Ilse Daly recently completed her PIPS placement with SCISYS (Space Division) in Bristol. Below is her account of her placement.
I spent my internship working with scientists and engineers in the space division of SCISYS in Bristol. We were working on a new piece of software to be used on the European Space Agency’s next generation Mars rovers. The system will allow the rover to autonomously detect unusual features or objects on the Martian surface that are likely to be of great interest to scientists.
My PhD involves investigating the peculiar and highly complex visual system of mantis shrimp. These amazing animals can not only see at least 15 colours, but are able to detect and distinguish both linear and circular polarised light â€“ the only animals known to have this ability. Given that I have a good working knowledge of one of the most complex visual systems known to science, it was natural that my contribution to the project would be to develop the initial detection method. In other words, I would be helping to the rover to see.
My main contribution to the project was to build a MATLAB implementation of a variety of different saliency algorithms and to evaluate their performance for use with this project. An algorithm which is outstanding at detecting blue cars, for instance, will not be of much use to a robot on the surface of Mars. However, I was surprised to learn just how difficult it is to come up with a good, robust working definition of a ‘good performance’. I also investigated possible ways of building on the performance of an algorithm to make it more suitable to the project. The final report summarised the findings of my virtual experiments and contained my suggested algorithm choice and application. This was sent to the chief scientist in the Autonomy and Robotics Group at ESA. My work was the first completed work in the project, and the first results which ESA received.
All in all I enjoyed my time at SCISYS. Not only did I get to see a group of scientists pretend that they were working with a robot on Mars from the comfort of an Oxford command centre (the robot was in fact in Chile, somewhat closer than Mars, but an infinitely easier place to run trials), but I also gained a useful insight into the way science is practised outside of academia. In addition, I picked up some useful practices that will improve the quality of my PhD work.