VIP20

本课程属于：006-Data Science and Engineering Analytics English Technical Courses · 共 20 门相关课程，组合学习更完整。

查看所属VIP全部课程 →

← 返回所属目录咨询 / 购买

Industrial Sensor Data Fusion and Analytics

30 Chapters

View Course Samples Online

Industrial sensor data fusion landscape

definitions, motivation, system architectures, and application domains.

Industriallandscape

Sensor characteristics and measurement principles

accuracy, precision, resolution, drift, noise, and calibration in industrial settings.

characteristicsmeasurement

Data acquisition and preprocessing pipelines

sampling, filtering, outlier removal, missing data handling, and synchronization.

acquisitionpreprocessing

Statistical foundations for sensor fusion

probability, Bayes theorem, covariance, correlation, and uncertainty propagation.

Statisticalfoundations

Kalman filtering for linear dynamic systems

prediction, update, tuning, and practical implementation for sensor fusion.

filteringprediction

Extended and unscented Kalman filters

nonlinear state estimation for industrial process and condition monitoring.

Extendedunscented

Particle filters and sequential Monte Carlo methods

handling non-Gaussian noise and multimodal distributions.

Particlesequential

Bayesian networks and probabilistic graphical models for multi-sensor diagnosis and decision support.

Bayesian networks and probabilistic graphical models for multi-sensor diagnosis and decision support.

Bayesiannetworks

Dempster-Shafer evidence theory

combining conflicting sensor evidence in fault detection and classification.

Dempster-Shaferevidence

Fuzzy logic and possibility theory

fusion under imprecision and linguistic uncertainty in industrial control.

possibilityimprecision

Multi-sensor data association and track-to-track fusion

nearest neighbor, JPDA, and distributed fusion architectures.

Multi-sensorassociation

Sensor registration and spatial alignment

coordinate transforms, time alignment, and extrinsic calibration.

registrationalignment

Feature-level and decision-level fusion

dimensionality reduction, feature selection, and classifier combination.

Feature-leveldecision-level

Deep learning for sensor fusion

CNNs, RNNs, autoencoders, and attention mechanisms for industrial time-series.

Graph neural networks and transformer architectures for multi-sensor and multi-modal industrial data.

Graph neural networks and transformer architectures for multi-sensor and multi-modal industrial data.

networkstransformer

Fusion of heterogeneous sensors

vibration, temperature, pressure, flow, acoustic, and vision data in practice.

heterogeneousvibration

Industrial IoT and edge computing for sensor fusion

protocols, gateways, latency, and bandwidth constraints.

Real-time stream processing and online learning for continuous industrial monitoring and anomaly detection.

Real-time stream processing and online learning for continuous industrial monitoring and anomaly detection.

Real-timeprocessing

Anomaly detection and fault diagnosis using fused sensor data

unsupervised, supervised, and semi-supervised methods.

detectiondiagnosis

Predictive maintenance and remaining useful life estimation

fusion-driven degradation modeling and prognostics.

Predictivemaintenance

Digital twins and cyber-physical systems

integrating fused sensor data for simulation, optimization, and control.

cyber-physicalintegrating

Data quality, trust, and fusion under adversarial conditions

sensor faults, cyberattacks, and robust estimation.

adversarialconditions

Scalable data management and analytics platforms

time-series databases, data lakes, and MLOps for industrial fusion.

Scalablemanagement

multi-sensor fusion for rotating machinery condition monitoring and fault classification.

multi-sensorrotating

fusion of process sensors for chemical or petrochemical plant anomaly detection and control.

chemicalpetrochemical

autonomous mobile robots and AGVs in smart factories using LiDAR, camera, IMU, and wheel odometry.

smart grid and energy systems: fusing PMU, SCADA, and smart meter data for state estimation.

Evaluation, validation, and benchmarking of sensor fusion systems

metrics, testbeds, and industrial standards.

Evaluationvalidation

Deployment challenges and best practices

edge-cloud orchestration, model compression, and lifecycle management.

Deploymentchallenges

Emerging trends and research frontiers

quantum sensing, 6G, explainable AI, and next-generation industrial fusion.

Emergingresearch