Dynamic Situational Awareness
Developing new methods to provide dynamic situational awareness for humans and machines to optimally understand each other’s situation as events evolve within dynamic missions, therefore, to work effectively together.
In order to work together effectively, humans and machines need to understand each other’s situation in an optimal manner (not too high or not too low). For example, a human working with a semi-autonomous vehicle needs a good understanding of where the vehicle is, its direction and speed of travel, its surroundings, and so on. Similarly, artificially intelligent systems must know how much information the human operator has and their needs, in order to make good decisions. Human operators must neither be kept in the dark nor overloaded with information if their expertise is to be properly used. Many recent approaches to delivering situational awareness in human-machine systems focus overly on reducing the information overload (‘cognitive loading’) problem. However, this often leads to “dumbing down” or marginalising the human’s role.
This project takes on the challenge of developing new methods to provide dynamic situational awareness, changing the amount of information, what is presented, and how it is presented as a response to the user’s current state, the machine’s state, and mission progress. It will research innovative ways to enable dynamic situational awareness, applied and demonstrated on CSIRO’s award-winning ‘DARPA human-robot team’ (https://www.csiro.au/en/research/technology-space/robotics/darpa-challenge ). It will consider how both the state of the user and the mission affect the user’s information requirements, and how these should be adjusted as the mission progresses. It will also consider how best to keep the user engaged, and how to respond when something goes wrong.
Optimal dynamic situational awareness allows humans to work with machines to boost human capabilities and enhance human decision making and performance, an important step towards the realisation of CINTEL “superteams” which combine human and machine intelligence.
Adaptive Interfaces and User Modeling
A project led by Hashini Senaratne, and supervised by Cecile Paris, Dana Kulic, David Howard, Jason Williams and Pavan […]