Could not download file: This paper is available to authorised users only.


K. Bryceson, A. Navas Borrero, C. Camilleri, F. Vasuian

The University of Queensland (AUSTRALIA)
The use of satellite data or airborne data has been used in agriculture for many years primarily as a useful way of collecting spatial variability data covering many hundreds of hectares at one time in order to use the knowledge so gained to manage crops, livestock and water more efficiently. The ultimate goal of such data collection being to provide a greater profitability to the users of the data. Unfortunately the cost of acquisition and processing of this type of remotely sensed data has proven prohibitive to most agricultural managers who have also had the difficulty of finding staff who are adequately trained in processing and using the data.

There are two issues here:
(i) the data collected – its cost (generally from private or government agencies), it’s revisit frequency (how often the satellite, plane etc returns to the same spot on the earth), its resolution (and thus its usefulness); its dependence on cloud/haze free atmosphere (for quality imagery) plus the necessary computing processing grunt required to deal with it; and
(ii) the lack of skilled personnel in the agricultural sector who can effectively make use of the data for increasing profitability on farm.

In terms of the data collected – in the last 5 years much satellite data can now be obtained free through NASA or various National government spatial databases, but there still remains the issue of resolution and useful (for agriculture) revisit frequency. During the same time frame however, there has been an exponential growth in the miniaturisation of electronic equipment that has driven the development and use of small drones. These drones have primarily been either multi-rotor or fixed wing flying with appropriate sensors payloads in order to obtain low cost imagery at useful revisit frequencies. Such data can then be used quite easily with readily available software on the equivalently increased computing hardware now available at everyone’s fingertips. However, in terms of availability of skilled personnel – this still lags behind as many agricultural programs are not incorporating remote sensing and associated spatial variability analysis into their curricula.

This paper will describe how with the setup of an Agricultural Remote Sensing Lab at The University of Queensland (UQ) in late 2014, driven by new undergraduate course developments in Precision Agriculture and the role of specifically designed drones with miniature sensors capable of recording agriculturally useful remotely sensed data, has proved successful in stimulating interest and engagement in both traditional agriculture / agribusiness students as well as non-traditional (Engineering and Science) students, to engage with these technologies and techniques.

Four ‘Design & Build’ projects will be described in the paper and the role they have played both for the industry partners and the students involved. These projects will include the ‘BeneficalBugDrone’, the ‘NetDrone’, the “UQMiniAgDrone’ and the ‘RFIDDrone’. Challenges both to the organisation in undertaking such developments and to the students in addressing subject matter outside of their core agricultural interest (for example, understanding the physics of light is a key component of understanding how remotely sensed data can be used for agriculture purposes, as is the need to understand some aspects of avionics and electrical engineering in order to know how to design and build a drone), will be discussed.