AI robot application的問題，透過圖書和論文來找解法和答案更準確安心。 我們找到下列免費下載的地點或者是各式教學

Intelligent Sustainable Systems: Selected Papers of WorldS4 2021, Volume 2

為了解決AI robot application的問題，作者 這樣論述：

Wall-distance Measurement for Indoor Mobile Robots.- Automating Cognitive Modelling Considering Non-Formalisable Semantics.- Using a Humanoid Robot to Assist Post-Stroke Patients with Standardised Neurorehabilitation Therapy.- Human Resource Information System in Healthcare Organizations.- Perfor

mance Prediction of Scalable Multi-Agent Systems using Parallel Theatre.- Dynamics of Epidemic Computer Subnetwork Models for Scan-based Worm Propagation: An Internet Protocol Addressing Configuration Perspective.- Security Analysis of Integrated Clinical Environment Using Attack Graph.- Smart Unive

rsity: Key factors for a cloud computing adoption model.- Agile governance supported by the frugal smart city.- Effect of Normal, Calcium Chloride Integral and Polyethene Sheet Membrane curing on the strength characteristics of Glass Fiber and Normal concrete.- Problems with health information syste

ms in Ecuador, and the need to educate university students in health informatics in times of pandemic.- Utilizing Technological Pedagogical Content (TPC) for Designing Public Service Websites.- А Multidimensional Rendering of Error Types in Sensor Data.- Optimization of the Overlap Shortest-Path Rou

ting for TSCH Networks.- Application of Emerging Technologies in Aviation MRO Sector to Optimize Cost, Utilization: The Indian Case.- User Evaluation of a Virtual Reality Application for safety training in Railway Level Crossing.- GreenMile - Gamification-Supported Mobile and Multimodal Route Planni

ng for a Sustainable Choice of Transport.- A Detailed Study for Bankruptcy Prediction by Machine Learning Technique.- Cyberbullying in Online/E-Learning Platforms Based on Social Networks.- Machine Learning and Remote Sensing Technique for Urbanization Change Detection in Tangail District.- Enabling

a Question-Answering System for COVID Using a Hybrid Approach based on Wikipedia and Q/A Pairs.- A Study of Purchase Behavior of Ornamental Gold Consumption.- Circularly Polarized Micro-strip Patch Antenna for 5G Applications.- An approach towards protecting Tribal lands through ICT Interventions.-

Urban Sprawl Assessment Using Remote Sensing and GIS Techniques: A Case Study of Ernakulam District.- Exploring the means and benefits of including Blockchain smart contracts to a smart manufacturing environment: Water bottling plant case study.- Extreme gradient boosting for predicting stock price

direction in context of Indian equity markets.- Performance of Grid-Connected Photovoltaic and its Impact: A Review.- A Novel Approach of Deduplication on Indian Demographic Variation for Large Structured Data.- Dual-message Compression with Variable Null Symbol Incorporation on Constrained Optimiz

ation based Multipath and Multihop Routing in WSN.- Spatio-temporal variances of Covid-19 active cases and genomic sequence data in India.- Significance of Thermoelectric Cooler approach of Atmospheric Water Generator for solving fresh water scarcity.- Fabrication of Energy Potent Data Center using

Energy Efficiency Metrics.- Performance Anomaly and Change Point Detection for Large-Scale System Management.- Artificial Intelligence Driven Monitoring, Prediction and Recommendation System (AIM-PRISM).- Financial Forecasting of Stock Market usingSentiment Analysis and Data Analytics.- A Survey on

Learning-Based Gait Recognition for Human Authentication in Smart Cities.- Micro-Arterial Flow Simulation for Fluid Dynamics: A Review.- Hip-Hop Culture incites Criminal Behavior: A Deep Learning Study.- Complex Contourlet Transform Domain Based Image Compression.- HWYL: An Edutainment Based Mobile

Phone Game Designed to Raise Awareness on Environmental Management.- IoT & AI based Advance LPG System (ALS).- ICT Enabled Automatic Vehicle Theft Detection System at Toll Plaza.- RBJ20 Cryptography Algorithm for Securing Big Data Communication using Wireless Networks.- Decision Tree for Uncerta

in Numerical Data Using Bagging and Boosting.- A Comparative Study on Various Sha

AI robot application進入發燒排行的影片

ThermalPose：基於熱影像深度學習人體姿態辨識技術之設計與實現

為了解決AI robot application的問題，作者蔡承翰 這樣論述：

現行的人體姿態辨識方法相當多樣，其中，多數使用RGB相機拍攝高解析度的圖像來取得人體特徵後進行骨幹評估，然而彩色圖像在人體姿態辨識容易受到燈光、環境所影響，導致無法準確的獲得關節點骨架，此外，彩色圖像的相機無法運用於具有隱私之場域，如：醫院、照護中心的廁所或浴室等。目前有許多研究為了達到去特徵化的人體姿態辨識，使用射頻訊號收發器、毫米波雷達等感測器進行人體姿態辨識，然而，這些方法雜訊過高與解析度不足，導致關節點骨架準確度低。本論文提出一種基於熱影像深度學習人體姿態辨識技術，稱為「ThermalPose」，可準確的辨識與追蹤人體關節與骨幹。ThermalPose包含兩個部分:骨幹辨識技術與動作

辨識演算法，骨幹辨識技術以熱像感測器、AI邊緣運算裝置與自蒐集熱影像資料集進行人體姿態辨識；而動作辨識演算法的目標是辨識日常生活中的動作，如:走路、跑步、坐地與彎腰。由實驗結果可證明，ThermalPose可在無RGB相機的情況下有效的使用熱影像辨識人體姿勢，因此可用於低光源與具有個人隱私環境的無人化應用。

Cognitive Systems and Signal Processing: 5th International Conference, Iccsip 2020, Zhuhai, China, December 25-27, 2020, Revised

為了解決AI robot application的問題，作者 這樣論述：

Award.- Quantized Separable Residual Network for Facial Expression Recognition on FPGA.- Hole-peg Assembly Strategy Based On Deep Reinforcement Learning.- EEG-based Emotion Recognition Using Convolutional Neural Network with Functional Connections.- Fast barcode detection method based on ThinYOLOv4.

- The Realtime Indoor Localization for Unmanned Aerial Vehicle.- Algorithm.- L1-norm and Trace Lasso based Locality Correlation Projection.- Episodic Training for Domain Generalization Using Latent Domains.- A Novel Attitude Estimation Algorithm Based on EKF-LSTM Fusion Model.- METAHACI: Meta-learni

ng for Human Activity Classification from IMU Data.- Fusing Knowledge and Experience with Graph Convolutional Network for Cross-task Learning in Visual Cognitive Development.- Factored Trace Lasso based Linear Regression Methods: Optimizations and Applications.- Path Planning and Simulation Based on

Cumulative Error Estimation.- MIMF: Mutual Information-driven Multimodal Fusion.- Application.- Spatial Information Extraction of Panax Notoginseng Fields Using Multi-algorithm and Multi-sample Strategy-based Remote Sensing Techniques.- Application of Convolution BLS in AI Face-changing Problem.- C

ognitive Calculation Studied for Smart System to Lead Water Resources Management.- A Robotic Arm Aided Writing Learning Companion for Children.- Design and Implement of Omnidirectional Mobile Robot Platform with Remote Visual Control.- AG-DPSO: Landing Position Planning Method for Multi-node Deep Sp

ace Explorer.- A New Paralleled Semi-supervised Deep Learning for Remaining Useful Life Prediction.- Balancing Task Allocation in Multi-robot Systems using AdpK-means Clustering Algorithm.- Reinforcement Learning for Extreme Multi-label Text Classification.- Manipulation.- Multimodal Object Analysis

with Auditory and Tactile Sensing using Recurrent Neural Networks.- A Novel Pose Estimation Method of Object in Robotic Manipulation using Vision-based Tactile Sensor.- Design and Implementation of pneumatic soft gripper with suction and grasp composite structure.- Movement Primitive Libraries Lear

ning for Industrial Manipulation Tasks.- Guided Deep Reinforcement Learning for Path Planning of Robotic Manipulators.- Large-scale multi-agent reinforcement learning based on weighted mean field.- Selective Transition Collection in Experience Replay.- Design and Grasping Experiments of Soft Humanoi

d Hand with Variable Stiffness.- Bioinformatics.- Prediction the age of human brains from gene expression.- Deep LSTM Transfer Learning for Personalized ECG Anomaly Detection on Wearable Devices.- Reflection on AI: The cognitive difference between libraries and scientists.- A Review of Research on B

rain-Computer Interface Based on Imagined Speech.- A Survey of Multimodal Human-Machine Interface.- A Logistic Regression Based Framework for Spatio-temporal Feature Representation and Classification of Single-trial EEG.- Biometric Traits Share Patterns.- Cervical Cell Detection Benchmark with Effect

ive Feature Representation.- Vision A.- Image Fusion for Improving Thermal Human Face Image Recognition.- Automatic Leaf Recognition Based on Attention DenseNet.- Semantic Segmentation for Evaluation of Defects on Smartphone Screens.- Image Clipping Strategy of Object Detection for Super Resolution

Image in Low Resource Environment.- Image Quality Assessment with Local Contrast Estimator.- Application of Broad learning system for image classification based on deep features.- Emotion Recognition Based on Graph Neural Networks.- A Light-Weight Stereo Matching Network with Color Guidance Refineme

nt.- Vision B.- Overview of Monocular Depth Estimation Based on Deep Learning.- Non-contact physiological parameters detection based on MTCNN and EVM.- Texture Classification of a Miniature Whisker Sensor with Varied Contact Pose.- Semantic-based Road Segmentation for High-Definition Map Constructio

n.- Transformer Region Proposal for Object

微型電網分層控制策略研究

為了解決AI robot application的問題，作者張泉泉 這樣論述：

微型電網(Microgrid)作為一種高效利用可再生能源分散式發電(Distributed Generation)的方法，可被用於解決偏遠地區的發電問題或為關鍵負荷提供不間斷供電。為了保證微型電網的可靠性和經濟運行，首要任務是維持系統電壓/頻率穩定和實現分散式發電單元之間功率的精確分配。微型電網通常運行於中低壓電力系統中，其線路阻抗主要呈現電阻電感性，傳統的P-f/Q-U下垂控制(Droop Control)性能不佳，雖然可通過採用虛擬複阻抗(Virtual Complex Impedance)的方法，使線路阻抗中的電阻分量被虛擬負電阻抵消。但由於存在線路阻抗參數漂移和估計誤差等問題，若虛擬

負電阻設計不當會導致系統不穩定。本文根據中低壓微型電網的線路參數特點，採用P-U/Q-f下垂控制，並且在控制迴路中引入由虛擬負電感和虛擬電阻組成的虛擬複阻抗，其中虛擬負電感用於減小系統阻抗中電感分量引起的功率耦合(Power Coupling)，虛擬電阻用於增強系統中的電阻分量，並且調整阻抗匹配度以提高功率分配精度。然而此作法功率分配仍然會受到系統線路阻抗參數的影響。此外，下垂控制結合虛擬阻抗方法易引起電壓偏差問題。因此本文研究了一種新型的基於虛擬複阻抗的穩壓均流控制方法，在不受線路阻抗參數變化影響的情況下實現精確的功率分配，並且提高電壓品質。本研究同時建立基於所提出方法的微型電網系統小信號模

型(Small-Signal Model)，用於分析系統的穩定性和動態性能，同時為控制器參數的設計提供理論依據。分析結果表明，所提出方法對線路阻抗參數漂移和估計誤差具有強健性，並且使系統具有較大的穩定裕度和較快的動態響應速度。再者，本文針對微型電網併聯逆變器的有功功率分配和電壓偏差問題探討，基於全域滑動模式控制(Total Sliding-Mode Control)技術重新設計功率-電壓下垂控制器和內迴路電壓調節器。首先，針對功率-電壓下垂控制回路，定義有功功率與公共耦合點(Point-of-Common-Coupling)電壓幅值之間的下垂控制關係誤差。然後通過採用全域滑動模式控制以獲得新的

下垂控制關係，從而同時實現有功功率分配和電壓幅值恢復。由於全域滑動模式控制方案可為系統提供快速的動態性能和強健性，高精度的暫態有功功率分配也可在不受線路阻抗影響的情況下被實現。更進一步，本文針對微型電網提出基於自我調整模糊類神經網路(Adaptive Fuzzy Neural Network)的分散式二級控制(Distributed Secondary Control)方案，以實現電壓/頻率恢復和最優功率分配。首先，建立微型電網動態系統模型，該模型由逆變器介面分散式電源模型和微型電網電力網絡模型組成，其中分散式電源模型可通過具有最優有功功率分配方案的初級控制器的動態模型來表示。微型電網電力網絡

模型由潮流動態模型和負荷模型組成。然後定義基於一致性演算法的誤差函數，並提出基於模型的全域滑動模式控制技術來處理同步和跟蹤問題。為達到無須詳細動態控制設計，本文設計自我調整模糊類神經網路方案來模擬全域滑動模式控制律，以繼承其快速動態響應性能和強健性。同時，所提出的自我調整模糊類神經網路控制方法可以解決全域滑動模式控制對微型電網動態模型精確資訊的依賴。藉由投影演算法(Project Algorithm)和李雅普諾夫穩定性(Lyapunov Stability)定理，推導模糊類神經網路的參數自我調整調節律，以保證基於自我調整模糊類神經網路的分散式二級控制系統的穩定性。本文所提出方法的有效性和優越性

將通過數值模擬和實驗進行驗證。