-push inventions.


2005 ◽  
Vol 128 (5) ◽  
pp. 1080-1091 ◽  
Author(s):  
Kerr-Jia Lu ◽  
Sridhar Kota

A unified approach to topology and dimensional synthesis of compliant mechanisms is presented in this paper as a discrete optimization problem employing both discrete (topology) and continuous (size) variables. The synthesis scheme features a design parameterization method that treats load paths as discrete design variables to represent various topologies, thereby ensuring structural connectivity among the input, output, and ground supports. The load path synthesis approach overcomes certain design issues, such as “gray areas” and disconnected structures, inherent in previous design schemes. Additionally, multiple gradations of structural resolution and a variety of configurations can be generated without increasing the number of design variables. By treating topology synthesis as a discrete optimization problem, the synthesis approach is incorporated in a genetic algorithm to search for feasible topologies for single-input single-output compliant mechanisms. Two design examples, commonly seen in the compliant mechanisms literature, are included to illustrate the synthesis procedure and to benchmark the performance. The results show that the load path synthesis approach can effectively generate well-connected compliant mechanism designs that are free of gray areas.


Author(s):  
Shashank Sharma ◽  
Anurag Purwar ◽  
Q. Jeffrey Ge

Fourier descriptor (FD)-based path synthesis algorithms for generation of planar four-bar mechanisms require assigning time parameter values to the given points along the path. An improper selection of time parameters leads to poor fitting of the given path and suboptimal four-bar mechanisms while also ignoring a host of mechanisms that could be potentially generated otherwise. A common approach taken is to use uniform time parameter values, which does not take into account the unique harmonic properties of the coupler path. In this paper, we are presenting a nonuniform parametrization scheme in conjunction with an objective function that provides a better fit, leverages the harmonics of the four-bar coupler, and allows imposing additional user-specified constraints.


Sign in / Sign up

Export Citation Format

Share Document