Monday 23 September 2024
Font Size
   
Thursday, 23 September 2010 19:00

The Future of Autonomous Mars Rovers

Rate this item
(0 votes)

By Emmet Cole, Wired UK

Researchers from across Europe — with a little help from experts at NASA’s Jet

Propulsion Laboratory — are working on a new, robotic exploration system that could enable future Mars rovers to independently explore the planet’s surface, identifying geological and biological samples and performing their own terrain hazard analysis.

Scientists hope that by enabling robotic explorers to autonomously scout their landscape for potential dangers and areas of interest, the PRoViScout system will overcome one of the primary obstacles to efficient robotic exploration of Mars — the time lag involved in sending and receiving messages to and from bots on the Martian surface. If successful, PRoViScout will enhance rovers’ mission-planning intelligence, dictating how resources should be deployed at any given time. Visual sensors, including cameras and a unique laser-fluorescence life-detection system that uses next generation Blu-ray technology will identify potential samples and hazards.

Meanwhile, samples already taken will be subjected to on-board scientific analysis. If the team achieve their goals, the technologies and system architecture developed during the 30-month, 2.69m Euro project, will be included on the European Space Agency’s (ESA) ambitious Mars Sample Return project, which is expected to launch in the early 2020s. The bandwidth between Earth and Mars is the equivalent of a few low-quality MP3s each day and messages take anywhere from five to 20 minutes to travel in either direction, so real-time “joy-sticking” of Mars rovers is an impossibility, says Mark Woods, who leads the robotics and autonomy group at SciSys, a UK-based technology company involved in the PRoViScout project.

“It depends on the orbits, but we’re roughly anywhere between 400,000 and 1 million kilometres from Mars at any given time and the bandwidth is really low. Sometimes it takes several days to receive the images taken as a rover spends just one day moving over the surface,” explains Woods. Using current technology, scientists try to overcome this problem by sending a bunch of commands at once and then waiting for the rover to confirm that the commands have been executed. This process takes “forever,” says Woods, and can potentially lead to bots missing sites of geological interest.

In 2009, NASA announced that its Opportunity rover had discovered “Block Island” — a large (60 cm) meteorite found on the Martian surface — an important discovery that revealed the Martian atmosphere was much thicker in the past. However, Opportunity almost missed the find because it didn’t have the on-board intelligence to pick up on the target, explains Woods. “If you look at what happened — and we know some of the guys working on this over at NASA — it was in the order of 10-17 days before the guys on the ground got some low-resolution images back, which just about show this rock in some of the images. They then decided that they needed to go back and visit it and confirmed the find,” says Woods.

“In many respects, they were very lucky, because, as their experience shows, you can have a rover go right past a pristine science target without knowing it and you’re looking at 10-plus days out before the guys on the ground realise what’s happened. Potentially, they might not even realise it — there’s always a risk that scientists on Earth might miss small targets that don’t show up clearly in the images.”

Mimicking the intuition and field practice expertise of a human geologist is going to be a huge challenge, says Woods. “We’re at the very beginning of the process of trying to replicate in software what geologists do using their intuition, experience and human intelligence. Geology is more of an art than a science in some ways and it’s open to subjective analysis, even for the professionals, with different geologists offering different opinions on what a rock looks like,” explains Woods.

PRoViScout will use cameras closely related to the PanCam — a panoramic camera designed for the 2018 ExoMars mission — to perform digital terrain mapping and search for clues indicating past biological activity preserved on the texture of surface rocks. PanCam consists of two wide angle cameras with multi-spectral stereoscopic imaging capabilities and a high-resolution camera for high-resolution colour imaging.

Meanwhile, a parallel European project called PRoVisG, is looking at building 3D models of landscapes from PanCam imagery and combining that with high-resolution images taken from space to further improve hazard detection. Central to PRoViScout’s chances of success is a unique life-detection system based on the use of laser-fluoresence, being developed by Jan-Peter Muller, of Univeristy College London’s Department Of Space & Climate Physics.

“It’s well known in biology and biochemistry that when subjected to particular wavelengths of radiation, organic materials will fluoresce at other wavelengths — wavelengths that are usually longer than the original ones,” says Muller, who leads the team tasked with developing PRoViScout’s organics and life-detection system. On the Martian surface, rovers will look for organic materials known as polycyclic aromatic hydrocarbons (PAHs), which are often touted as molecular candidates for early forms of life.

PAHs turn yellow or green when subjected to UV-light. Muller’s team has built a prototype system using laser diodes originally designed for reading next-generation Blu-Ray discs — a decision led by economies of energy and size. “The reason why we’re using these lasers is because nowadays you can get tiny little sold state lasers which are basically just a few milimetres across, and use tiny amounts of power … like half a watt,” explains Muller.

Rovers equipped with lasers can scan the Martian landscape and, potentially, detect minute traces of organic matter amongst the rocks and dust. Muller hopes that a modified version of his laser-fluorescence system will be included on the joint NASA-ESA ExoMars mission, which is expected to launch in 2018. “We hope to use it on ExoMars on the material that is dug up by the drill. But in PRoViScout, we’re primarily concerned with scouting. That is, exploring the terrain and looking for potential targets that might be looked at in much greater detail by other instruments, including Raman/LIBS and the Life Marker Chip,” explains Muller.

Those supporting technologies are critical because the ProViScout team faces several challenges ahead when it comes to identifying organic life-forms using laser-fluorescence says Alan Waggoner, Director of the Molecular Biosensor and Imaging Center at Carnegie Mellon.

“Lasers in the UV range create background fluorescence from some types of minerals, so you have to watch out for that. I hope that the organic molecules Muller is looking for are in a high enough concentration to overcome background signal issues,” says Waggoner, who has developed a flash-lamp and dye based system for detecting organic material as part of a NASA-sponsored project.

Combining laser-fluorescence for large-scale surveys with technologies that enable more detailed analysis will make it possible to find the hard evidence for life on Mars, says Nilton Renno, from the University of Michigan’s Department of Atmospheric, Oceanic and Space Sciences, who is not involved in PRoViScout. And with evidence of both methane and water on Mars already acquired, we can expect to find proof of bacterial life on the red planet in the next 10-12 years, says Renno, who led the Atmospheric Team Group on NASA’s Phoenix Mission to Mars.

“On Earth, everywhere we find liquid water, there is life. It doesn’t matter how acidic it is or how saline the water is, if there is liquid water, there is bacterial life. And Mars is the most Earth-like planet in the solar system. I am pretty confident, based on the data that we have that there is liquid water on Mars today. So, there is life,” says Renno. Renno recently submitted a proposal to NASA for a Trace Gas Microwave Radiometer (TGMR) designed to reveal the processes that lead to the production of methane on Mars. Renno hopes his device will be incorporated in a joint NASA-ESA mission to Mars in 2016, (aided in no small measure by the TGMR theme song). PRoViScout is not the only system on the block that promises greater rover autonomy.

Last winter, NASA scientists uploaded software to the Mars Opportunity rover that enables it to autonomously identify sites of potential geological interest, photograph them and send the resulting images to Earth. The Autonomous Exploration for Gathering Increased Science (AEGIS) system also allows scientists to change the criteria used for choosing potential targets. For example, in some environments, rocks that are dark and angular can be identified as higher-priority targets than rocks that are light and rounded.

For UCL’s Muller, the most difficult challenge is “as always, financial”. “We’re trying to do 21st century science with late 20th century funding. That’s a good challenge, and a lot of the engineering that we’re doing is certainly state-of-the-art. But we’re doing it with a fraction of the resources that would be available to our colleagues and competitors in the U.S.,” says Muller. As part of the PRoViScout project, European researchers will exchange information with experts at NASA’s JPL who have worked on the Mars Spirit and Opportunity rovers as well as the Mars Science lander.

“The Americans have been doing this for very many years and we are playing catch-up in Europe. But we want to play catch-up on our own terms. They get something out of it, because they learn the ideas we have and how we’re planning to implement them. We benefit from the fact that they are the early-adopters and hopefully we’re the ones that can learn from their mistakes and not make too many of our own,” says Muller.

PRoViScout in action

In July, to help prepare for eventual field trials, testing and algorithmic development, Dr. Derek Pullan from the University of Leicester surveyed a beach in West Wales called Clarach Bay. As part of this work Pullan outlined an example route that he would like an autonomous bot to explore and identified the kind of targets he would expect the bot to detect and examine. (Note: the yellow cable is for power, not control.)

PRoViScout also includes experts from Aberystwyth University, the Czech Technical University, the University of Leicester, DLR(The German Aerospace Centre), the University of Strathclyde, https://www.joanneum.at/en/jr.html, CSEM, GMV, and TraSys.

Image: Artist’s impression of ESA’s ExoMars Rover./ European Space Agency

See Also:

Authors: Emmet Cole

to know more click here

French (Fr)English (United Kingdom)

Parmi nos clients

mobileporn