This module covers the field of AI, including its main paradigms, historical perspective, state-of-the-art, and applications to engineering and automation problems. Theoretical underpinnings of AI such as knowledge representation, logics, machine learning, approximate reasoning, biologically inspired computational methods, and collective and decentralised intelligence will be examined in detail, in the context of their main implementations such as classical and fuzzy expert systems, shallow and deep neural networks, simulated annealing, evolutionary and swarm algorithms. The teaching heavily draws from current and past research at the Department of Mechanical Engineering, and includes for each topic the examination of practical implementations to engineering problems in fields like robotics, manufacturing, motor control, etc.
Learning Outcomes
By the end of the module students should be able to:
Apply a comprehensive knowledge of the main artificial intelligence paradigms, their historical perspective, and the current state-of-the-art, including state-of-the-art of biologically inspired computational intelligence methods for solving complex optimisation problems in engineering; discussing limitations of the techniques employed and reaching substantiated conclusions.
Apply a comprehensive knowledge of the working principles of the main artificial intelligence tools, machine learning paradigms, and agent-based nature-inspired algorithms to formulate and analyse complex problems in artificial intelligence to reach substantiated conclusions; discussing the limitations of the techniques employed.
Select and apply appropriate techniques to choose and configure appropriate artificial intelligence tools, for solving complex problems, such as classification, regression, clustering, control, optimisation, image recognition, and expert/recommender systems for a broad range of applications such as in robotics, manufacturing, control, systems engineering, product design; discussing the limitations of the techniques employed.
Assessment
Assessment Methods & Exceptions
Assessment: 2hr Centrally Timetabled Written Unseen exam 80% and Coursework test 20%
Reassessment: Supplementary to match the main assessment method with due consideration made to any restrictions imposed at the time of reassessment. Students can carry forward passed assessment components from the main assessment.
