¹û¶³Ó°Ôº

IoT is a growing field encompassing digital and physical technologies with applications across different industrial sectors, and spanning manufacturing, consumer electronics, security, mining, design, transport, exploration and healthcare.

Ìý

The implementation of such systems encompasses great engineering challenges, both related to the sensing and control of the real world (e.g., cyber-physical systems), and to the need to deal with large collections of sensing/actuating nodes (agents), distributed across the real environment (factory, field, building).

Ìý

This module will provide you with the knowledge required to creation, deploy and exploit such distributed networks of agents. We will investigate this challenge through the lens of multi-agent systems, an area of information technology dealing with the management and control of systems composed of multiple interacting software/hardware components (known as agents). Mutli-agent systems rely on and exploit cooperation among agents/ systems, as to allow them to jointly solve problems that are beyond the abilities of any individual member and are key to exploiting the heterogeneous combination of systems typically involved in IoT systems.

Ìý

In this module you will learn most extended multi-agent approaches (algorithms). However, rather than exploring such systems from a theoretical point of view, we will strongly focus on their actual deployment and real-world applications. More specifically, we will make use of the range of distributed sensor/actuator networks available across ¹û¶³Ó°Ôº East and our dedicated studio/ lab, using them to illustrate the deployment, management and use of multi-agent techniques.

Aims:

The aims of this module are to:

• Understand the concept of an agent and a multi-agent system, and their main scope of application.
• Critically analyse the main issues surrounding the creation and deployment of multi-agent systems.
• Acquire main approaches and techniques for enabling communication and cooperation in such multi-agent systems, via direct analysis of real deployments across ¹û¶³Ó°Ôº East.
• Apply multi-agent techniques to real world multi-agent deployments existing within ¹û¶³Ó°Ôº East.

Intended learning outcomes:

On successful completion of the module, a student will be able to:

1. Understand the notion of an agent, how agents are distinct from other software paradigms (e.g., objects), and understand the characteristics of applications that lend themselves to an agent-oriented solution.
2. Describe the key issues associated with constructing agents capable of intelligent autonomous action and identify most suitable approaches taken to developing such agents.
3. Recognise key factors governing high-level communication in multi-agent systems (infrastructure, communication, power, resilience, redundancy) and develop techniques for automated decision-making for multi-agent systems according to these factors.
4. Apply techniques to create real world multi-agent systems, deploying them in ¹û¶³Ó°Ôº East.

Indicative content:

The following are indicative of the topics the module will typically cover:

Delivery will cover most common multi-agent techniques (e.g., deductive reasoning, practical reasoning, reactive and hybrid architectures, coalition formation), covered in related textbooks. However, instead of following the theoretic and application agnostic approach in most textbooks, each multi-agent technique will be presented in the context of a specific real-world sensor deployment. This will describe the application context, infrastructure and technology stack used, and then use it to introduce one specific multi-agent technique and how it is used in this specific context. Technical descriptions of existing sensor deployments across ¹û¶³Ó°Ôº East will be used wherever possible. Applications of the technique to other domains/ sensor networks will also be introduced during the lecture, but only after having motivated them within the context of a specific real-world deployment.

Requisites:

To be eligible to select this module as an optional or elective, a student must be registered on a programme and year of study for which it is formally available.