Skip to main navigationSkip to main content
The University of Southampton
Centre of Excellence for In situ and Remote Intelligent Sensing

Shared Intelligence for Machines in Extreme Environments

Concept of machine intelligence shared between different locations for real-time remote awareness over limited communication infrastructure
Concept of machine intelligence

 

What?

This PhD will investigate how to make sensors smarter in order to make more informative measurements in remote and extreme environments, such as the ocean and space.

Why?

Robotics and sensing technology have massively increased our ability to collect data in even the most extreme and remote environments of our planet. However, even though 90% of all data that exists has been generated in the last two years, this doesn't necessarily mean that more knowledge is being generated. To generate knowledge, data needs to be processed and analysed. In built environments, such as cities, homes and offices, where our infrastructure has abundant power and communication bandwidth, most processing takes place in remote datacentres and the sensors themselves do not need to be smart to be useful (e.g. Amazon echo). In environments that lack this kind of power and communication infrastructure, it is necessary to process information closer to the data's point of origin (i.e. in situ) so that compressed information can be transmitted so that centralised systems can make rapid, sensible decisions when they are needed. This PhD will look at how intelligence can be split between in situ and remote locations in order to optimise the trade off for different energy and constraints in a flexible way.

How?

In this PhD, the student will develop an approach to determine how much data processing should take place onboard a sensor/robot for a 'given communication bandwidth', 'availability of processing power'. 'speed of decision needed', 'flexibility of decision needed'. The student will apply these methods to improve intelligent data acquisition and remote awareness for:

These will be designs for the different energy and communication bandwidths associated with:

Key Skills

Engineering background with a passion for field and space robotics, willing to work with embedded systems, machine learning/artificial intelligence, computer vision, data management.

Register for PhD Engineering and the Environment

Supervisors

Blair Thornton (Maritime Engineering) - b.thornton@soton.ac.uk        

Jon Hare (Electronics and Computer Science) - jsh2@ecs.soton.ac.uk  

APPLY HERE

 

Privacy Settings