Carbon fiber reinforced polymers (CFRP) have been used in many high-performance applications where strength to weight ratio is an important characteristic. The goal of this research was to analyze the effects of Mode II, also known as shear loading, on the displacement fields surrounding a crack for unidirectional carbon fiber composites. Tensile and fatigue experiments were conducted on angled unidirectional CFRP coupled with digital image correlation (DIC) to analyze the full field displacement. Angled CFRP cracks experienced mixed mode loading which required addition insight due to the complex stresses on the fiber/matrix interface. The experimental displacement fields acquired from DIC were used as inputs for an anisotropic regression analysis to determine the mode I and mode II stress intensity factor ranges. The results from the regression analysis were used to predict the displacement fields around a crack. When comparing the experimental results with the predicted results, the inclusion of Mode II increased the agreement between predicted and experimental displacement fields around a crack tip for two different fiber orientation angles. Crack growth rate analysis and analytical stress intensity factor ranges were used to expand on the agreement of the results as well as bring to light CFRP specific fracture mechanisms that lead to disagreements.
{"title":"The Importance of the Mode II Term on the Analysis of Angled Cracks in Unidirectional Carbon Fiber Composites","authors":"Jacob Biddlecom, G. Pataky","doi":"10.1115/detc2021-67236","DOIUrl":"https://doi.org/10.1115/detc2021-67236","url":null,"abstract":"\u0000 Carbon fiber reinforced polymers (CFRP) have been used in many high-performance applications where strength to weight ratio is an important characteristic. The goal of this research was to analyze the effects of Mode II, also known as shear loading, on the displacement fields surrounding a crack for unidirectional carbon fiber composites. Tensile and fatigue experiments were conducted on angled unidirectional CFRP coupled with digital image correlation (DIC) to analyze the full field displacement. Angled CFRP cracks experienced mixed mode loading which required addition insight due to the complex stresses on the fiber/matrix interface. The experimental displacement fields acquired from DIC were used as inputs for an anisotropic regression analysis to determine the mode I and mode II stress intensity factor ranges. The results from the regression analysis were used to predict the displacement fields around a crack. When comparing the experimental results with the predicted results, the inclusion of Mode II increased the agreement between predicted and experimental displacement fields around a crack tip for two different fiber orientation angles. Crack growth rate analysis and analytical stress intensity factor ranges were used to expand on the agreement of the results as well as bring to light CFRP specific fracture mechanisms that lead to disagreements.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":"41 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81121007","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
R. Malhan, R. Joseph, P. Bhatt, Brual C. Shah, Satyandra K. Gupta
3D reconstruction technology is used in a wide variety of applications. Currently, automatically creating accurate pointclouds for large parts requires expensive hardware. We are interested in using low-cost depth cameras mounted on commonly available industrial robots to create accurate pointclouds for large parts automatically. Manufacturing applications require fast cycle times. Therefore, we are interested in speeding up the 3D reconstruction process. We present algorithmic advances in 3D reconstruction that achieve a sub-millimeter accuracy using a low-cost depth camera. Our system can be used to determine a pointcloud model of large and complex parts. Advances in camera calibration, cycle time reduction for pointcloud capturing, and uncertainty estimation are made in this work. We continuously capture point-clouds at an optimal camera location with respect to part distance during robot motion execution. The redundancy in pointclouds achieved by the moving camera significantly reduces errors in measurements without increasing cycle time. Our system produces sub-millimeter accuracy.
{"title":"Fast, Accurate, and Automated 3D Reconstruction Using a Depth Camera Mounted on an Industrial Robot","authors":"R. Malhan, R. Joseph, P. Bhatt, Brual C. Shah, Satyandra K. Gupta","doi":"10.1115/detc2021-71725","DOIUrl":"https://doi.org/10.1115/detc2021-71725","url":null,"abstract":"\u0000 3D reconstruction technology is used in a wide variety of applications. Currently, automatically creating accurate pointclouds for large parts requires expensive hardware. We are interested in using low-cost depth cameras mounted on commonly available industrial robots to create accurate pointclouds for large parts automatically. Manufacturing applications require fast cycle times. Therefore, we are interested in speeding up the 3D reconstruction process. We present algorithmic advances in 3D reconstruction that achieve a sub-millimeter accuracy using a low-cost depth camera. Our system can be used to determine a pointcloud model of large and complex parts. Advances in camera calibration, cycle time reduction for pointcloud capturing, and uncertainty estimation are made in this work. We continuously capture point-clouds at an optimal camera location with respect to part distance during robot motion execution. The redundancy in pointclouds achieved by the moving camera significantly reduces errors in measurements without increasing cycle time. Our system produces sub-millimeter accuracy.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":"45 8","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91503981","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Estimating the form and functional performance of a design in the early stages can be crucial for a designer for effective ideation Humans have an innate ability to guess the size, shape, and type of a design from a single view. The brain fills in the unknowns in a fraction of a second. However, humans may struggle with estimating the performance of designs in the early stages of the design process without making prototypes or doing back-of-the-envelope calculations. In contrast, machines need information about the full 3D model of a design to understand its structure. Machines can estimate the performance using pre-defined rules, expensive numerical simulations, or machine learning models. In this paper, we show how information about the form and functional performance of a design can be estimated from a single image using machine learning methods. Specifically, we leverage the image-to-image translation method to predict multiple projections of an image-based design. We then train deep neural network models on the predicted projections to provide estimates of design performance. We demonstrate the effectiveness of our method by predicting the aerodynamic performance from images of aircraft models. To estimate ground truth aero-dynamic performance, we run CFD simulations for 4045 3D aircraft models from the ShapeNet dataset and use their lift-to-drag ratio as the performance metric. Our results show that single images do carry information for both form and functional performance. From a single image, we are able to produce six additional images of a design in different orientations, with an average Structural Similarity Index score of 0.872. We also find image-translation methods provide a promising direction in estimating the performance of design. Using multiple images of a design (gathered through image-translation) to predict design performance yields a recall value of 47%, which is 14% higher than a base guess, and 3% higher than using a single image. Our work identifies the potential and provides a framework for using a single image to predict the form and functional performance of a design during the early-stage design process. Our code and additional information about our work are available at http://decode.mit.edu/projects/formfunction/.
{"title":"Design Form and Function Prediction From a Single Image","authors":"K. M. Edwards, Vaishnavi L. Addala, Faez Ahmed","doi":"10.1115/detc2021-71853","DOIUrl":"https://doi.org/10.1115/detc2021-71853","url":null,"abstract":"\u0000 Estimating the form and functional performance of a design in the early stages can be crucial for a designer for effective ideation Humans have an innate ability to guess the size, shape, and type of a design from a single view. The brain fills in the unknowns in a fraction of a second. However, humans may struggle with estimating the performance of designs in the early stages of the design process without making prototypes or doing back-of-the-envelope calculations. In contrast, machines need information about the full 3D model of a design to understand its structure. Machines can estimate the performance using pre-defined rules, expensive numerical simulations, or machine learning models. In this paper, we show how information about the form and functional performance of a design can be estimated from a single image using machine learning methods. Specifically, we leverage the image-to-image translation method to predict multiple projections of an image-based design. We then train deep neural network models on the predicted projections to provide estimates of design performance. We demonstrate the effectiveness of our method by predicting the aerodynamic performance from images of aircraft models. To estimate ground truth aero-dynamic performance, we run CFD simulations for 4045 3D aircraft models from the ShapeNet dataset and use their lift-to-drag ratio as the performance metric. Our results show that single images do carry information for both form and functional performance. From a single image, we are able to produce six additional images of a design in different orientations, with an average Structural Similarity Index score of 0.872. We also find image-translation methods provide a promising direction in estimating the performance of design. Using multiple images of a design (gathered through image-translation) to predict design performance yields a recall value of 47%, which is 14% higher than a base guess, and 3% higher than using a single image. Our work identifies the potential and provides a framework for using a single image to predict the form and functional performance of a design during the early-stage design process. Our code and additional information about our work are available at http://decode.mit.edu/projects/formfunction/.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85030713","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
With the dynamic arrival of production orders and ever-changing shop-floor conditions within a production system, production scheduling presents a challenge for manufacturing firms to ensure production demands that are met with high productivity and low operating cost. Before a production schedule is generated to process the incoming production orders, the production planning stage must take place. Given the large number of input parameters involved in production planning, it is important to understand the interactions of input parameters between production planning and scheduling. This is to ensure that production planning and scheduling could be determined effectively and efficiently in achieving the best or optimal production performance with minimizing cost. In this study, by utilizing the capabilities of data pervasiveness in smart manufacturing setting, we propose an approach to develop a surrogate model to predict the production performance using the input parameters from a production plan. Based on three categories of input parameters, namely current production system load, machine-based and product-based parameters, the prediction is performed by developing a surrogate model using multivariate adaptive regression spline (MARS). The effectiveness of the proposed MARS model is demonstrated using an industrial case study of a wafer fabrication production through the random sampling of varying numbers of training data set.
{"title":"Prediction of Production Performance in Smart Manufacturing Using Multivariate Adaptive Regression Spline","authors":"P. C. Chua, S. K. Moon, Y. Ng, H. Ng","doi":"10.1115/detc2021-69632","DOIUrl":"https://doi.org/10.1115/detc2021-69632","url":null,"abstract":"\u0000 With the dynamic arrival of production orders and ever-changing shop-floor conditions within a production system, production scheduling presents a challenge for manufacturing firms to ensure production demands that are met with high productivity and low operating cost. Before a production schedule is generated to process the incoming production orders, the production planning stage must take place. Given the large number of input parameters involved in production planning, it is important to understand the interactions of input parameters between production planning and scheduling. This is to ensure that production planning and scheduling could be determined effectively and efficiently in achieving the best or optimal production performance with minimizing cost. In this study, by utilizing the capabilities of data pervasiveness in smart manufacturing setting, we propose an approach to develop a surrogate model to predict the production performance using the input parameters from a production plan. Based on three categories of input parameters, namely current production system load, machine-based and product-based parameters, the prediction is performed by developing a surrogate model using multivariate adaptive regression spline (MARS). The effectiveness of the proposed MARS model is demonstrated using an industrial case study of a wafer fabrication production through the random sampling of varying numbers of training data set.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":"30 4 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82979030","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Augmented reality (AR) technologies present immense potential for the design and manufacturing communities. However, coordinating traditional engineering data representations into AR systems without loss of context and information remains a challenge. A major barrier is the lack of interoperability between manufacturing-specific data models and AR-capable data representations. In response, we present a pipeline for porting standards-based Product Manufacturing Information (PMI) with three-dimensional (3D) model data into an AR scene. We demonstrate our pipeline by interacting with annotated parts while continuously tracking their pose and orientation. Our work provides insight on how to address fundamental issues related to interoperability between domain-specific models and AR systems.
{"title":"Visualizing Model-Based Product Definitions in Augmented Reality","authors":"Teodor Vernica, R. Lipman, W. Bernstein","doi":"10.1115/detc2021-71329","DOIUrl":"https://doi.org/10.1115/detc2021-71329","url":null,"abstract":"\u0000 Augmented reality (AR) technologies present immense potential for the design and manufacturing communities. However, coordinating traditional engineering data representations into AR systems without loss of context and information remains a challenge. A major barrier is the lack of interoperability between manufacturing-specific data models and AR-capable data representations. In response, we present a pipeline for porting standards-based Product Manufacturing Information (PMI) with three-dimensional (3D) model data into an AR scene. We demonstrate our pipeline by interacting with annotated parts while continuously tracking their pose and orientation. Our work provides insight on how to address fundamental issues related to interoperability between domain-specific models and AR systems.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":"94 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84253754","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The purpose of this paper is to propose a new method for the automatic composition of both encoding schemes and search operators for system architecture optimization. The method leverages prior work that identified a set of six patterns that appear often in system architecture decision problems (down-selecting, combining, assigning, partitioning, permuting, and connecting). First, the user models the architecture space as a directed graph, where nodes are decisions belonging to one of the aforementioned patterns, and edges are dependencies between decisions that affect architecture enumeration (e.g., the outcome of decision 1 affects the number of alternatives available for decision 2). Then, based on a library of encoding scheme and operator fragments that are appropriate for each pattern, an algorithm automatically composes an encoding scheme and corresponding search operators by traversing the graph. The method is demonstrated in two case studies: a study integrating three architectural decisions for constructing a portfolio of earth observing satellite missions, and a study integrating eight architectural decisions for designing a guidance navigation and control system. Results suggest that this method has comparable search performance to hand-crafted formulations from experts. Furthermore, the proposed method drastically reducing the need for practitioners to write new code.
{"title":"Automatic Composition of Encoding Scheme and Search Operators in System Architecture Optimization","authors":"Gabriel Apaza, Daniel Selva","doi":"10.1115/detc2021-71399","DOIUrl":"https://doi.org/10.1115/detc2021-71399","url":null,"abstract":"\u0000 The purpose of this paper is to propose a new method for the automatic composition of both encoding schemes and search operators for system architecture optimization. The method leverages prior work that identified a set of six patterns that appear often in system architecture decision problems (down-selecting, combining, assigning, partitioning, permuting, and connecting). First, the user models the architecture space as a directed graph, where nodes are decisions belonging to one of the aforementioned patterns, and edges are dependencies between decisions that affect architecture enumeration (e.g., the outcome of decision 1 affects the number of alternatives available for decision 2). Then, based on a library of encoding scheme and operator fragments that are appropriate for each pattern, an algorithm automatically composes an encoding scheme and corresponding search operators by traversing the graph. The method is demonstrated in two case studies: a study integrating three architectural decisions for constructing a portfolio of earth observing satellite missions, and a study integrating eight architectural decisions for designing a guidance navigation and control system. Results suggest that this method has comparable search performance to hand-crafted formulations from experts. Furthermore, the proposed method drastically reducing the need for practitioners to write new code.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89865406","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Vivian Wen Hui Wong, M. Ferguson, K. Law, Y. T. Lee, P. Witherell
Additive manufacturing (AM) provides design flexibility and allows rapid fabrications of parts with complex geometries. The presence of internal defects, however, can lead to deficit performance of the fabricated part. X-ray Computed Tomography (XCT) is a non-destructive inspection technique often used for AM parts. Although defects within AM specimens can be identified and segmented by manually thresholding the XCT images, the process can be tedious and inefficient, and the segmentation results can be ambiguous. The variation in the shapes and appearances of defects also poses difficulty in accurately segmenting defects. This paper describes an automatic defect segmentation method using U-Net based deep convolutional neural network (CNN) architectures. Several models of U-Net variants are trained and validated on an AM XCT image dataset containing pores and cracks, achieving a best mean intersection over union (IOU) value of 0.993. Performance of various U-Net models is compared and analyzed. Specific to AM porosity segmentation with XCT images, several techniques in data augmentation and model development are introduced. This work demonstrates that, using XCT images, U-Net can be effectively applied for automatic segmentation of AM porosity with high accuracy. The method can potentially help improve quality control of AM parts in an industry setting.
{"title":"Segmentation of Additive Manufacturing Defects Using U-Net","authors":"Vivian Wen Hui Wong, M. Ferguson, K. Law, Y. T. Lee, P. Witherell","doi":"10.1115/detc2021-68885","DOIUrl":"https://doi.org/10.1115/detc2021-68885","url":null,"abstract":"\u0000 Additive manufacturing (AM) provides design flexibility and allows rapid fabrications of parts with complex geometries. The presence of internal defects, however, can lead to deficit performance of the fabricated part. X-ray Computed Tomography (XCT) is a non-destructive inspection technique often used for AM parts. Although defects within AM specimens can be identified and segmented by manually thresholding the XCT images, the process can be tedious and inefficient, and the segmentation results can be ambiguous. The variation in the shapes and appearances of defects also poses difficulty in accurately segmenting defects. This paper describes an automatic defect segmentation method using U-Net based deep convolutional neural network (CNN) architectures. Several models of U-Net variants are trained and validated on an AM XCT image dataset containing pores and cracks, achieving a best mean intersection over union (IOU) value of 0.993. Performance of various U-Net models is compared and analyzed. Specific to AM porosity segmentation with XCT images, several techniques in data augmentation and model development are introduced. This work demonstrates that, using XCT images, U-Net can be effectively applied for automatic segmentation of AM porosity with high accuracy. The method can potentially help improve quality control of AM parts in an industry setting.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":"95 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79242676","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The unstructured data available on the websites of manufacturing suppliers can provide useful insights into the technological and organizational capabilities of manufacturers. However, since the data is often represented in an unstructured form using natural language text, it is difficult to efficiently search and analyze the capability data and learn from it. The objective of this work is to propose a set of text analytics techniques to enable automated classification and ranking of suppliers based on their capability narratives. The supervised classification and semantic similarity measurement methods used in this research are supported by a formal thesaurus that uses SKOS (Simple Knowledge Organization System) for its syntax and semantics. Normalized Google Distance (NGD) was used as a metric for measuring the relatedness of terms. The proposed framework was validated experimentally using a hypothetical search scenario. The results indicate that the generated ranked list shows a high correlation with human judgment specially if the query concept vector and supplier concept vector belong to the same class. However, the correlation decreases when multiple overlapping classes of suppliers are mixed together. The findings of this research can be used to improve the precision and reliability of Capability Language Processing (CLP) tools and methods.
{"title":"Capability Language Processing (CLP): Classification and Ranking of Manufacturing Suppliers Based on Unstructured Capability Data","authors":"Kimia Zandbiglari, F. Ameri, Mohammad Javadi","doi":"10.1115/detc2021-71308","DOIUrl":"https://doi.org/10.1115/detc2021-71308","url":null,"abstract":"\u0000 The unstructured data available on the websites of manufacturing suppliers can provide useful insights into the technological and organizational capabilities of manufacturers. However, since the data is often represented in an unstructured form using natural language text, it is difficult to efficiently search and analyze the capability data and learn from it. The objective of this work is to propose a set of text analytics techniques to enable automated classification and ranking of suppliers based on their capability narratives. The supervised classification and semantic similarity measurement methods used in this research are supported by a formal thesaurus that uses SKOS (Simple Knowledge Organization System) for its syntax and semantics. Normalized Google Distance (NGD) was used as a metric for measuring the relatedness of terms. The proposed framework was validated experimentally using a hypothetical search scenario. The results indicate that the generated ranked list shows a high correlation with human judgment specially if the query concept vector and supplier concept vector belong to the same class. However, the correlation decreases when multiple overlapping classes of suppliers are mixed together. The findings of this research can be used to improve the precision and reliability of Capability Language Processing (CLP) tools and methods.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88802320","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The generation of footpaths for additive manufacturing (AM), a process commonly known as “slicing,” has a strong impact on the performance of both the associated hardware systems and the resulting objects. Available slicers invariably produce discontinuous tootpaths, featuring jumps or so-catted “travel moves” during which the deposition of material or/and energy must be hatted. For AM processes using slowly solidifying feedstock materials, such as thermosetting polymers or cementitious mixtures such as concrete, these tootpath discontinuities are highly undesirable due to the artifacts they generate. This renders existing sticers difficult to use in such applications, and presents a road-block to the adoption of AM for such material systems. In the present work, this difficulty is addressed by the development of a simple geometric criterion for the existence of continuous tool-paths that are capable of producing a specified input geometry. This development is based on the principles of morphological geometric analysis and graph theory. It is shown that, for any geometric feature with a characteristic thickness at least twice the extrusion width, a continuous toolpath exists. Furthermore, a general-purpose algorithm for continuous toolpath generation, for arbitrarily shaped objects satisfying this criterion, is developed and demonstrated on a representative test problem. Finally, conclusions and the path forward for the usage of this approach with existing AM systems is explored.
{"title":"Generation of Continuous Toolpaths for Additive Manufacturing Using Implicit Slicing","authors":"J. Steuben, J. Michopoulos, A. Iliopoulos","doi":"10.1115/detc2021-69320","DOIUrl":"https://doi.org/10.1115/detc2021-69320","url":null,"abstract":"\u0000 The generation of footpaths for additive manufacturing (AM), a process commonly known as “slicing,” has a strong impact on the performance of both the associated hardware systems and the resulting objects. Available slicers invariably produce discontinuous tootpaths, featuring jumps or so-catted “travel moves” during which the deposition of material or/and energy must be hatted. For AM processes using slowly solidifying feedstock materials, such as thermosetting polymers or cementitious mixtures such as concrete, these tootpath discontinuities are highly undesirable due to the artifacts they generate. This renders existing sticers difficult to use in such applications, and presents a road-block to the adoption of AM for such material systems. In the present work, this difficulty is addressed by the development of a simple geometric criterion for the existence of continuous tool-paths that are capable of producing a specified input geometry. This development is based on the principles of morphological geometric analysis and graph theory. It is shown that, for any geometric feature with a characteristic thickness at least twice the extrusion width, a continuous toolpath exists. Furthermore, a general-purpose algorithm for continuous toolpath generation, for arbitrarily shaped objects satisfying this criterion, is developed and demonstrated on a representative test problem. Finally, conclusions and the path forward for the usage of this approach with existing AM systems is explored.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":"26 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89223000","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
To execute non-repetitive tasks, robots need to learn on the tasks to improve task performance. The performance model cannot be built in advance for such non-repetitive tasks. The robot can execute a small portion of the task with certain process parameters and attractively update the process parameters based on the observations of the task performance. To make the learning process efficient, the trade-off between exploration and exploitation should be explicitly considered. Too much exploration may lead to the waste of time without significant improvement on task performance. On the other hand, stopping exploration prematurely may lead to suboptimal task performance. This paper describes a sequential decision making approach to select the set of parameters to improve task performance. The overall learning approach uses feasibility biased sampling, surrogate model construction and greedy optimization. We implement our approach in the simulation of robotic sanding. We also compare our method with other design of experiments methods.
{"title":"Learning to Improve Performance During Non-Repetitive Tasks Performed by Robots","authors":"Yeo Jung Yoon, Satyandra K. Gupta","doi":"10.1115/detc2021-67627","DOIUrl":"https://doi.org/10.1115/detc2021-67627","url":null,"abstract":"\u0000 To execute non-repetitive tasks, robots need to learn on the tasks to improve task performance. The performance model cannot be built in advance for such non-repetitive tasks. The robot can execute a small portion of the task with certain process parameters and attractively update the process parameters based on the observations of the task performance. To make the learning process efficient, the trade-off between exploration and exploitation should be explicitly considered. Too much exploration may lead to the waste of time without significant improvement on task performance. On the other hand, stopping exploration prematurely may lead to suboptimal task performance. This paper describes a sequential decision making approach to select the set of parameters to improve task performance. The overall learning approach uses feasibility biased sampling, surrogate model construction and greedy optimization. We implement our approach in the simulation of robotic sanding. We also compare our method with other design of experiments methods.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":"34 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74093837","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：