Assembly Instructions Research Articles

Assessing assembly instructions quality using operator behavior

Assessing assembly instructions quality using operator behavior

Generating the assembly instructions of helicopter subassemblies using the hierarchical pruning strategy and large language model

Assembly instructions are process documents in detail describing the operation steps, materials, tools, fixtures, and assembly sequences in assembly procedures. Due to assembly instructions including numerous contents, and the content being easy for workers to understand, process designers need to spend lots of time thinking and authoring assembly instructions to ensure that workers can complete the assembly task according to the assembly instructions. Focusing on the difficulties of the variety of assembly instructions and the process factors implicit in the standard languages of assembly instructions, a method of assembly instruction generation for helicopter subassemblies is proposed. First, a data representation model of multi-source heterogeneous knowledge and information based on knowledge graphs is designed and established. Then, a hierarchical pruning VF3 algorithm is presented to reuse assembly instructions according to hybrid similarity. Finally, a process factor revision model based on RoBERTa-BiLSTM-CRF is proposed to generate revised assembly instructions. Helicopter subassemblies, which contain 11,240 assembly procedures, are used to evaluate the performance of the method for generating assembly instructions. The proposed method greatly reduces the time cost of assembly instruction authoring and promotes the intelligent development of assembly process design.

Should I Evaluate My Augmented Reality System in an Industrial Environment? Investigating the Effects of Classroom and Shop Floor Settings on Guided Assembly.

Numerous prior studies have investigated real-time assembly instructions using Augmented Reality (AR). However, most such experiments were conducted in laboratory settings with simplistic assembly tasks, failing to represent real-world industrial conditions. To ascertain to what extent results obtained in a laboratory environment may differ from studies in actual industrial environments, we carried out a user study with 32 manufacturing apprentices. We compared assembly task execution results in two settings, a classroom and an industrial workshop environment. To facilitate the experiments, we developed AR-guided manual assembly systems for simple and more complex assets. Our findings reveal a significantly improved task performance in the industrial workshop, reflected in faster task completion times, fewer errors, and subjectively perceived higher flow. This contradicted participants' subjective ratings, as they expected to perform better in the classroom environment. Our results suggest that the actual manufacturing environment is critical in evaluating AR systems for real-world industrial applications.

Zero day ransomware detection with Pulse: Function classification with Transformer models and assembly language

Finding automated AI techniques to proactively defend against malware has become increasingly critical. The ability of an AI model to correctly classify novel malware is dependent on the quality of the features it is trained with and the authenticity of the features is dependent on the analysis tool. Peekaboo, a Dynamic Binary Instrumentation tool defeats evasive malware to capture its genuine behaviour. The ransomware Assembly instructions captured by Peekaboo, follow Zipf’s law, a principle also observed in natural languages, indicating Transformer models are particularly well-suited to binary classification. We propose Pulse, a novel framework for zero day ransomware detection with Transformer models and Assembly language. Pulse, trained with the Peekaboo ransomware and benign software data, uniquely identify truly new samples with high accuracy. Pulse eliminates any familiar functionality across the test and training samples, forcing the Transformer model to detect malicious behaviour based solely on context and novel Assembly instruction combinations.

Design and Assembly of a Miniature Catheter Imaging System for In Vivo Heart Endoscopic Imaging.

In this paper, we present the design and fabrication of a novel chip-on-tip catheter, which uses a microcamera and optical fibres to capture in vivo images in a beating porcine heart thanks to a saline flush to clear the blood field. Here, we demonstrate the medical utility and mechanical robustness of this catheter platform system, which could be used for other optical diagnostic techniques, surgical guidance, and clinical navigation. We also discuss some of the challenges and system requirements associated with developing a miniature prototype for such a study and present assembly instructions. Methods of clearing the blood field are discussed, including an integrated flush channel at the distal end. This permits the capture of images of the endocardial walls. The device was navigated under fluoroscopic guiding, through a guiding catheter to various locations of the heart, where images were successfully acquired. Images were captured at the intra-atrial septum, in the left atrium after a trans-septal cross procedure, and in the left ventricle, which are, to the best of our knowledge, the first images captured in an in vivo beating heart using endoscopic techniques.

Empirische Ergebnisse zum Einsatz von Assistenzsystemen in der Montage

Abstract As part of the survey, data was collected from 158 study participants in order to gain an overview of the importance of the technologies used in assembly and the objectives pursued with their use. The results of the survey indicate that assistance systems are considered important by the respondents and will be part of the original equipment in new assembly areas in the future. Technologies that are already established include the provision of digital assembly instructions via a screen and pick-by-light, which are linked to the relevant experience. In addition to these technologies, human-robot collaboration (HRC) and the provision of visual information by means of projection or data glasses, possibly supported by augmented reality, will also become established in the future. At present and in the future, the main objectives will be to reduce work errors, increase productivity, reduce the training time for new employees, and cut assembly costs.

MDGraph: A novel malware detection method based on memory dump and graph neural network

Malware detection is of great importance to computer security. Although the malware detection approaches have made great progress in recent years, these methods are still limited in regard to identifying the advanced malware that conceals their malicious activities. To address this problem, we present a novel malware detection solution, MDGraph, which is based on the memory dump and graph neural network. MDGraph first dynamically grabs a memory dump file for a target process. Then, it applies the recursive disassembling technique to extract the program functions composed of assembly instruction sequences and the invocation relationship between functions from the memory dump. Next, the program functions are vectorized using the doc2vec model. Based on the vectorized functions and their connections, MDGraph leverages a graph neural network model for malware detection. The evaluation shows our method can identify unpacked and packed malware effectively, and it is superior to the recent malware detection methods based on the memory dump.

A multi-modal context-aware sequence stage validation for human-centric AR assembly

Augmented reality (AR) has demonstrated its superior performance to benefit manual assembly tasks by delivering intuitive guidance on the workbench directly, alleviating the mental load while leading to time-saving operations. Nevertheless, current AR-assisted assembly mainly focuses on superimposing visual instructions onto real scenarios and assumes that the worker performs correctly as instructed, ignoring the confirmation of the actual assembly execution processes, and it is still difficult to avoid operating errors on the shop floor. To this end, this paper aims to propose a multi-modal context-aware on-site assembly stage recognition for human-centric AR assembly. Firstly, a sim-real point cloud-based semantic understanding method for assembly stage identification is presented, which can recognize the current sequence stage during the AR assembly process even when encountering weakly textured workpieces. In addition, a 2D image-based semantic recognition for on-site images is applied as compensation from the RGBD camera, resulting in a robust multi-modal context-aware assembly stage validation for the ongoing AR assembly tasks. Followed by a context-aware closed-loop AR assembly system that provides actual assembly result confirmation automatically, relieving the mental load for workers to activate the next assembly instruction, as well as confirm the current status during the actual operation. Finally, extensive experiments are carried out, and the results illustrate that the proposed context-aware AR assembly system can monitor the on-site sequence stage while providing a human-centric AR assembly action.

O-114 Instructions for assembly of early mammalian embryos

Abstract The Plachta lab has pioneered the use of live-imaging approaches to discover how the preimplantation embryo forms in real time. Our recent findings help to transform the field of preimplantation development by showing how diverse processes occurring at the level of transcription dynamics, cytoskeletal organization, and the action of mechanical forces are integrated in real time to ensure proper embryo development. We extend our studies from mouse to human embryos, which enabled us to discover differences in embryo patterning between species and some of the events causing developmental errors, such as aneuploidy within the intact-developing system.

Digital Customization for Product Design and Manufacturing: A Case Study within the Furniture Industry

Computational design together with the digitization of most fabrication processes play an important role in many research areas. Digital tools such as 3D modeling and computational design have been increasingly used. Computational design combines traditional 3D product design together with programming a general-purpose CAD system in order to promote system integration. In essence, using CAD-based textual or visual programming languages a series of products can be designed with accuracy and take advantage of product customization and automation of downstream applications. The present paper aims at customizing furniture design based on automating both the design and the fabrication procedures. The customer is able to define a series of geometrical characteristics, i.e., width, length, internal dimensions, and various other properties. The outcome consists of automating a great deal of processes, i.e., 3D modeling and assembling, visualization, creating the bill of materials (BOM), producing assembly instructions for the user, drawings and prototyping files, weight estimation.

Design and Development of Self Balancing Two Wheeler

The Self-balancing robots represent a fascinating intersection of technology and practical application. This project aims to design and develop a self-balancing two- wheeler robot capable of maintaining stability and balance autonomously. Drawing inspiration from principles observed in human balance and locomotion, the robot employs a combination of sensors, actuators, and control algorithms to achieve its objective. The project begins with an in-depth literature survey, exploring existing research and methodologies in the field of self-balancing robotics. Leveraging insights from key publications, the project outlines fundamental principles, sensor fusion techniques, advanced control methodologies, and challenges inherent in self-balancing robot design. Key components including an Arduino Uno board, accelerometers, gyroscopes, DC motors, batteries, and optional reaction wheels are carefully selected and integrated into the robot's hardware architecture. A systematic approach to hardware selection, assembly, sensor integration, microcontroller programming, and motor control ensures a coherent and functional design. Throughout the development process, emphasis is placed on rigorous testing, calibration, and iterative improvement. The robot is subjected to various environmental conditions and scenarios to validate its stability and performance. Calibration procedures refine sensor accuracy and control algorithms, enhancing the robot's ability to maintain balance under dynamic conditions. The project culminates in the creation of a comprehensive documentation package, detailing schematics, code explanations, assembly instructions, and testing methodologies. Additionally, a presentation showcases the robot's functionality, design rationale, and potential applications. By meticulously following the outlined steps and leveraging insights from existing research, this project aims to contribute to the advancement of self-balancing robotics and serve as a foundation for future exploration and innovation in this exciting field

Development of an Educational Kit Using CAD Software for Simple Machine Learning

Traditional simple machine topic is frequently theoretical and does not provide enough hands-on practise. Students may as a result fail to comprehend the fundamentals of simple machines and their applications. This research presents an innovative educational kit for simple machine learning. Students can utilize this kit to understand the fundamental concepts behind simple machines like levers, gears, and pulleys. This study investigates the creation of a simple machine kit using CAD software as a teaching tool for fundamental mechanics. A survey study is employed as part of this technique in order to ascertain audience needs and expectations. Beginning with idea generation and followed by rigorous evaluation, the process of developing the kit focuses on aligning pedagogical viability, marketability, and profitability. The selected concept is then transformed into a detailed technical drawing specifying dimensions, components, and assembly instructions, ensuring coherence between educational goals and machine learning integration. Iterative prototyping and testing further refine the kit's design, ensuring seamless interaction between machine learning elements and LEGO-based components. This approach culminates in an innovative educational kit that effectively imparts foundational machine learning principles through an engaging platform. The simple machine kit made through this research can then be utilized as an educational tool in primary, secondary, or higher education environments to facilitate students' grasp of basic mechanics. Students learning product design may find this novel approach both enjoyable and useful in their studies.

MRNA: a Physiological Multipurpose Tool for Hitherto Hard-to-treat Neurological and Neuro-Ophthalmological Diseases

Interdisciplinary translational research between physiologists and clinicians based on mRNA technology has opened and shall open treatment perspectives for a broad range of diseases, many of them hitherto hard to treat or not treatable at all. In this overview, various options for applications of mRNA technology are presented for the treatment of quite different neurological and neuro-ophthalmological disease categories such as metabolic and neurodegenerative diseases, infectious diseases, and tumors, illustrated by a few selected examples of current clinical trials and experimental approaches. mRNA technology allows for developing tailor-made individualized therapies by which the body may even produce the remedies itself according to the physiological recipe of the mRNA assembly instructions.

NeuroPorator: An open-source, current-limited electroporator for safe in utero gene transfer

BackgroundElectroporation is an effective technique for genetic manipulation of cells, both in vitro and in vivo. In utero electroporation (IUE) is a special case, which represents a fine application of this technique to genetically modify specific tissues of embryos during prenatal development. Commercially available electroporators are expensive and not fully customizable. We have designed and produced an inexpensive, open-design, and customizable electroporator optimized for safe IUE. We introduce NeuroPorator. MethodWe used off-the-shelf electrical parts, a single-board microcontroller, and a cheap data logger to build an open-design electroporator. We included a safety circuit to limit the applied electrical current to protect the embryos. We added full documentation, design files, and assembly instructions. ResultNeuroPorator output is on par with commercially available devices. Furthermore, the adjustable current limiter protects both the embryos and the uterus from overcurrent damage. A built-in data acquisition module provides real-time visualization and recordings of the actual voltage/current pulses applied to each embryo. Function of NeuroPorator has been demonstrated by inducing focal cortical dysplasia in mice. Significance and conclusionThe simple and fully open design enables quick and cheap construction of the device and facilitates further customization. The features of NeuroPorator can accelerate the IUE technique implementation in any laboratory and speed up its learning curve.

Automated arduino robot design with multi-level natural language processing chains

This research explores the potential of leveraging OpenAIs GPT-3.5-Turbo API for automating Arduino robot design through a structured multi-level language processing chain termed LangChain. The system breaks down the design process into six stages, from preliminary design sketching to failure analysis. Each stage generates robot design components using user inputs, progressively building upon the previous outputs. The results highlight the models capabilities in generating functional preliminary designs, suggesting hardware and software components, and offering assembly instructions. While demonstrating a commendable level of technical knowledge, the system requires further refinement in numerical analysis and design reviews. This approach provides a foundational framework for bridging the knowledge gap between amateurs and professionals in robotic design.

Assembly Function Recognition in Embedded Systems as an Optimization Problem

Many different aspects of software system development and verification rely on precise function identification in binary code. Recognition of the source Assembly functions in embedded systems is one of the fundamental challenges in binary program analysis. While numerous approaches assume that the functions are given a priori, correct identification of the functions in binaries remains a great issue. This contribution addresses the problem of uncertainty in binary code in identification of functions, which were optimized during compilation. This paper investigates the difference between debug and optimized functions via modeling of these functions. To do so, we introduce an extensible model-centred hands-on approach for examining similarities between binary functions. The main idea is to model each function using a set of predetermined, experimentally discovered features, and then find a suitable weight vector that could give impact factor to each such a feature. After finding the weight vector, the introduced models of such desired functions can be identified in binary software packages. It means that we reduce the similarity identification problem of the models to a classical version of optimization problems with one optimization criterion. Using our implementation, we found that the proposed approach works smoothly for functions, which contain at least ten Assembly instructions. Our tool guarantees success at a very high level.

Malware2ATT&CK: A sophisticated model for mapping malware to ATT&CK techniques

MITER Corporation presents the ATT&CK Matrix, which maps malware behavior to different tactics, techniques, and procedures (TTPs) providing a comprehensive view to clarify the inner mechanism of malware more accurately. However, manual mapping methods are time-consuming, while rule-based feature-based mapping methods often under- or misreport many attacks. Inspired by the successful application of image multi-label classification techniques, we propose a method called Malware2ATT&CK to automatically map malware to ATT&CK techniques. The method applies pre-trained models to extract features from the two pieces of information — the static analysis information of assembly instruction and API calls from malware. The malicious techniques are identified by a multi-label classifier based on the graph neural network and knowledge graph. In the experiments over two test sets, Malware2ATT&CK shows excellent performance achieving an average F1 score of 83.6% in the technology mapping task. Further evaluation indicates that the high accuracy of the prediction is due to our ability to accurately capture the correlation between behaviors.

MouseVUER: video based open-source system for laboratory mouse home-cage monitoring

Video monitoring of mice in the home-cage reveals behavior profiles without the disruptions caused by specialized test setups and makes it possible to quantify changes in behavior patterns continually over long time frames. Several commercial home-cage monitoring systems are available with varying costs and capabilities; however there are currently no open-source systems for home-cage monitoring. We present an open-source system for top-down video monitoring of research mice in a slightly modified home-cage. The system is designed for integration with Allentown NexGen ventilated racks and allows unobstructed view of up to three mice, but can also be operated outside the rack. The system has an easy to duplicate and assemble home-cage design along with a video acquisition solution. The system utilizes a depth video camera, and we demonstrate the robustness of depth video for home-cage mice monitoring. For researchers without access to Allentown NexGen ventilated racks, we provide designs and assembly instructions for a standalone non-ventilated rack solution that holds three systems for more compact and efficient housing. We make all the design files, along with detailed assembly and installation instructions, available on the project webpage (https://github.com/NIH-CIT-OIR-SPIS/MouseVUER).

Recognition of Hand Activities for Automatic Generation of Assembly Instructions

The manual generation of instructions for assembly processes requires time and effort. This is especially true for assembly processes that are characterized by a high variance of products at low quantities, as they are typically found in small and medium-sized enterprises (SMEs). Consequently, in SMEs, expert knowledge about the correct assembly of products is often only implicitly available among few experienced workers. Due to demographic changes, capturing this implicit knowledge becomes increasingly relevant to prevent expertise loss. A promising approach to reduce the generation effort is the automatic generation of assembly instructions based on process recordings. In prior work, the authors presented a concept of a system for automated instruction generation. The system processes video recordings of assembly processes with several independent modules for object, hand, and activity recognition and converts the results into assembly instructions. Among the individual processing modules, the hand activity recognition plays a special role, since the recognized activities form an important basis for the final assembly instruction. In this work, the utilization of Graph Convolutional Networks (GCNs) for the recognition of basic hand activities, such as rotating, is presented. Since the system for generating assembly instructions is intended to operate on arbitrary manual assembly use cases, the authors investigate the transferability of the proposed approach by training GCN models on the publicly available Assembly101 dataset and own recordings. The trained models are evaluated on respective test sets, mutually, and on a real-life example to assess the applicability of the proposed approach in practice. The identified challenges for reliable activity detection are discussed and offer important insights for future work.

Framework for Propagating Product Variability to Digital Assembly Instructions

As the amount of product iterations within high variety low volume assembly environments increases, the need for efficient management of variability and engineering changes within assembly instructions rises. This work proposes a methodology and a system architecture that facilitates the interplay between product lifecycle management (PLM) systems and digital assembly instructions, using an ISA-95 based data model. The framework consists of three main components: (1) a product variability model that integrates PLM information, (2) a digital instructions authoring tool, and (3) an assembly instruction variability model that connects to instruction platforms. Finally, the approach is applied to a validation case.