Menu
Menu
'Freeing' robots: From 3D mapping to autonomous systems

'Freeing' robots: From 3D mapping to autonomous systems

The CSIRO's Michael Bruenig looks forward to a future where robotic systems will be able to easily navigate complex environments, whether a disaster site or a factory floor

The CSIRO's Zebedee. Image credit: CSIRO.

The CSIRO's Zebedee. Image credit: CSIRO.

We’ve recently used the breakthrough technology to create the first 3D map of the interior of Italy’s landmark Leaning Tower of Pisa. Previously, tight spaces and the repetitive nature of the internal structure prevented it from being captured. Our detailed record may one day be critical in being able to reconstruct the site if it was to suffer catastrophic damage due to natural disasters such as a fire or an earthquake.

In 2012, CSIRO worked with 3D Laser Mapping, a global developer of laser scanning solutions to commercialise the research into the ZEB1 product. It is being used to increase efficiencies and improve productivity in a number of different industries. For example, the technology is assisting mining companies to better manage their operations and helping security forces to quickly scan crime scenes. It has been used for cave mapping as well as for mapping cultural heritage sites.

However, the crucial step of bringing the technology back to the robot still has to be completed. Imagine what is possible when you remove the barrier of using an autonomous vehicle to enter unknown environments (or actively collaborating with humans) by equipping robots with such mobile 3D mapping technologies.

Due the simplicity, they can be significantly smaller and cheaper while still being robust in terms of localisation and mapping accuracy. Imagine how small such a robot could be using just a laser scanner and an inertial unit. If needed, the available 3D information can be easily augmented with vision or hyper spectral imaging, providing a comprehensive understanding of the surroundings, for example real time 3D heat maps.

Driving innovation in agile manufacturing

A specific area of interest for this robust mapping and localisation is the manufacturing sector where non-static environments are becoming more and more common and where the cost and complexity for each device has to be kept to a minimum. With a trend towards more agile manufacturing setups, the technology enables lightweight robots that are able to navigate safely and quickly through unstructured and dynamic environments like conventional manufacturing workplaces.

The steps towards simpler and more robust robotic systems almost certainly involve maximizing the information that can be derived from individual sensors. While using multiple sensors and fusing the sensor information is a necessary step to exploit the information available to a robotic system, one might wonder if this step is often too early in the design phase.

Currently, the resulting systems carry more sensors than necessary, are difficult to optimize and are less robust than leaner robotic systems for the same task. The increasing computing power available to small systems further exacerbates the problem as robotic systems can be designed without much consideration of what information power individual sensors provide.

However, it is worth pushing the boundaries of what information can be extracted from very simple systems. Similar to the Zebedee system, it may be the feasibility of innovative system designs that opens the door to entirely new applications and markets.

Michael Bruenig is deputy chief of CSIRO’s Computational Informatics Division

Join the CIO Australia group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.

Join the newsletter!

Or
Error: Please check your email address.

Tags robotsrobotics

More about CSIROCSIROLaser

Show Comments
Computerworld
ARN
Techworld
CMO