Industry 4.0 III: Advanced Computational Technologies

Introduce machine learning fundamentals, its relation to artificial intelligence, and the three learning types—supervised, unsupervised, and reinforcement—showing data-driven models and predictions in industry.

Explore supervised learning and linear regression with labeled data, modeling relationships to predict continuous outputs and classify data, using a best-fit line and gradient descent to minimize squared error.

Explore how classification, a supervised learning method, assigns discrete labels from labeled data by mapping input features to predefined classes, with applications such as spam detection and medical diagnosis.

Learn how decision trees train a supervised learning model that splits data by features to build a tree for classification or regression, using entropy, gini impurity, and variance reduction.

Explore unsupervised learning with clustering, focusing on k-means and DBscan, to form data clusters, handle outliers as noise, and enable customer segmentation, anomaly detection, and pattern recognition.

demonstrates predictive maintenance for a pump by building a logistic regression model on synthetic sensor data to predict failures using temperature, vibration, and pressure readings.

Explore model execution and result analysis for a logistic regression classifier, including predictions, confusion matrix visualization, feature importance, and live predictions with scaled data.

Explore machine learning models for fault detection in rotating and moving parts using smart sensor data, accelerometers, and vibration patterns to enable predictive maintenance and anomaly detection in industrial settings.

Explore real-world industrial artificial intelligence applications in predictive maintenance, including gear wear prediction and bearing failure analysis, using vibration, temperature, acoustic data, and sensor analytics.

Explore CAD generative design, an AI-driven process that generates multiple geometry options under predefined constraints for additive manufacturing, with engineers validating designs via simulations and testing.

Explore practical generative design for mechanical components, defining seed designs, preserved geometries, obstacle regions, and mass optimization within additive manufacturing constraints using Fusion 360.

Learn how convolution operations drive industrial computer vision by applying filters over images to produce feature maps for edge detection, texture recognition, and robust object segmentation.

Explore edge detection in image processing to identify boundaries and simplify analysis by highlighting regions of rapid pixel change with Sobel and Canny, enabling robust computer vision and object detection.

Explore feature extraction using the Fourier transform to convert images from the spatial domain to the frequency domain, revealing edges, textures, and patterns for image compression and pattern recognition.

Explore object detection and recognition with YOLO and Faster R-CNN, comparing real-time efficiency and accuracy for applications such as autonomous vehicles, security, and industrial automation.

Differentiate ai, ml, and deep learning by clarifying ai as the broad field and ml as learning from data; deep learning uses neural networks for complex tasks.

Explore how graphs encode information with nodes and edges to model networks and enable efficient querying and pattern recognition, while logic-based systems use formal logic for automatic reasoning and inference.

Reinforcement learning trains an agent to maximize rewards by interacting with an environment, improving automation tasks like robotics, autonomous vehicles, and industrial processes through iterative feedback and learning.

Explore how cobots use reinforcement learning to autonomously learn optimal assembly paths by interacting with their environment, improving efficiency, speed, and adaptability in dynamic manufacturing.

Explore how AI and machine learning apply in industry 4.0 and how to integrate software concepts with physical systems to monitor, predict, and optimize processes.