Atacama Field Trial 2014 - Interview with Mark Woods

How Autonomous Robotic Explorers Can Raise the Science Benefit of Future Exploration Missions

Interview with SCISYS' Robotics expert Mark Woods on the upcoming Atacama Desert Field Trial.

Imagine a new type of robotic explorer which can autonomously adapt to a changing environment in order to conserve energy and navigate more accurately, safely and reliably. This is the goal of Chameleon – a project funded by the UK Space Agency through their CREST2 programme and awarded to SCISYS and the University of Oxford. In this Interview Mark Woods, Head of Autonomy and Robotics Group at SCISYS, explains the main objectives of the Atacama field trial 2014.

Mark, you were part of the SCISYS team during the SAFER project that successfully demonstrated advanced navigation and self-localization technologies as well as the interaction between the rover in Chile and the control centre in the United Kingdom. What are the goals for this new activity called CHAMELEON?

Mark: The main objective of the CHAMELEON project is to conduct fundamental research on future technology solutions for sensing and autonomous navigation applied to extra-terrestrial exploration rovers. In the upcoming field trial we will investigate techniques which allow mobile platforms to adapt and optimise their sensing and perception approach so that they accommodate local terrain conditions in a dynamic and autonomous way. By doing so, we hope to enable future rovers such as ESA’s Sample Fetch Rover to use their scarce energy supplies more optimally and therefore travel further allowing them to meet their mission goals.

Our work also includes the remote navigation of a rover during night time which is –due to limited power supply – currently not possible. This is mainly because most rovers are heavily restricted by their dependence on the Sun for both energy and lighting to help them to see their way as they traverse this incredible environment. At present, these rovers usually only travel during the middle of the day when the sunlight provides maximum surface illumination. This limits the time available for exploration significantly.

Finally we are also investigating the use of Unmanned Aerial Vehicles (UAV’s) to provide high-resolution overhead data. We are able to build 3d maps of the terrain using such data which enables us to explore future concepts such as co-operative aerial and ground surveying of remote terrains.

Is this work only relevant to Space Robotics?

Mark: Absolutely not. The end product for this research will be a rover system which can build 3D maps/images of an area 24/7, by autonomously navigating without GPS using very little battery power. This  is of interest to many terrestrial applications in domains such as mining, nuclear and the oil and gas industry where remote assets are difficult to reach but must be monitored on a continual basis.

Why is the Atacama Desert particularly suitable for field trials in robotic research activities like CHAMELEON?

Mark: First of all, we can find a very Mars like terrain in the Atacama. By developing an understanding of how certain active and passive sensors behave under expected Mars conditions it will be possible to develop basic models which in turn enable the two major goals of this work namely planning to perceive and intelligent sensor selection and fusion. Experience has shown that such trials greatly enhance the downstream impact and credibility of this work. By testing in one of the most extreme environments on Earth we are able to build realistic and robust solutions for both space and terrestrial applications. Another important aspect is the nearly guaranteed availability of sunlight as the main power source during daytime. As you can expect from a desert, neither clouds nor rain occur particularly frequent in the Atacama.

You already mentioned the limitation of time for exploration due to the dependence on sunlight. Does the CHAMELEON project also particular focus on this issue?

Mark: Definitely yes. Sensors and their associated software are essentially the eyes and ears of a rover but each sensor has an associated energy cost. For example the rover might choose to use information from its wheels which tell it how many wheel ticks have been recorded since it started to move. This is an easy way of working out how far the rover has gone. It’s not very accurate though on some terrain such as soft sand for example. An alternative is to use its cameras and sophisticated software to work out the current rover location. This is much more accurate and isn’t fooled by sandy terrain in the same way but it is a lot more expensive in terms of energy. One of the major objectives of CHAMELEON is to characterise and baseline new low-power and low-mass active sensors which permit operation beyond middle of the day lighting conditions. By understanding how the sensors perform and how much energy they use, we can develop autonomy technologies that allow the rover to select different modes for different types of terrain.

How can interested readers find out more about CHAMELEON and the 2014 Atacama field trials?

Mark: Like in 2013, there will be a Blog that displays recent messages and information about the activities and achievements of the Atacama field trial 2014. Furthermore, you can follow SCISYS Space on Twitter (@SCISYS_SPACE).


Christoph Wichmann
Business Communications
Tel. +49 234 9258 115
marketing @scisys .de