Semi-Autonomous Robots Pave the Way for Faster Planetary Exploration

Current planetary surface missions, such as those involving rovers on Mars, operate cautiously due to significant communication delays and data transfer limitations between Earth and the rovers. These factors necessitate advance planning for operations and typically restrict exploration to a small portion of a landing site, limiting the collection of diverse geological data. Rovers are designed for energy efficiency and safety, moving slowly across varied terrain, often traveling up to a few hundreds of meters per day.

A new approach involving a semi-autonomous robotic explorer has been tested, offering a solution to accelerate planetary discovery.

This robot is designed to investigate multiple targets sequentially and collect data without constant human oversight. The results indicate that semi-autonomous robots, equipped with compact instruments, could significantly accelerate resource prospecting and the search for biosignatures on planetary surfaces. This method allows a robot to approach multiple targets and autonomously perform measurements at each location, rather than requiring continuous human supervision for investigating a single rock.

The primary research objective was to determine if a robot with a simple scientific payload could rapidly study several targets while still producing meaningful scientific results. The findings successfully demonstrated that compact instruments are capable of achieving the full scientific objective, including identifying rocks relevant for astrobiology and resource exploration.

Robotic Field Test Details

To test this innovative concept, the quadrupedal robot 'ANYmal' was utilized. It was equipped with a robotic arm carrying two key instruments: the microscopic imager MICRO and a portable Raman spectrometer. This research was a collaborative effort involving institutions such as the Robotic Systems Lab at ETH Zurich, ETH Zurich | Space, the University of Zurich, and the University of Bern.

Experiments took place in the 'Marslabor' facility at the University of Basel, which simulates planetary surface conditions using analogue rocks, regolith materials, and controlled lighting. During the tests, the robot autonomously approached selected targets, deployed its instruments, and subsequently returned images and spectra for analysis.

The system successfully identified various rock types highly relevant to planetary exploration, including gypsum, carbonates, basalts, dunite, and anorthosite. Some of these, such as lunar-analog rocks like dunite and anorthosite, and oxides like rutile, may indicate valuable resources for future space missions.

Comparison of Exploration Approaches

The study directly compared two distinct operational methods: the traditional single-target exploration, which is closely guided by scientists for every step, and a semi-autonomous multi-target strategy, where the robot performs measurements at several locations consecutively.

The semi-autonomous missions consistently exhibited faster completion times.

Multi-target missions required between 12 and 23 minutes, while a human-guided mission took 41 minutes to complete comparable analyses.

Despite this increased speed, the robot achieved high scientific success rates. In one particular test run, all selected targets were correctly identified.

This approach could enable future missions to quickly survey large areas of planetary surfaces. Scientists could then analyze the incoming data and prioritize the most promising locations for detailed investigation. This method suggests that robots could traverse terrain, scan rocks, and collect data without requiring human command for every step, potentially accelerating scientific discovery on planetary surfaces by allowing rapid exploration and characterization of many rocks.

Future Exploration Preparation

The study indicates that relatively simple instruments can provide valuable scientific information when integrated into autonomous robotic systems. Instead of relying solely on extensive instrument suites, future missions could deploy agile robots that rapidly scan environments and identify promising targets for further investigation.

As space agencies plan upcoming missions to the Moon, Mars, and other destinations, such semi-autonomous systems could significantly assist scientists in surveying larger areas more efficiently.

This capability would support both crucial resource prospecting and the vital search for evidence of past life.