SEARCH · Engineering Papers
Results for “machine vision”
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Welding technology transfer task/laser based weld joint tracking system for compressor girth welds
Sensors to control and monitor welding operations are currently being developed at Marshall Space Flight Center. The laser based weld bead profiler/torch rotation sensor was modified to provide a weld joint tracking system for compressor girth welds. The tracking system features a precision laser based vision sensor, automated two-axis machine motion, and an industrial PC controller. The system benefits are elimination of weld repairs caused by joint tracking errors which reduces manufacturing costs and increases production output, simplification of tooling, and free costly manufacturing floor space.
NASA Centennial Challenge: Three Dimensional (3D) Printed Habitat, Phase 2
The NASA Centennial Challenges: 3D-Printed Habitat Challenge seeks to develop the fundamental technologies necessary to manufacture an off-world habitat using mission recycled materials and/or local indigenous materials. The vision is that autonomous habitat manufacturing machines will someday be deployed to the Moon or Mars to construct shelters for human habitation. NASA and Bradley University, are holding a new US$ 2.5 million competition to design and build a 3-D printed habitat for deep space exploration, including the agency's journey to Mars. The multi-phase 3-D Printed Habitat Challenge, part of NASA's Centennial Challenges program, is designed to advance the additive construction technology needed to create sustainable housing solutions for Earth and beyond. The first phase of the competition ran through Sept. 27, 2015. This phase, a design competition, called on participants to develop state-of-the-art architectural concepts that take advantage of the unique capabilities 3-D printing offers. The top 3 prizes with a prize purse of $40,000 were awarded at the 2015 World Maker Faire in New York. The second phase of the competition is called the Structural Member Competition and it is divided into three levels happening in the spring and summer of 2017. The Compression Test Competition (Level 1) focuses on the fabrication technologies needed to manufacture structural components from a combination of indigenous materials and recyclables, or indigenous materials alone. For Level 1, teams will develop 3D printable materials, build a 3D printing machine, and print two specimens: a truncated cone and a cylinder. The Level 2 Beam Member Competition is the second of three sub-competitions within the overall Structural Member Competition. For Level 2, teams will print a beam that will be tested. The Level 3 Head to Head Competition is the third of three sub-competitions within the overall Structural Member Competition. For Level 3, teams will develop 3D printable materials, use a 3D printing machine, and print three compression specimens of the elected material, three flexural specimens of the elected material, and one dome structure. Tests conducted on the specimens and the dome structure will determine Level 3 scores and awards. On Earth these same habitat manufacturing capabilities could be used to produce housing wherever affordable housing is needed and access to conventional building materials and skills is limited. Terrestrially, it is envisioned that local indigenous materials (dirt, clay, sand, etc.) could be combined with readily available recyclable materials and used to construct semi-permanent shelters against environmental elements for human habitation. The goal of the 3D-Printed Habitat Challenge is to foster the development of new technologies necessary to additively manufacture a habitat using local indigenous materials with, or without, recyclable materials. This paper will summarize the Level 2 results of this NASA Centennial Challenge competition and it will discuss related technology advancement.
Impact Ice Microstructure Segmentation Using Transfer Learned Model
A process of using machine learning to segment impact ice microstructure is presented and analyzed. The segmentation was conducted with the goal of obtaining average grain size estimations. The model was trained on a set of micrographs of impact ice grown at NASA Glenn’s Icing Research Tunnel. The model leveraged a model pre-trained on a large set of micrographs of various materials as a starting point. Post-processing of the segmented images was done to connect broken boundaries. An automatic method of determining grain size following an ASTM standard was implemented. Segmentation results using different training sets as well as different encoder and decoder pairs are presented. Calculated sizes are compared to manual grain size measurement methods. Results show promise in accuracy as well as a possible improvement in repeatability and consistency. Next steps for improving the model are suggested.
Future Model-Based Systems Engineering Vision and Strategy Bridge for NASA
A vision for the future of model-based systems engineering (MBSE) at NASA in 2029 and a strategy bridge towards that future are presented. Strategic thinking and leading change concepts were used to analyze reports and presentations on global trends and visionary thinking about the future of systems and digital engineering. The context, strategic time horizon, stakeholders, strategic challenges, strategic advantages, driving forces, and opportunities were considered. The analysis resulted in a future vision of MBSE that shows what NASA systems engineers and digital machines will do to perform rapid, extraordinary, and unprecedented missions. The NASA systems engineer, in this future vision, works with a global project team in a virtual and collaborative environment, engineers the system, and uses digital approaches as the routine and default way of working. The digital machines provide data-driven and automated mission designs; have a backbone of program and project management, systems engineering, and product life-cycle management; and are a knowledge-sharing infrastructure. The NASA systems engineer and the systems engineering team are envisioned to use digital machines to plan and perform rapid exploration missions, develop a digital twin that lasts across the life cycle, and develop enduring and adaptable systems. NASA has an engineering enterprise and a life-cycle management framework that endure, adapt, and respond. A strategy bridge based on the Baldrige Criteria for Performance Excellence Framework and lessons learned from a recent MBSE initiative illuminates a way forward from today to this desired future. The bridge lays out a strategy for leaders and recommends investments of today for immediate benefits and for benefits in 2029.
Human factors issues in the use of night vision devices
An account is given of the critical human factors that arise in field data on the differences between night vision displays and unaided day vision. Attention is given to the findings of empirical studies of performance on rotorcraft-flight-relevant perceptual tasks in which depth and distance perception are critical factors. Suggestions are made for man-machine-critical component design modifications in current night vision systems.
Affordable Artificial Intelligence-Assisted Machine Supervision System for the Small and Medium-Sized Manufacturers
With the rapid concurrent advance of artificial intelligence (AI) and Internet of Things (IoT) technology, manufacturing environments are being upgraded or equipped with a smart and connected infrastructure that empowers workers and supervisors to optimize manufacturing workflow and processes for improved energy efficiency, equipment reliability, quality, safety, and productivity. This challenges capital cost and complexity for many small and medium-sized manufacturers (SMMs) who heavily rely on people to supervise manufacturing processes and facilities. This research aims to create an affordable, scalable, accessible, and portable (ASAP) solution to automate the supervision of manufacturing processes. The proposed approach seeks to reduce the cost and complexity of smart manufacturing deployment for SMMs through the deployment of consumer-grade electronics and a novel AI development methodology. The proposed system, AI-assisted Machine Supervision (AIMS), provides SMMs with two major subsystems: direct machine monitoring (DMM) and human-machine interaction monitoring (HIM). The AIMS system was evaluated and validated with a case study in 3D printing through the affordable AI accelerator solution of the vision processing unit (VPU).
Safeguards-Informed Hybrid Imagery Dataset [Poster]
Deep Learning computer vision models require many thousands of properly labelled images for training, which is especially challenging for safeguards and nonproliferation, given that safeguards-relevant images are typically rare due to the sensitivity and limited availability of the technologies. Creating relevant images through real-world staging is costly and limiting in scope. Expert-labeling is expensive, time consuming, and error prone. We aim to develop a data set of both realworld and synthetic images that are relevant to the nuclear safeguards domain that can be used to support multiple data science research questions. In the process of developing this data, we aim to develop a novel workflow to validate synthetic images using machine learning explainability methods, testing among multiple computer vision algorithms, and iterative synthetic data rendering. We will deliver one million images – both real-world and synthetically rendered – of two types uranium storage and transportation containers with labelled ground truth and associated adversarial examples.
Popnet : computer vision based deep learning model for forecasting gridded population
Here, this study introduces Popnet, a deep learning model for forecasting 1 km-gridded populations, integrating U-Net, ConvLSTM, a Spatial Autocorrelation module and deep ensemble methods. Using spatial variables and population data from 2000 to 2020, Popnet predicts South Korea’s population trends by age groups (under 14, 15-64 and over 65) up to 2040. In validation, it outperforms traditional machine learning and state-of-the-art computer vision models. The output of this model discovered significant polarisation: population growth in urban areas, especially the capital region, and severe depopulation in rural areas. Popnet is a robust tool for offering significant insights to policymakers and related stakeholders about the detailed future population, which allows them to establish detailed, localised planning and resource allocations.
Advancing Vision-based Feedback and Convolutional Neural Networks for Visual Outlier Detection
Machine learning has matured into a technology that has immediate applicability to the surveillance needs of nuclear material storage containers. These containers at LANL are the barrier preventing release of radioactive material to the workers, public, and environment during the storage period of the material. Annual surveillance activities can only provide coverage on a handful of containers. There is a significant need for surveillance tools to identify potential issues and precursors to containment failure that can be used during opportunistic inspections and, more generally, outside of annual surveillance activities. In this report we provide details on the advancement of our proposed embodiment that combines an automation system for taking pictures and a high-accuracy machine learning-driven object detection software. We further showcase the improvements on the software side with progress on extracting unique identification features and advances in detecting damage. The current state of the system captures subject matter expert training and a space-conscious design whose implementation is envisioned in the near future.
Contaminant Investigation and Pre‐Processing Opportunities for Textile‐To‐Textile Recycling
Millions of metric tons of textiles are landfilled or incinerated each year in the United States, with less than 1% of textiles recycled into new clothing or fabrics. To counter this trend, a growing number of companies and researchers are exploring how a circular economy can be applied to support textile‐to‐textile recycling. A significant barrier they face comes down to quickly and efficiently extracting pure feedstock material from post‐consumer garments that feature a mix of natural and synthetic fibers. Textile recyclers prefer pure feedstocks, as working with mixed sources typically means lower throughput, higher risk of equipment failure, and diminished business margins. To facilitate a circular economy for textiles, methods, and technologies are needed that can efficiently separate out materials and contaminants from end‐of‐life textiles to increase the flow of pure feedstocks to recyclers. This paper summarizes findings from interviews with a cross section of textile recyclers and from a review of literature to define basic feedstock requirements. In addition to our qualitative research, we deconstruct a bale of post‐consumer textiles and analyze them using computer‐vision imaging, Fourier transform infrared spectroscopy (FTIR), and machine learning. The resulting data are used to set system‐level design inputs for an automated contaminant removal system to process post‐consumer clothing into appropriate feedstocks for recycling. To set the system's levels for automated real‐time near‐infrared analysis, we identify the minimum percentage of primary material that any single garment in a load of used clothing must contain for the average of the full output stream to meet the target purity levels of recyclers. Here, the envisioned automated system can also address undesirable trace materials that might contaminate the processed stream by using imaging cameras coupled with artificial intelligence to identify sections of clothing for de‐trimming. Proof‐of‐concept machine learning algorithms are evaluated to locate and identify trims or garment areas with hidden contaminant materials. Integrating these methods into automated textile cutting systems can provide a cost‐effective means for increasing feedstock purity from used clothing, which can advance circularity for textiles by helping recyclers to reach production volumes and quality targets that were not possible solely with manual dismantling operations.
Utilizing computer vision and artificial intelligence algorithms to predict and design the mechanical compression response of direct ink write 3D printed foam replacement structures
Additive Manufacturing (AM) of porous polymeric materials, such as foams, recently became a topic of intensive research due their unique combination of low density, impressive mechanical properties, and stress dissipation capabilities. Conventional methods for fabricating foams rely on complex and stochastic processes, making it challenging to achieve precise architectural control of structured porosity. In contrast, AM provides access to a wide range of printable materials, where precise spatial control over structured porosity can be modulated during the fabrication process enabling the production of foam replacement structures (FRS). Current approaches for designing FRS are based on intuitive understanding of their properties or an extensive number of finite element method (FEM) simulations. These approaches, however, are computationally expensive and time consuming. As such, in this work, we present a novel methodology for determining the mechanical compression response of direct ink write (DIW) 3D printed FRS using a simple cross-sectional image. By obtaining measurement data for a relatively small number of samples, an artificial neural network (ANN) was trained, and a computer vision algorithm was used to make inferences about foam compression characteristics from a single cross-sectional image. Finally, a genetic algorithm (GA) was used to solve the inverse design problem, generating the AM printing parameters that an engineer should use to achieve a desired compression response from a DIW printed FRS. The methods developed herein present an avenue for entirely autonomous design and analysis of additively manufactured structures using artificial intelligence.
Regulation compliant AI for fusion: explainable image-based feedback control of divertor detachment in DIII-D tokamak
While artificial intelligence (AI) has been promising for fusion control, its inherent black-box nature will make compliant implementation in regulatory environments a challenge. This study implements and validates a real-time AI-enabled linear and interpretable control system for successful divertor detachment control with the DIII-D lower divertor camera. Using D 2 gas, we demonstrate successful feedback divertor detachment control with a mean absolute difference of 2% from the target for both detachment and reattachment. This automatic training and linear processing framework can be extended to any image-based diagnostic for future fusion reactors.
Real-time semantic segmentation on FPGAs for autonomous vehicles with hls4ml
In this paper, we investigate how field programmable gate arrays can serve as hardware accelerators for real-time semantic segmentation tasks relevant for autonomous driving. Considering compressed versions of the ENet convolutional neural network architecture, we demonstrate a fully-on-chip deployment with a latency of 4.9 ms per image, using less than 30% of the available resources on a Xilinx ZCU102 evaluation board. The latency is reduced to 3 ms per image when increasing the batch size to ten, corresponding to the use case where the autonomous vehicle receives inputs from multiple cameras simultaneously. We show, through aggressive filter reduction and heterogeneous quantization-aware training, and an optimized implementation of convolutional layers, that the power consumption and resource utilization can be significantly reduced while maintaining accuracy on the Cityscapes dataset.
Event-to-Video Conversion for Overhead Object Detection
Collecting overhead imagery using an event camera is desirable due to the energy efficiency of the image sensor compared to standard cameras. However, event cameras complicate downstream image processing, especially for complex tasks such as object detection. In this paper, we investigate the viability of event streams for overhead object detection. We demonstrate that across a number of standard modeling approaches, there is a significant gap in performance between dense event representations and corresponding RGB frames. We establish that this gap is, in part, due to a lack of overlap between the event representations and the pre-training data that the object detectors were initially trained on through a number of experiments. Then, apply an off-the-shelf event-to-video conversion tool that converts event streams into gray-scale video to close this gap. We demonstrate that this approach results in a large performance increase, outperforming even event-specific object detection techniques on our overhead target task. These results suggest that better aligning event representations with existing large pre-trained models may result in greater short-term performance gains compared to end-to-end event-specific architectural improvements.
Identifying Critical Infrastructure in Imagery Data Using Explainable Convolutional Neural Networks
To date, no method utilizing satellite imagery exists for detailing the locations and functions of critical infrastructure across the United States, making response to natural disasters and other events challenging due to complex infrastructural interdependencies. This paper presents a repeatable, transferable, and explainable method for critical infrastructure analysis and implementation of a robust model for critical infrastructure detection in satellite imagery. This model consists of a DenseNet-161 convolutional neural network, pretrained with the ImageNet database. The model was provided additional training with a custom dataset, containing nine infrastructure classes. The resultant analysis achieved an overall accuracy of 90%, with the highest accuracy for airports (97%), hydroelectric dams (96%), solar farms (94%), substations (91%), potable water tanks (93%), and hospitals (93%). Critical infrastructure types with relatively low accuracy are likely influenced by data commonality between similar infrastructure components for petroleum terminals (86%), water treatment plants (78%), and natural gas generation (78%). Local interpretable model-agnostic explanations (LIME) was integrated into the overall modeling pipeline to establish trust for users in critical infrastructure applications. The results demonstrate the effectiveness of a convolutional neural network approach for critical infrastructure identification, with higher than 90% accuracy in identifying six of the critical infrastructure facility types.
Systematic Evaluation of Backdoor Data Poisoning Attacks on Image Classifiers
Backdoor data poisoning attacks have recently been demonstrated in computer vision research as a potential safety risk for machine learning (ML) systems. Traditional data poisoning attacks manipulate training data to induce unreliability of an ML model, whereas backdoor data poisoning attacks maintain system performance unless the MLmodel is presented with an input containing an embedded“trigger” that provides a predetermined response advantageous to the adversary. Our work builds upon prior back-door data-poisoning research for ML image classifiers and systematically assesses different experimental conditions including types of trigger patterns, persistence of trigger patterns during retraining, poisoning strategies, architectures (ResNet-50, NasNet, NasNet-Mobile), datasets (Flowers, CIFAR-10), and potential defensive regularization techniques (Contrastive Loss, Logit Squeezing, Manifold Mixup,Soft-Nearest-Neighbors Loss). Experiments yield four key findings. First, the success rate of backdoor poisoning at-tacks varies widely, depending on several factors, including model architecture, trigger pattern and regularization technique. Second, we find that poisoned models are hard to detect through performance inspection alone. Third, regularization typically reduces backdoor success rate, although it can have no effect or even slightly increase it, depending on the form of regularization. Finally, backdoors inserted through data poisoning can be rendered ineffective after just a few epochs of additional training on a small set of clean data without affecting the model’s performance.
Artificial Intelligence/Machine Learning Data Management Software System for NSRCs: Phase I Final Technical Report
Explore the source record for details and available documents.