In the near future, robots will need to be deployed in large numbers, coordinated across multiple agencies and with fewer operators, if we are to make the best use of their capabilities. This project will therefore focus on the design of AI algorithms and interfaces which will ensure that human operators aren't overloaded and that they understand the automated actions taken by the swarms. The creation of such human-machine teams will help speed up the deployment of such swarms in disaster response, where time and human resources are limited and there's an urgent need for rapidly gathering situational awareness.
Explaining the science
Artificial intelligence algorithms that are used to coordinate swarms of drones can efficiently search through millions of route plans for each member of a swarm, but the end-result of such computations may not be obvious to human operators nor easily modifiable. It is therefore important to provide tools to human operators to understand such machine-based decisions while at the same time design the algorithms to suit human interventions.
This project aims to establish a set of use cases, underpinned by real-world or synthetic datasets. Benchmarking existing solutions within such use-cases, will help to draw out the main research questions that arise when designing, implementing and deploying such human-machine teams.
The project will involve collaborating with key practitioners from the disaster response domain, with the ambition to develop tools and techniques that will help them carry out aerial surveys and situational awareness missions faster and better. Drawing upon their feedback, this project will help establish the basis for a larger programme of work.
Thales is providing use cases for the use of drones as well as feedback on the solutions developed.
This work applies to a number of significant domains where drones are used to carry out surveys. This includes precision agriculture, disaster response, smart grids and transportation. Specifically, there will be a focus on disaster response as a key area where it is important to rapidly deploy swarms and where human operators are in short supply, driving the need for high degrees of automation.