News

Where am I? Place recognition for robots.

Our upcoming article in Robotics and Autonomous Systems describes a method by which a mobile robot can determine, based on laser range data, if it has visited a location before; and - if so - can orient itself within the existing map. This is the robot's equivalent of you being left blindfolded in an unknown location within a city and looking around to determine where you are so that you can find your way home. The technology can also be used to efficiently determine if a loop has been closed while incrementally building a map, or to merge maps generated at different times or by multiple vehicles.

Building a mathematical model for generic neural microcircuits

Reservoir computing networks have recently received attention as a potential mathematical model for generic neural microcircuits. They also show promise as a computational technique for technical applications such as time series prediction, however they suffer significant variation in performance when applied to different problems and require the development of problem-specific optimisation procedures. We have investigated methods for addressing this issue in collaboration with researchers in Japan and Taiwan, and our findings have recently been published in the Human Frontier Science Program Journal. Our results have both potential for improving engineering applications, as well as implications for understanding how biological neural networks are shaped to be optimally adapted to the requirements of their environment.

Modelling the movement and behaviour of cattle

The journal Ecological Modelling has published results from experiments using high sample rate GPS data to develop a model of cattle movement and behavioural state (relocating, grazing, and bedding). This novel animal modelling methodology can successfully recognise and model each individual animal, as well as their herding behaviour. It not only shows that cattle graze pastures in a non-homogeneous way but also begins to identify some of the underlying processes that lead to uneven grazing pressure. Understanding herbivore landscape grazing interactions can lead to improved management intervention strategies such as identifying the optimal location of watering points to prevent localised overgrazing effects.

Autonomous helicopter flies beyond visual range

The maturity of ARCAA - CSIRO rotary UAV platform was on display last week as it successfully demonstrated Beyond Visual Range (BVR) flight with the remote control transmitter powered down and the helicopter flying autonomously out of range of the backup pilot. Flight trials also demonstrated various failsafe modes such as automatic return on loss of communications and in the event that the UAV finds itself "out of bounds" (ie. outside a pre-defined mission boundary). This is a significant milestone in the development of a dependable autonomous helicopter. ARCAA (the Australian Research Centre for Aerospace Automation) is a joint venture of CSIRO and Queensland University of Technology.

Cloud Computing student competition and internships

The Australian Academy of Technological Sciences and Engineering is calling for for proposals advocating novel applications of computing clouds answering the questions: how could a cloud platform be used to solve a previously intractable problem or to deliver services in a new and scalable way?

Continuous Monitoring of Reservoir Water Quality

The September 2009 issue of the Journal of the Australian Water Association features an article on the Lake Wivenhoe Integrated Wireless Sensor Network. This project is a collaboration between CSIRO, Seqwater and the University of Queensland, and involves a deployment of floating sensor nodes, land-based nodes, and an autonomous boat. The system monitors processes that affect water quality and provides this information to operators and modelers to allow early warning of deteriorating water quality.

Efficient 3D mapping for mobile robots

Our work on efficient 3D mapping for mobile robots has been published in the Autonomous Robots Journal. Whereas most current navigation algorithms in mobile robotics produce 2D maps from data provided by 2D sensors, we have developed a novel algorithm for efficiently constructing 3D maps. Using this technique we mapped a 200 square meter office environment in 3D at 0.02m resolution and demonstrated that a robotic vehicle with only a 2GHz processor and no prior knowledge of its position could locate itself within 5 seconds.

UAV technology offers hope to lost bushwalker Outback Joe

Students from Aviation High and Brisbane Grammar were the big winners at the UAV Challenge – Outback Rescue competition for 2009. Conceived by researchers from CSIRO and QUT as a way of fostering and promoting the unmanned aerial vehicle (UAV) industry in Australia, the event is one of the largest robotics challenges in the world, and the highest stakes UAV challenge.

Dust storm delivers unexpected opportunity

The dust storm that engulfed south east Queensland yesterday provided an unexpected opportunity for researchers from the Autonomous Systems Laboratory to test our autonomous vehicle navigation systems in adverse conditions.

Twin wins at the Engineering Excellence Awards

The ICT Centre has had two wins at the 2009 Australian Engineering Excellence Awards.

Automation technology for the aluminium industry

Dr Ash Tews presented our experience with automating hot metal carrier vehicles at the Aluminium Cast House Technology Conference on Queensland's Gold Coast. The presentation was enthusiastically received by senior executives in the aluminium industry.

Robust outdoor visual localization

Our research towards a robust outdoor visual localization system for autonomous vehicles in outdoor industrial applications has been published in the Journal of Field Robotics. The paper shows that reliable visual localization can be achieved in challenging outdoor conditions by combining an invariant map, a robust multihypothesis localization algorithm, and an intelligent exposure control algorithm.

Comparison of sensors in dust and rain conditions.

The Journal of Field Robotics recently published results from our comparative evaluation of millimeter-wave radar and two-dimensional scanning lasers in dust and rain conditions. Research challenges included the development of a robust and reliable method for measuring the level of suspended dust (or other obscurant media) by determining the transmission coefficient. This method was used for quantitative assessment of sensor performance.