F21ZL - Dependable and Deployable Artificial Intelligence for Robotics 2

Course leader(s):

Aims

This course aims to:

Introduce the students to different approaches to system design, concentrating on issues arising in the design of closed-loop robotics systems operating in human-centred and dynamic field environments.
Ensure students understand how responsible research and innovation issues can and should be incorporated at the stages of concept development and system design, with a specific focus on issues arising in RAS domains.
Make students aware of methodologies for measurement, testing and evaluation of the performance of robotics and autonomous systems (RAS) in field environments.
Through case studies and real-world examples, expose students to dependability and deployability issues arising in the field. This includes discussion and reflection on how design choices, integration and testing methodologies help or hinder safe field deployments of RAS technologies.

Syllabus

1. Introduction to design frameworks for robotics and autonomous systems; setting the RAS context of diverse interacting subsystems and uncertainty in operating conditions (1.1 The design of autonomous systems is increasingly complex, and involves multiple components and tools, and faces challenges of safety, security, sustainability, regulation. The students will learn to appreciate the current evolving landscape.)

2. Study Principles of some selected design frameworks such as human-centred design, biologically inspired design, etc. (2.1 Interactive autonomous systems need to be developed with user-centred methods, that may range for user-friendly interactions to cyber-physical properties of the systems. The students will learn to deploy this range of methods.)

3. Study Principles of Responsible Research and Innovation (RRI) (3.1 There is a huge societal and government demand for industries and academia to develop autonomous and intelligent systems responsibly. The students will learn how to apply principles of RRI in robotics)

4. Master RRI toolkits, and their application to RAS (4.1 The students will learn how to use RRI kits in autonomous systems in practice)

5. Conduct case studies addressing design/testing failures and challenges (5.1 The students will earn to design appropriate case studies for testing failures of complex autonomous systems)

6. Introduction to different methods for testing, including functional testing, regression testing, performance testing, etc. (6.1 Testing methods are widely deployed in industry and are widely varied across sectors. The students will learn to appreciate the range and apply most prominent methods in prcatice.)

7. Practice Data-driven and quantitative performance evaluation methods for RAS (7.1 The students will be trained in deploying quantitative and statistical methods for evaluating performance of robotic systems.)

8. Practice to understand safety cases and standards-driven development methods (8.1 The standards for autonomous and robotics systems are constantly evolving. The students will learn to understand this link, and deploy , 8.2 this knowledge in practical scenarios.)

Learning outcomes

By the end of the course, students should be able to do the following:

Implement principles of design frameworks, e.g., human-centred design or biologically inspired design, in Robotics applications.
Formulate principles of responsible innovation and innovation (RRI) and apply RRI concepts and strategies in particular RAS design contexts.
Comparatively assess system level design choices with respect to dependability and deployability.
Critically assess state-of-the-art methods for field testing of robotics and autonomous systems, including understanding of their strengths and limitations in assessing dependability and deployability, especially in the context of AI components.
Devise and apply quantitative performance evaluation methods for a particular RAS concept and design.
Sketch safety case sections pertaining to design and testing of AI-enabled robotics and autonomous systems.

Further details

Curriculum explorer: Click here

SCQF Level: 11

Credits: 30