Mars rover drives while scientists choose drill sites

A Mars rover’s scarce operating time can be used for choosing drill sites instead of steering around every obstacle. In Spain’s Tabernas desert, the prototype Charlie travelled more than 60 metres autonomously with less than 1% navigation error while scientists selected targets from its data. If the full system remains dependable after a 2030 landing, a control room in Turin could approve a scientific target while the rover crosses safe ground on its own.

A rover on Mars cannot be rescued after a bad route choice. Its controllers must work out slopes, hazards and useful scientific targets from limited data sent across space. A route that looks safe in a picture may trap or damage the vehicle, and a lost day cannot be recovered. The difficult part is not only making the rover move. It is turning each day’s measurements into the next set of decisions before the mission’s time runs out.

Why driving is only half the problem

Mars rover missions work in command cycles. A team receives data, interprets it, chooses a plan and sends instructions for the rover to carry out later. This delay means scientists cannot simply steer around every stone as if they were holding a remote-control car.

Software for autonomous hazard avoidance can help the rover move safely toward a nearby place. But it cannot decide which rock or patch of ground is most valuable to investigate. Searching for signs of ancient microbial life requires geological context and measurements from several instruments. The rover must reach the right kind of place, not merely any place it can reach.

That makes the scientists’ work closely tied to the driving. They need to read images, subsurface radar and spectral analysis together, then choose both a target and the instruments to use there.

A desert rehearsal for Mars

In September, about 100 scientists and engineers tested this combined process in Spain’s Tabernas desert. The prototype rover, named Charlie, used the guidance, navigation and control software intended for the Rosalind Franklin rover. The science team worked from the Rover Operations Control Centre in Turin, Italy, about 1,200 kilometres away.

The test joined flight-like navigation software, a prototype set of instruments and a team spread between the desert and the control room. Scientists used Charlie’s data to select places to search for evidence of life. Operators planned routes at the same time.

On the sixth day, Charlie travelled more than 60 metres on its own with less than 1% navigation error, according to the rover operations lead. The result does not mean the Mars mission is ready to run without people. It means that continuous, joystick-style driving is no longer the only way to handle a short stretch of ground.

The team also saw why different kinds of data matter. A slope first judged risky from the rover’s data was shown by a digital elevation map to be a gentle incline. A two-dimensional image had made the terrain look more threatening than it was. Better information can turn a cautious stop into a safe route.

The exercise lasted 20 days and included daily data downlinks, technology checks and operating procedures in a demanding environment. Rain interrupted the tests on one day. That interruption was on Earth; on Mars, the larger challenge will be making the complete command process work with the real terrain and communications system.

A morning in the control room

If these procedures remain reliable in increasingly realistic tests, the Rosalind Franklin team could spend more of its Mars operations time choosing where to drill and analyse instead of manually steering around every obstacle. In a control room in Turin, a geologist might begin the morning with a rover panorama and a radar trace, approve a target and find that the rover has already crossed a safe stretch of ground overnight.

That future depends on several parts working together after Rosalind Franklin’s expected landing in 2030: the rover, its instruments, the command chain, autonomous navigation and communications through the ExoMars Trace Gas Orbiter. The next field test, planned for summer 2027, will add challenges involving that orbiter.

The shift is therefore not from humans to machines. It is from humans steering every metre to humans deciding what the rover should investigate while software handles nearby movement. The benefit, if the system holds, is more time for the questions that only the science team can answer. The daily science-command process and autonomous navigation must first remain dependable under actual Mars communication and terrain conditions.

出典

European Space Agency, “European rover trains for Mars in the desert”