This paper describes the application of SLAM in complex indoor environment. Through the joint work of sensor, LiDAR and visual camera, the real-time self-positioning and map construction are realized at the same time. However, in the practical application, there are also many challenges, the layout of the environment, the brightness of the environment, the presence of noise, the presence of dynamic objects and other environmental factors will lead to different degrees of data errors in SLAM technology. The small errors generated by a long time of work will gradually accumulate into potentially dangerous huge errors, and this is one of the biggest challenges for the current SLAM technology. Not only has that, the high configuration conditions required by SLAM technology also become one of the reasons why people do not choose it. However, the advantages of SLAM technology far outweigh the disadvantages. And almost all of these problems have corresponding solutions, SLAM technology is still very reliable, very widely used, the future of a great prospect of a technology.

Power quality classification is essential for identifying and categorizing various power quality issues within electrical systems. With the increasing integration of electronic devices into power grids, including converters, rectifiers, and inverters, the significance of real-time monitoring and fault prediction has grown. This paper explores the application of image classification techniques, specifically Convolutional Neural Networks (CNN) and Multilayer Perceptron (MLP), in addressing power quality issues. Using a dataset consisting of signals representing different power quality conditions, models were trained and evaluated based on accuracy and loss metrics. The results indicate that while CNN achieved the highest accuracy and lowest loss, MLP models demonstrated efficiency in terms of computational resources. Additionally, the study discusses the potential of image classification technology in enhancing power quality monitoring and management, with implications for improving energy utilization efficiency and supply quality. The findings highlight the importance of leveraging deep learning techniques for addressing complex power quality challenges and paving the way for the intelligent and automated development of power systems.

The occurrence of extreme weather events affects the natural economic and social security, and has a serious impact on some industries, so the prediction and evaluation of disaster risk level is particularly important. This paper collected data on nine indicators from 25 regions and divided them into four areas: extreme weather, natural, social, and economic. the ARIMA prediction model is used to predict the probability and intensity of extreme weather in the future. The combined weights of each index were calculated by Composite Weighting Method, including principal component analysis, entropy weight method and coefficient of variation method, and the Risk Assessment Index (RAI) formula was constructed, and then the WNSE prediction and evaluation model was constructed. It is divided into four risk levels by fuzzy cluster analysis. We use WNSE prediction and evaluation model to calculate New York City and Switzerland’s RAI and coverage levels. Using this prediction, we can assess the severity of the consequences of natural disasters and prevent or remedy them. The assessment of regional disaster risk levels can give relevant reference to the government, and the government can make reasonable resource allocation.

Understanding the intricate dynamics of the Great Lakes water levels, shaped by environmental factors like rainfall and evaporation, is crucial for devising sustainable water management strategies. This study develops a model to manage the Great Lakes' water levels by analyzing the interconnectedness of the lake network. Utilizing principal component analysis, key variables like rainfall and evaporation were evaluated for their impact on water levels. System dynamics models integrated with flow conservation laws allowed for a detailed hydrodynamic assessment. The study introduces a "narrow seam method" for dry river scenarios, enhancing model accuracy. Annual water level trends from 2000 are depicted, offering insights into future water management practices in response to environmental changes.

This paper uses particle swarm and neural network algorithms to construct an efficient submarine search and rescue (SAR) solution based on an unmanned underwater detector cluster, aiming to provide stronger safety guarantees for tourists who experience the submarine project. Specifically, this article constructs a position prediction model and a search optimization model respectively. The former will accurately simulate the submarine's motion trajectory and various parameters through a combination of interpolation fitting and neural network learning. The latter will use the above data to Particle swarm algorithm optimizes the group behavior of unmanned detectors to derive the optimal search and rescue plan. Finally, upon analyzing the success rate of this strategy, it’s surprised to find that its increasing trends complies with Logistic growth law. This indicates that the strategy has promising advantages in terms of time efficiency and accuracy. The research conducted in this article provides an application direction for the particle swarm algorithm and a feasible model for the construction of deep-sea search and rescue strategies, which has a great use value.

Brain tumors represent a severe threat to human health due to their potential to disrupt vital functions controlled by the brain, such as movement, sensation, and cognition. This study aims to classify a comprehensive dataset of 7,023 magnetic resonance imaging (MRI) images from Figshare, SARTAJ, and BR35H into four distinct classes (Glioma, Meningioma, Pituitary, and No-Tumor). After evaluating the state-of-the-art models including VGG16, DenseNet201, ResNet50, and V3 Inception, the study proposes a novel Convolutional Neural Network (CNN) model with multiple inception layers. As a result, the model achieved an accuracy of 98.71% and high AUC scores for all four categories. These results suggest the potential of this model in enhancing diagnostic accuracy and reducing radiologist workload in clinical applications. Further improvements could focus on advanced feature extraction and integration of multi-parametric MRI data. The model's robustness and effectiveness underscore its potential utility in real-world clinical settings for accurate tumor classification.

This paper examines the relationship between "Momentum" and victory in the imbledon 2023 men's tournament, particularly between the semi-finals and the final. The definition and ormula of the Pearson's correlation coefficient are presented, and the Pearson's correlation oefficient is analysed based on the match data between Carlos Alcaraz and Novak Djokovic. The results show that there is a positive correlation between "Momentum" and winning. Then the logistic regression model was used to analyse the factors affecting the variation of "kinetic energy", such as onsecutive points, decisive points, key points, fatigue level and double faults. By training and testing the data set, a logistic regression model was developed for predicting the changes in "kinetic energy", and some suggestions based on the differences in "Momentum" were made to help players and coaches better cope with the changes in "Momentum" in the game. This paper also proposes some suggestions based on "Momentum" differences to help players and coaches better cope with "Momentum" changes in matches.

Movie, also known as moving or moving pictures, or "reflections," is a form of visual art that communicates ideas, stories, cognitions, emotions, values, or various atmospheric simulated experiences through the use of moving images. As the mainstay of cultural industry, film has brought huge economic and social benefits, and the box office is the most important index to measure the economic benefits of film. In this study, the prediction results of the random forest analysis model are acceptable, it is well in line with the structure of the data, which is conducive to the analysis of the data, and the data is displayed with charts and graphs at the same time, so as to show the readers the indicators and data used in this topic more clearly and clearly, which increases the scientific nature of this paper. This study provides more scientific and accurate strategies for decision-makers and creators, and contributes to the high-quality dissemination of short videos.

This paper introduces the Adaptive Thompson Sampling (AdaTS) algorithm, specifically designed for the challenges posed by opportunistic bandit problems. Unlike conventional multi-armed bandit scenarios where the optimality of actions remains relatively static, opportunistic bandit environments are significantly influenced by fluctuating external conditions. These variations necessitate a highly adaptive strategy for decision-making. The AdaTS algorithm meets this requirement through a novel integration of real-time system load assessments, which dynamically adjust the balance between exploration and exploitation. This method not only ensures more effective adaptation to changing conditions but also considerably enhances performance metrics. By incorporating system load into the decision-making process, AdaTS effectively reduces regret and maximizes reward, particularly under scenarios of variable load intensities. This is particularly evident in environments characterized by binary-valued loads and stochastic reward distributions. The results highlight AdaTS's robustness and efficiency, making it a promising approach for complex environments where adaptability is crucial. This research marks a significant advancement in the field of adaptive algorithms for opportunistic bandit problems.

Automatic Speech Recognition (ASR) is a technology that leverages artificial intelligence to convert spoken language into written text. It utilizes machine learning algorithms, specifically deep learning models, to analyze audio signals and extract linguistic features. This technology has revolutionized the way that people interact with voice-enabled devices, enabling efficient and accurate transcription of human speech in various applications, including voice assistants, captioning, and transcription services. Among previous works for ASR, Long Short-Term Memory (LSTM) networks and Transformer-based methods are typical solutions towards effective ASR. In this paper, the author focuses on an in-depth exploration of the progression and comparative analysis of deep learning innovations within the ASR domain. This work starts with a foundational historical perspective, mapping the evolution from pioneering ASR systems to the current benchmarks: LSTM networks and Transformer-based models. The study meticulously evaluates these technologies, dissecting their strengths, weaknesses, and the potential they hold for future advancements in ASR.