CARDyAL: Cooperative Aerodynamics and Radio-based DYnamic Animal Localisation

Lead Research Organisation: University College London
Department Name: Computer Science

Abstract

The primary scientific objectives of the CARDyAL project lie (i) in the challenges inherent in creating very lightweight sensing devices capable of accurate localisation; (ii) in the use of those devices in measuring the dynamically changing relative location of a cooperative group of animals in a range of different contexts for which GPS-based localisation techniques are inappropriate; (iii) relating that dynamic behaviour to scientific questions of cooperative fluid dynamics, energetics, and social biology using appropriate physiological data; and (iv) using the measured information about team organisation to inform a particle swarm optimisation model that will be used as a potential way of reducing energy consumption within and outside this application domain. The ability to make such measurements, and thus to undertake science in these areas, is currently very limited and the problem of achieving this is both challenging in research and engineering terms, and likely to be dependent on the environment for which the particular solutions are created. However, the applications for such technologies are numerous, varied and of significant scientific and strategic importance to the biological sciences community and to the public at large.The technology developed in this project is intended to be inherently transferrable: the hardware and software base for tags capable of relative localisation (a) indoors (b) in long-term deployments (c) in high dynamic environments has applications both to other animal models - understanding the dynamics of groups for purposes of conservation, ecology, welfare or epidemiology - and to the monitoring of humans for better facilities management, workplace design, emergency service and military use, amongst other things. The data reduction and analysis methodologies are applicable to a range of situations in which dynamic and social structure is important, and the relationship between true biological data and a derived PSO model is of substantial scientific interest, and of potential applicability to optimisation problems of many sorts, e.g. energy minimisation, in the computer graphics industry in terms of more accurately representing swarm movements, and in swarm robotics of various types, in path planning and control. In terms of the biology, the availability of this technology in a usable form factor allows research to be performed in a way that is not currently possible, and, consequently, allows the potential to ask scientific questions that are not currently capable of being answered.Thus CARDyAL involves research and engineering in the field of lightweight wireless sensing and wireless localisation that would lack specificity without the constraints and demands of a real application domain, and it involves research in the field of animal sciences that it would not be possible to conduct without the sensing devices and their associated algorithms. There are advances to be made in each of the fields that could not be made without the active engagement of the other and this co-dependency is innately both translational (high risk) and high reward since it is founded on a truly symbiotic programme of research in which neither of the constituent research fields dominates.

Publications

10 25 50
 
Description We have discovered how and why birds fly in a V formation, when cheetah hunt, how disease is likely to propagate through flocks of hefted animals, and whether male leopards avoid each other.

In the process of this, we have captured data about animal behaviour that has never before been held and have developed new mathematical techniques the help in the analysis of that data.
Exploitation Route The work undertaken is substantially scientific in nature; consequently, it is of interest largely in terms of the facts that it uncovers about nature.

The mathematical techniques, published and still in development, are useful in analysing animal movement and behavioural data, and the time series analysis technique developed by an allied PhD student has the potential to be useful in analysing financial trends.
Sectors Financial Services, and Management Consultancy

 
Description The findings to date are essentially scientific in nature. They have been published in high impact journals.