EAI Endorsed Transactions on AI and Robotics

Designing Automation for Pickup and Delivery Tasks in Modern Warehouses Using Multi Agent Path Finding (MAPF) and Multi Agent Reinforcement Learning (MARL) Based Approaches

Shambhavi Mishra — 2024-03-18

A warehouse pickup and delivery problem finds its solution using multi agent path finding (MAPF) approach. Also, the problem has been used to showcase the capabilities of the multi agent reinforcement learning (MARL). The warehouse pickup and delivery work needs the agent to pick up a requested item and successfully deliver it to the intended location within the warehouse. The problem has been solved based on two approaches that include single shot and lifelong problem solution. The single shot solution has the delivery as the final goal and thus once it reaches the delivery address, it stops whereas in case of lifelong, the agent needs to deliver the item which it had picked, deliver it to the required place and then again pick up new item until requests are satisfied. The strategy used by multi agent path finding (MAPF) approach aims at constructing collision free paths to reach the delivery location but in case of multi agent reinforcement learning (MARL), the agents’ decision making tactics (or policies) are learned which are then used to help agents decide path to be followed based on environment state and agent’s position. The results show that the lifelong conflict based search (CBS) is a better option when the agents are less in number as in that case, the re-planning will take overall less time but when the agents are large in number then this re-planning can take very long to produce conflict free paths from source to goal nodes. In this case, shared experience action critic (SEAC) which is based on multi agent reinforcement learning (MARL) approach can be more efficient choice as it takes the current environment state to give the most suitable action for that time t. For this study the agents taken for learning are homogeneous in nature that can pickup and deliver any type of requested item. We can address the same pickup and delivery problem when the agents are not all same and differ in their capabilities and the type of item they can handle.

Implementation of GPT models for Text Generation in Healthcare Domain

Anirban Karak — 2024-04-09

INTRODUCTION: This paper highlights the potential of using generalized language models to extract structured texts from natural language descriptions of workflows in various industries like healthcare domain

OBJECTIVES: Despite the criticality of these workflows to the business, they are often not fully automated or formally specified. Instead, employees may rely on natural language documents to describe the procedures. Text generation methods offer a way to extract structured plans from these natural language documents, which can then be used by an automated system.

METHODS: This paper explores the effectiveness of using generalized language models, such as GPT-2, to perform text generation directly from these texts

RESULTS: These models have already shown success in multiple text generation tasks, and the paper's initial results suggest that they could also be effective in text generation in healthcare domain. In fact, the paper demonstrates that GPT-2 can generate comparable results to many current text generation methods.

CONCLUSION: This suggests that generalized language models can increase the efficiency and accuracy in text generation, where workflows are repetitive and sequential.

A Comparison of the Performance of Six Machine Learning Algorithms for Fake News

Rafah H Al-Furaiji — 2024-03-20

INTRODUCTION: This research focuses on the increasing importance of social media websites as versatile platforms for entertainment, work, communication, commerce, and accessing global news. However, it emphasizes the need to use this power responsibly.

OBJECTIVES: The objective of the study is to evaluate the performance of artificial intelligence algorithms in detecting fake news.

METHODS: Through a comparison of six machine learning algorithms and the use of natural language processing techniques,

RESULTS: The study identifies four algorithms with a 99% accuracy rate in detecting fake news.

CONCLUSION: The results demonstrate the effectiveness of the proposed method in enhancing the performance of artificial intelligence algorithms in addressing the problem of fake news detection.

Improved Hybrid Preprocessing Technique for Effective Segmentation of Wheat Canopies in Chlorophyll Fluorescence Images

Ankita Gupta — 2024-01-08

Precision agriculture heavily relies on accurately segmenting wheat canopies from chlorophyll fluorescence (CHF) images. However, these images often face challenges due to inherent noise and illumination variations, primarily induced by the thermal activity of photons emitting a fluorescence effect. The unique nature of fluorescence introduces variations in illumination, especially during the crop's dark adaptation before experimentation. This adaptation aims to capture the full fluorescence effect, starting from minimum fluorescence and progressing to maximum fluorescence. In the initial stages of fluorescence, images tend to appear darker compared to those progressing towards maximum fluorescence. This variability necessitates the development of a sophisticated hybrid approach to eliminate noise and enhance contrast collaboratively, maximizing the benefits derived from CHF images. This paper introduces a novel hybrid preprocessing approach designed to address these challenges. The proposed method integrates five denoising techniques, namely Discrete Cosine Transform, Block Matching-3D, Low-Rank Matrix Approximation, Wiener Filtering, and Median Filtering, to mitigate the impact of noise in CHF images. Simultaneously, two enhancement techniques, Adaptive Histogram Specification and Gamma Correction, are employed to accentuate critical features, compensating for inherent variations in illumination during the fluorescence process. The hybrid preprocessing technique was proposed after analysing different combinations of denoising and enhancement techniques. Through qualitative and quantitative analysis of the results, it was observed that Block Matching-3D denoising with Gamma Correction produced the best output, with an Average PSNR of 0.54 and Average MSE of 0.07. This cascaded approach not only emphasizes noise reduction but also prioritizes the enhancement of crucial information within CHF images. By synergistically combining denoising and enhancement methods, the proposed approach optimizes the overall quality of the images, laying a foundation for improved wheat canopy segmentation. This research contributes a comprehensive and innovative solution to the challenges associated with CHF images in precision agriculture. The proposed hybrid approach holds promise for advancing the accuracy and reliability of wheat canopy segmentation, thereby enhancing the efficacy of precision agricultural practices.

Utilizing Fundamental Analysis to Predict Stock Prices

Akshay Khanpuri — 2024-03-22

Portfolio management involves the critical task of determining the optimal times to enter or exit a stock in order to maximize profits in the stock market. Unfortunately, many retail investors struggle with this task due to unclear investment objectives and a lack of a structured decision-making process. With the vast number of stocks available in the market, it can be difficult for investors to determine which stocks to invest in. As a result, there is a growing need for the development of effective investment decision support systems to assist investors in making informed decisions. Researchers have explored various approaches to building such systems, including predicting stock prices using sentiment analysis of news, articles, and social media, as well as historical trends and patterns. However, the impact of financial reports filed by companies on stock prices has not been extensively studied. This paper aims to address this gap by using machine learning techniques to develop a more accurate stock prediction model based on financial reports from companies in the Nifty 50. The financial reports considered include quarterly reports, annual reports, cash flow statements, and ratios.

Improving recognition accuracy for facial expressions using scattering wavelet

Mehdi Davari — 2024-03-13

One of the most evident and meaningful feedback about people’s emotions is through facial expressions. Facial expression recognition is helpful in social networks, marketing, and intelligent education systems. The use of Deep Learning based methods in facial expression identification is widespread, but challenges such as computational complexity and low recognition rate plague these methods. Scatter Wavelet is a type of Deep Learning that extracts features from Gabor filters in a structure similar to convolutional neural networks. This paper presents a new facial expression recognition method based on wavelet scattering that identifies six states: anger, disgust, fear, happiness, sadness, and surprise. The proposed method is simulated using the JAFFE and CK+ databases. The recognition rate of the proposed method is 99.7%, which indicates the superiority of the proposed method in recognizing facial expressions.

Prediction of short circuit current of wind turbines based on artificial neural network model

Ebrahim Aghajari — 2024-07-17

The growth of renewable energy on a global scale is making significant strides in power plants. This is due to the increasing concern about climate change, the rising demand for electricity, and the necessity to reduce reliance on fossil fuels. Ensuring the successful integration of new energy resources into the existing network is just as crucial as it requires the system to be reliable and adaptable. For instance, wind energy, which is one of the renewable sources, has an intermittent nature that necessitates the ability to synchronize its actions to achieve the desired system performance. The objective of this study is to utilize a new neural network system to calculate the short circuit current of power plants. Specifically, the focus is on identifying and categorizing the short circuit faults that occur between the stator coils of the squirrel cage induction generator used in wind power generation. To achieve this, a system was developed to simulate turbine data. Subsequently, four feature extraction techniques and machine learning algorithms were employed to enable early detection of short circuit faults. The numerical results obtained from the simulation demonstrated the high efficiency and accuracy of the proposed model.

A Hybrid Approach for Robust Object Detection: Integrating Template Matching and Faster R-CNN

Hewa Majeed Zangana — 2024-10-07

Object detection is a critical task in computer vision, with applications ranging from autonomous vehicles to medical imaging. Traditional methods like template matching offer precise localization but struggle with variations in object appearance, while deep learning approaches such as Faster R-CNN excel in handling diverse and complex datasets but often require extensive computational resources and large amounts of labeled data. This paper proposes a hybrid approach that integrates template matching with Faster R-CNN to leverage the strengths of both techniques. By combining the accuracy of template matching with the robustness and generalization capabilities of Faster R-CNN, our method achieves superior performance in challenging scenarios, including objects with occlusions, varying scales, and complex backgrounds. Extensive experiments demonstrate that the hybrid model not only enhances detection accuracy but also reduces computational load, making it a practical solution for real-world applications.

Hybrid Image Denoising Using Wavelet Transform and Deep Learning

Hewa Majeed Zangana — 2024-11-20

In this paper, we propose a hybrid image denoising method that combines wavelet transform and deep learning techniques to effectively remove noise from digital images. The wavelet transform is applied to each color channel of the noisy image, decomposing it into different frequency components. The approximation coefficients are then denoised using a convolutional neural network (CNN), specifically designed for this task. The denoised coefficients are subsequently reconstructed to form the final denoised image. Our experimental results demonstrate that this hybrid approach outperforms traditional denoising methods, achieving superior noise reduction while preserving image details. The proposed method is validated using synthetic noisy images, and the results are visually and quantitatively evaluated to confirm its effectiveness.

Fog Cloud Computing and IoT Integration for AI enabled Autonomous Systems in Robotics

Kiran Deep Singh — 2024-03-12

Fog Cloud Computing and the Internet of Things are transforming robotics by empowering AI-enabled autonomous systems. This study analyzes the benefits, drawbacks, and uses of this integration. AI-enabled autonomous robots can use edge computing and cloud resources for real-time data processing and decision-making, improving their performance and adaptability. Communication protocols, data management, security, and scalability are examined in the ecosystem. Case studies reveal how this confluence affects robotics applications. This research shows how FCC, IoT, and AI may improve robotic systems' efficiency, intelligence, and autonomy. The article covers AI-enabled autonomous systems in transportation, manufacturing, healthcare, agriculture, and smart cities. These technologies can improve productivity and safety in many fields, from self-driving automobiles to surgical robots. Integrating these technologies raises safety, ethical decision-making, data privacy, and security concerns. The report emphasizes transparent and ethical AI algorithms, unbiased decision-making, and regulatory frameworks to enable responsible integration and mitigate dangers. In the future, AI-enabled autonomous systems will be shaped by improved AI algorithms, multi-modal sensing, human-robot collaboration, and edge intelligence. It emphasizes the necessity of interdisciplinary collaboration and ethical considerations in responsible technology development. This study concludes with a detailed analysis of fog/cloud computing, IoT, and AI in robotics, revealing the immense promise and problems of AI-enabled autonomous systems. Responsible development and collaboration can help us negotiate this transformational frontier and create a safer, more efficient, and innovative society with AI-driven autonomous systems.

Interdisciplinary Approaches: Fog/Cloud Computing and IoT for AI and Robotics Integration

Prabh Deep Singh — 2024-01-08

Fog/Cloud Computing and the Internet of Things have created intriguing opportunities for AI and robotics integration. This study examines interdisciplinary approaches that combine FCC, IoT, AI, and Robotics to construct sophisticated autonomous systems. These integrated systems may efficiently and intelligently conduct complicated tasks by using edge devices and cloud resources. Communication protocols, data management, security, and interoperability are studied in this interdisciplinary environment. Real-world case studies demonstrate the practicality and benefits of this integration. This study shows how interdisciplinary approaches will change AI and robotics integration. In conclusion, the intersection of Fog/Cloud Computing, IoT, AI, and Robotics is influencing autonomous systems. Edge devices and the cloud enable robots to become intelligent, adaptable, and essential parts of many industries. This research encourages researchers, practitioners, and policymakers to collaborate on innovation and widespread adoption of disruptive technologies. Interdisciplinary techniques are essential to maximizing AI and robotics integration and launching a new era of intelligent automation

Computer vision recognition in the teaching classroom: A Review

Hui Jiang — 2024-01-08

Artificial intelligence introduces computer vision recognition into the teaching classroom, and computer vision recognition technology lays a solid foundation for the intelligent teaching classroom. Through the classroom camera video stream to the classroom student information data collection, voice, posture, facial, physiological signal data recognition analysis processing to extract and define the characteristics of student behaviour, automatic classification behaviour and then record and display student behaviour, thus effectively help teachers to grasp the students learning state and emotions, to promote the quality of teaching has far-reaching significance.

Methods and Strategies for 3D Content Creation Based on 3D Native Methods

Shun Fang — 2024-05-27

The present paper provides a comprehensive overview of three neural network models, namely Point·E, 3DGen, and Shap·E, with a focus on their overall processes, network structures, loss functions, as well as their strengths, weaknesses, and potential future research opportunities. Point·E, an efficient framework, generates 3D point clouds from complex text prompts, leveraging a text-to-image diffusion model followed by 3D point cloud creation. 3DGen, a novel architecture, integrates a Variational Autoencoder with a diffusion model to produce triplane features for conditional and unconditional 3D object generation. Shap·E, a conditional generative model, directly generates parameters of implicit functions, enabling the creation of textured meshes and neural radiance fields. While these models demonstrate significant advancements in 3D generation, areas for improvement include enhancing sample quality, optimizing computational efficiency, and handling more complex scenes. Future research could explore further integration of these models with other techniques and extend their capabilities to address these challenges.

Exploring the Capabilities of NeRF in Generating 3D Models

Shun Fang — 2024-04-22

This review paper presents a comprehensive analysis of three cutting-edge techniques in 3D content synthesis: EG3D, DreamFusion, and Magic3D. EG3D, leveraging geometry-aware representations and generative adversarial networks, enables the generation of high-quality 3D shapes. DreamFusion integrates text-to-image diffusion models with neural rendering, opening new horizons for creative expression. Magic3D, on the other hand, extends text-to-image synthesis principles to 3D content creation, synthesizing realistic and detailed models. We delve into the theoretical frameworks, neural network architectures, and loss functions of these techniques, analyzing their experimental results and discussing their strengths, weaknesses, and potential applications. This review serves as a valuable resource for researchers and practitioners, offering insights into the latest advancements and pointing towards future directions for exploration in 3D content synthesis.

A Survey of Data-Driven 2D Diffusion Models for Generating Images from Text

Shun Fang — 2024-04-22

This paper explores recent advances in generative modeling, focusing on DDPMs, HighLDM, and Imagen. DDPMs utilize denoising score matching and iterative refinement to reverse diffusion processes, enhancing likelihood estimation and lossless compression capabilities. HighLDM breaks new ground with high-res image synthesis by conditioning latent diffusion on efficient autoencoders, excelling in tasks through latent space denoising with cross-attention for adaptability to diverse conditions. Imagen combines transformer-based language models with HD diffusion for cutting-edge text-to-image generation. It uses pre-trained language encoders to generate highly realistic and semantically coherent images, surpassing competitors based on FID scores and human evaluations in DrawBench and similar benchmarks. The review critically examines each model's methods, contributions, performance, and limitations, providing a comprehensive comparison of their theoretical underpinnings and practical implications. The aim is to inform future generative modeling research across various applications.

A Comprehensive Survey of Text Encoders for Text-to-Image Diffusion Models

Shun Fang — 2024-07-18

In this comprehensive survey, we delve into the realm of text encoders for text-to-image diffusion models, focusing on the principles, challenges, and opportunities associated with these encoders. We explore the state-of-the-art models, including BERT, T5-XXL, and CLIP, that have revolutionized the way we approach language understanding and cross-modal interactions. These models, with their unique architectures and training techniques, enable remarkable capabilities in generating images from textual descriptions. However, they also face limitations and challenges, such as computational complexity and data scarcity. We discuss these issues and highlight potential opportunities for further research. By providing a comprehensive overview, this survey aims to contribute to the ongoing development of text-to-image diffusion models, enabling more accurate and efficient image generation from textual inputs.

Micro robot as the feature of robotic in healthcare approach from design to application: the State of art and challenges

Ata Jahangir Moshayedi — 2024-04-12

Micro robots, miniature robotic devices typically ranging from micrometers to a few millimeters in size, hold immense potential in various fields, particularly healthcare. Their diminutive stature enables access to intricate anatomical regions previously unreachable, facilitating targeted drug delivery, localized treatment, and precise monitoring. These robots offer numerous advantages, including enhanced maneuverability, reduced invasiveness, and minimized tissue damage. By navigating through complex biological environments, micro robots can deliver therapies with unprecedented precision, improving treatment efficacy and patient outcomes. Additionally, their small size allows for minimally invasive procedures, reducing recovery times and enhancing patient comfort. Overall, micro robots represent a groundbreaking technological advancement with the potential to revolutionize healthcare delivery and significantly benefit human well-being.
Their small size enables access to intricate anatomical regions for targeted drug delivery, localized treatment, and precise monitoring. Despite challenges like size constraints and navigation complexities, innovative solutions and interdisciplinary collaboration are driving their advancement in improving healthcare outcomes.

Systematic review on Artificial Intelligence in the editorial management of scientific journals

Carlos Rafael Araujo Inastrilla — 2024-06-04

INTRODUCTION: scientific journals play a crucial role in the dissemination and validation of scientific knowledge, and editorial management ranges from conceptualization to post-publication of content. Artificial intelligence (AI) has had a great impact on scientific communication and editorial management of scientific journals. AI can offer solutions and benefits for editorial management, but it also poses technical, economic, social and ethical challenges that should be considered and approached with caution and responsibility.

OBJECTIVE: to describe the benefits and limitations of the use of AI in the editorial management of scientific journals through a systematic literature review.

METHOD: a systematic literature review was conducted based on the PRISMA methodology. An information search was carried out in different bibliographic database systems, indexing systems and search engines, and inclusion and exclusion criteria were applied to the identified studies.

RESULTS: the information search allowed retrieving a total of 2750 sources, of which 10 articles that met the stated criteria were included. Benefits such as the facilitation of writing, translation, review and editing tasks were identified, as well as limitations related to ethical issues, bias, errors and plagiarism generated by AI.

CONCLUSIONS: While AI can streamline the production and analysis of information distribution, it also poses challenges in terms of reliability, ethics and authenticity of published content. It requires the critical involvement of human intelligence for proper exploitation.

Robots in Agriculture: Revolutionizing Farming Practices

Ata Jahangir Moshayedi — 2024-06-20

The integration of robotics in modern agriculture represents a revolutionary paradigm shift, enhancing efficiency and sustainability in food production. Agricultural robots, designed to automate various tasks, play a pivotal role in addressing the challenges faced by the industry. These robots are purpose-built for activities such as precision planting, weeding, and harvesting, streamlining processes that were traditionally labor-intensive. Their implementation leads to increased productivity, reduced operational costs, and minimized environmental impact through optimized resource utilization. This paper delves into the intricate landscape of robotic structures employed in agriculture, unraveling the diverse mechanisms and designs that underpin their functionality. It meticulously examines and elucidates the structural nuances of agricultural robots, shedding light on the engineering marvels that enable precision farming. From articulated arms to autonomous drones, the paper navigates through a spectrum of robot architectures, dissecting their roles in automating tasks critical to modern agriculture.

Empowering financial futures: Large language models in the modern financial landscape

Xinwei Cao — 2024-07-25

In this paper, we delve into the transformative influence of Large Language Models (LLMs) in the financial sector. Through meticulous exploration, we uncover the multifaceted applications of LLMs, ranging from elevating customer support and fortifying fraud detection to reshaping market analysis and prediction. LLMs, with their unparalleled ability to process extensive textual data, bring forth innovative solutions and insights. However, we also address critical challenges such as user trust and ethical considerations, emphasizing the need for responsible integration. Collaborative efforts between industry stakeholders and researchers are essential prerequisites for making a pivotal stride towards a future where LLMs redefine financial practices, with efficiency, accuracy, and ethical precision shaping the industry’s evolution.

Empirical Analysis of Widely Used Website Automated Testing Tools

Balqees Sani — 2024-10-10

In today's software development, achieving product quality while minimising cost and time is critical. Automated testing is crucial to attaining these goals by lowering inspection efforts and discovering faults more effectively. This paper compares widely used automated testing tools, such as Selenium, Appium, Java Unit (JUnit), Test Next Generation (TestNG), Jenkins, Cucumber, LoadRunner, Katalon Studio, Simple Object Access Protocol User Interface (SoapUI), and TestComplete, based on functionality, ease of use, platform compatibility, and integration capabilities. Our findings show that no single tool is inherently superior, with each excelling in certain areas such as online, mobile, Application Programming Interface (API), or performance testing. While Selenium and Appium are the dominant online and mobile testing frameworks, TestComplete and Katalon Studio offer complete, user-friendly cross-platform testing solutions. Despite the benefits of automation, obstacles such as tool maintenance, scalability, and cost issues remain. The report finishes with advice for picking the best tool for the project and offers potential approaches for enhancing testing frameworks, such as AI-driven optimisation, cloud-based testing, and greater Continuous Integration/ Continuous Deployment (CI/CD) integration. This study offers useful information for developers and testers looking to optimise their testing methods and increase software quality.

A Comprehensive Review of Electromyography in Rehabilitation: Detecting Interrupted Wrist and Hand Movements with a Robotic Arm Approach

Kimia Savoji — 2024-10-08

Electromyography (EMG) is a diagnostic technique that measures the electrical activity generated by skeletal muscles. Utilizing electrodes placed either on the skin (surface EMG) or inserted directly into the muscle(intramuscular EMG), it detects electrical signals produced during muscle contractions. EMG is widely employed in clinical and research settings to assess muscle function, diagnose neuromuscular disorders,and guide rehabilitation therapy. Over the years, EMG has evolved from a basic measurement tool into an essential technology within clinical and research environments, propelled by advances in recording techniques and digital innovations. The integration of wearable technology and artificial intelligence (AI) has significantly expanded its applications, particularly in rehabilitation and sports science. By capturing muscle electrical activity through surface or intramuscular electrodes, EMG benefits from enhanced signal processing that improves accuracy and data analysis. Despite challenges such as signal interference and the complexities of movement patterns—especially in wrist and hand rehabilitation—EMG combined with robotic systems offers real-time feedback for precise and personalized therapy. However, obstacles like cost, complexity, and variability among patients still remain. Future advancements aim to make EMG more accessible and to integrate AI for tailored rehabilitation strategies, alongside improvements in sensors and wireless communication to enhance reliability and performance. This review explores various facets of EMG,from its fundamental principles to its application in detecting disrupted wrist and hand movements through robotic approaches. It provides a comprehensive analysis of EMG’s historical and technological evolution, recent innovations like AI and wearable devices, and its extensive applications in rehabilitation and sports science. Detailed case studies illustrate its effectiveness in areas such as stroke recovery and spinal cord injury rehabilitation. Additionally, the review addresses challenges like technical limitations and patient variability while emphasizing the integration of EMG with robotic systems for personalized therapy. It also discusses the significance of real-time feedback, future enhancements in AI and sensor technology, and the pressing need for more affordable, user-friendly solutions to improve therapeutic outcomes.

An Overview of OpenAI's Sora and Its Potential for Physics Engine Free Games and Virtual Reality

Zuyan Chen — 2024-03-06

Sora, OpenAI's latest text-to-video model, is particularly skilled at understanding the physical world, and all of the content it generates mostly consistent with the laws of physics. This indicates that Sora already has the beginnings of a world model and has the potential to become an excellent physics engine in the near future. This paper analyses and explains in detail the potential applications of Sora in physics engines and virtual reality. In addition, its advantages and disadvantages over traditional physics engines are compared based on its unique behavioural characteristics. Finally, it looks forward to the application of Sora in other fields.

RETRACTED ARTICLE Mapping Generative Artificial Intelligence (GAI's) Exciting Future: From Gemini to Q* and Beyond

Zarif Bin Akhtar — 2024-08-15

RETRACTED: This article has been retracted at the request of our research integrity team. The retraction notice can be found here https://doi.org/10.4108/airo.7168

Retraction Notice: Mapping Generative Artificial Intelligence (GAI's) Exciting Future: From Gemini to Q* and Beyond

Zarif Bin Ahktar — 2024-09-02

The article Mapping Generative Artificial Intelligence (GAI's Exciting Future: From Gemini to Q* and Beyond has been retracted at the request of EAI's Research Integrity Committee. The paper is being removed on the grounds plagiarism and misrepresentation of academic affiliation.