Unit Name | Machine Learning For Industrial Automation |
Unit Code | ME605 |
Unit Duration | 12 weeks |
Award |
Graduate Diploma of Engineering (Industrial Automation)
|
Year Level | 2nd |
Unit Creator / Reviewer | Hadi Harb |
Core/Sub-Discipline: | Core |
Pre/Co-requisites | None |
Credit Points |
|
Mode of Delivery | On-Campus or Online |
Unit Workload | 10 hours per week: Lecture - 1 hour Tutorial Lecture - 1 hour Practical / Lab - 1 hour (where applicable) Personal Study recommended - 7 hours (guided and unguided) |
Unit Description and General Aims
This unit addresses machine learning and its application to industrial automation.
In this unit the student will be introduced to supervised learning, clustering, regression and time-series analysis. Data pre-processing and system evaluation will be explored.
A series of sub-topics will address the characteristics of commonly used algorithms such as Decision Trees, K-Nearest Neighbours, Neural Networks, linear regression, and K-Means Clustering.
Different applications of machine learning to industrial automation will be explored. This will include condition monitoring, system identification, and image processing for autonomous vehicles. Software tools that can be used to implement machine learning algorithms will be presented. The student will be able to use such tools to apply machine learning to a particular industrial automation problem.
Learning Outcomes
On successful completion of this subject/unit, students are expected to be able to:
- Judge the applicability of machine learning to an industrial automation problem.
Bloom’s Level 5. - Plan and execute data pre-processing to a machine learning problem.
Bloom’s Level 6 - Evaluate a machine learning algorithm.
Bloom’s Level 5. - Design and implement a machine learning system to solve an industrial automation problem.
Bloom’s Level 6
Student assessment
Assessment Type (e.g. Assignment - 2000 word essay (specify topic) Examination (specify length and format)) |
When assessed (e.g. After Topic 5) |
Weighting (% of total unit marks) | Learning Outcomes Assessed |
Assignment 1
|
After Topic 6 | 15% | 1, 2 |
Assignment 2
|
After Topic 8 | 25% | 1, 2, 4 |
Assignment 3 Type: Presentation, discussion, group work, exercises, self-assessment/reflection, case study analysis, application.
|
After Topic 10 | 15% | 1, 2 |
Assignment 4 – Final Project
Embedded practical component: |
Final Week | 40% | 1, 2, 3, 4 |
Attendance / Tutorial Participation Example: Presentation, discussion, group work, exercises, self-assessment/reflection, case study analysis, application. |
Continuous | 5% |
Prescribed and Recommended Readings
Required textbook
- [Available on Knovel] Bonnin, Rodolfo. (2017). Machine Learning for Developers. ISBN 978-1-78646-987-8, Packt Publishing.
Recommended textbooks
- [Available on Knovel] Lucci, Stephen Kopec, Danny. (2016). Artificial Intelligence in the 21st Century. 2nd Edition. ISBN 978-1-942270-00-3, Mercury Learning and Information.
- [Available on Knovel] Davies, E. R. (2018). Computer Vision - Principles, Algorithms, Applications, Learning. 5th Edition. ISBN 978-1-942270-00-3, Elsevier.
Reference MaterialsNumber of peer-reviewed journals and websites (advised during lectures).
Some examples are listed below.
- Engineering Applications of Artificial Intelligence, Elsevier
- Machine Learning and Knowledge Extraction, MDPI
- IEEE Transactions on Evolutionary Computation, IEEE
- IEEE Transactions on Pattern Analysis and Machine Intelligence, IEEE
- Internet of Things: Engineering Cyber Physical Human Systems, Elsevier
- Other material advised during the lectures.
Unit Content
One topic is delivered per contact week, with the exception of part-time 24-week units, where one topic is delivered every two weeks.
Topics 1
Introduction to Machine Learning
- Classification
- Regression
- Time-series analysis
- Supervised learning
- Clustering
- Knowledge representation
Topics 2
Data pre-processing and system evaluation
- Feature vectors
- Feature selection
- Dimensionality reduction (Principle Component Analysis)
- Data preparation into training, validation and test datasets
- Cross-validation
- ROC (Receiver Operating Characteristic) curve
- Recall-precision curves
- Evaluating numeric prediction: root mean-squared error, root relative squared error, correlation coefficient
- Overfitting and generalization
Topics 3
Python for Machine Learning
- Review of Pandas, Numpy and Matplotlib
- Scikit-learn
- Statsmodels
- Tensorflow
Topics 4
Machine Learning Techniques - 1
- K-Nearest Neighbours
- Naïve Bayes
- Decision Trees
- Association rules
- K-Means clustering
Topics 5
Machine Learning Techniques - 2
- Numeric prediction: Linear Regression
- Decision boundaries
- Linear classification: Logistic Regression
- Linear classification: The Perceptron
Topics 6
Machine Learning Techniques - 3
- Neural Networks
- Multilayer Perceptron
- Gradient descent
- Training using error backpropagation
- Neural Networks as classifiers
- Neural Networks to learn functions
Topics 7
Machine Learning Techniques - 4
- Deep Learning
- Deep Feedforward Networks
- Convolutional Neural Networks
- Recurrent Neural Networks
- Deep Learning applications
Topics 8
Machine Learning Software Tools
- WEKA
- WEKA Explorer
- Data pre-processing
- Building and testing classifiers
- Clustering
- Creating Association rules
- Matlab/Octave
- R
- Cloud-based platforms
- Amazon
- Microsoft
- IBM
Topics 9
Machine Learning Applications in Industrial Automation - 1
- Industrial IoT
- Artificial Intelligence for Industrial IoT
- Smart instruments as data generators
- Product demand forecasting
For each cited application, the problem will be analysed, the machine learning solution presented, and the obtained results will be discussed.
Topics 10
Machine Learning Applications in Industrial Automation - 2
- Predictive maintenance
- Condition monitoring
- System identification
For each cited application, the problem will be analysed, the machine learning solution presented, and the obtained results will be discussed.
Topics 11
Machine Learning Applications in Industrial Automation - 3
- Autonomous vehicles
- Localization
- Movement planning
- Scene understanding
- Image processing for autonomous vehicles
- Classification
- Clustering
For each cited application, the problem will be analysed, the machine learning solution presented, and the obtained results will be discussed.
Topics 12
Review
In the final weeks students will have an opportunity to review the contents covered so far. Opportunity will be provided for a review of student work and to clarify any outstanding issues. Instructors/facilitators may choose to cover a specialized topic if applicable to that cohort.
Engineers Australia
The Australian Engineering Stage 1 Competency Standards for the Professional Engineer, approved as of 2013. This table is referenced in the mapping of graduate attributes to learning outcomes and via the learning outcomes to student assessment.
Stage 1 Competencies and Elements of Competency |
|
1. |
Knowledge and Skill Base |
1.1 |
Comprehensive, theory based understanding of the underpinning natural and physical sciences and the engineering fundamentals applicable to the engineering discipline. |
1.2 |
Conceptual understanding of the mathematics, numerical analysis, statistics, and computer and information sciences which underpin the engineering discipline. |
1.3 |
In-depth understanding of specialist bodies of knowledge within the engineering |
1.4 |
Discernment of knowledge development and research directions within the engineering discipline. |
1.5 |
Knowledge of engineering design practice and contextual factors impacting the engineering discipline. |
1.6 |
Understanding of the scope, principles, norms, accountabilities and bounds of sustainable engineering practice in the specific discipline. |
2. |
Engineering Application Ability |
2.1 |
Application of established engineering methods to complex engineering problem-solving. |
2.2 |
Fluent application of engineering techniques, tools and resources. |
2.3 |
Application of systematic engineering synthesis and design processes. |
2.4 |
Application of systematic approaches to the conduct and management of engineering projects. |
3. |
Professional and Personal Attributes |
3.1 |
Ethical conduct and professional accountability. |
3.2 |
Effective oral and written communication in professional and lay domains. |
3.3 |
Creative, innovative and pro-active demeanour. |
3.4 |
Professional use and management of information. |
3.5 |
Orderly management of self, and professional conduct. |
3.6 |
Effective team membership and team leadership. |
Software/Hardware Used
Software
- Python with Pandas, Numpy, Matplotlib, Scikit-learn, Statsmodels, Tensorflow
- WEKA: https://waikato.github.io/weka-wiki/downloading_weka/
- R with Caret package: https://www.r-project.org https://cran.r-project.org/web/packages/caret/index.htm
- MATLAB with Deep Learning Toolbox
Hardware
- N/A