The aim of this project is the design of sensor networks with sensing and computing capabilities to control dynamic systems. The research will focus on two main areas: UAVs and distributed parameter systems.

The safe operation of UAVs demands the same or better accuracy from control and navigation systems than is needed by their manned equivalents. This in turn requires accurate sensors for determining location and motion. The size of UAVs is often considerably smaller than manned vehicles and necessitates that these sensors be small, lightweight and rugged. These requirements tend to make the current generation of sensors too expensive for cost-effective implementation on UAV platforms, which in all other respects can be manufactured more cheaply. There is therefore a real need for development of new sensors for use on UAVs. The lack of affordable sensing for small UAVs is prohibiting their widespread use.

The areas of research that should bear the greatest rewards appear to be:

- development of more accurate inertial sensors using MEMS technology,

- better sensor fusion algorithms to implement useable control algorithms with existing cheap sensors,

- control of flight using robust biologically inspired visual sensing methods,

- development of higher accuracy positioning systems to outperform GPS type networks.

Distributed parameter systems like flexible structures and acoustical systems are easy to control if information about individual vibration modes can be obtained. Recently there papers have appeared that use sensor fusion and averaging techniques to achieve this. Unfortunately the work is confined to structures that have a regular geometry and ideal boundary conditions. This work is hard to extend for practical situation. In this research systems will be designed with cheap MEMS sensors in mind. The idea is to embed hundreds of these sensors on flexible structures and then use combinations of those sensors to obtain the required mode vibration information. Firstly it needs to be determined as to which combination gives which vibration modes. Secondly the output from these combinations can be used to provide adequate control. The entire process can be automated and since cheap MEMS sensors can be embedded the cost of sensor and control combination can be kept very low.

Unassisted landing of UAVs on moving platforms can eliminate pilots from several applications where lightweight UAVs are used. This will reduce the cost of operation to a small fraction of the present costs. The design methods for distributed system parameters will have significance for aircraft structure, submarine body, and other hi-tech applications.

The coming together of people in the sensor networks (SN) and control community will provide an extraordinary collaboration. The SN community will have a new area of application. The present applications of SNs are limited to static systems. Many control applications cannot improve their performance unless aided by intelligent sensors and actuators. In the case of distributed parameter systems like aircraft structure, it is not uncommon to need thousands of sensors for control applications. This type of technology can only be developed by the collaboration of sensor and control community.

Could put people working on real-UAV platform development and distributed parameter systems in touch with sensor and signal processing experts who may be able to assist in the development of more accurate sensing techniques.

With such a big coastline, Australia can benefit greatly from automating surveillance tasks using autonomous aerial vehicles. Small UAVs can be employed in many counter-terrorist or law-enforcement applications. For example, could fly a micro-UAV next to the window of a skyscraper to peer in at the situation unfolding inside. Australia is a leading player in acoustic noise control and this technology will directly