A depiction of a rover exploring an airless moon. An Earth like world rises over the horizon. Elements of this image furnished by NASA.

How Mars rovers use artificial intelligence

Posted on: September 14, 2023

In the past century, space exploration has been one of humanity’s most daring pursuits – and its expeditions to Mars have been among its most ambitious and high-profile missions.

Chief among these missions to the so-called red planet have been the Mars rovers expeditions. Rovers are vehicles that can traverse the surface of another planet with the objective of collecting data and images that are transmitted back to Earth for analysis, and they have played an invaluable role in enhancing human understanding of the Martian surface and its geological features.

Over the years, NASA (the National Aeronautics and Space Administration) and other agencies have pioneered the use of artificial intelligence (AI) in these missions, enabling rovers to navigate and explore autonomously, and even make crucial decisions without constant human intervention. 

Mars rovers: How they work and what they do

Mars exploration rovers are robotic, autonomous explorers that are transported to the planet through space. They are designed to travel along Mars’ surface and gather data to send back to Earth, and they are equipped with advanced instruments, such as spectrometers and cameras, to identify and study various geological features.

Each planetary rover is designed to withstand the harsh Martian environment, which includes dust storms and extreme temperatures, while carrying out its mission.

To date, there have been six successful Mars rover missions:

  • Sojourner (NASA)
  • Spirit (NASA)
  • Opportunity (NASA)
  • Curiosity (NASA)
  • Perseverance (NASA)
  • Zhurong (CNSA, the China National Space Administration).

What are the benefits of exploring Mars?

The exploration of Mars can help scientists better understand the planet’s history, geology, climate, and potential for life.

It can also offer valuable insight into wider questions, such as those around the formation and evolution of planets. For example, it’s believed that Mars – the planet most similar to Earth in our solar system – was previously home to water before its climate changed, so better understanding Mars can help us to better understand our own planet, and how it can evolve.

“Evidence suggests that Mars was once full of water, was warmer, and had a thicker atmosphere, offering a potentially habitable environment,” says the European Space Agency. “While life arose and evolved on Earth, Mars experienced serious climate change. Planetary geologists can study rocks, sediments and soils for clues to uncover the history of the surface. Scientists are interested in the history of water on Mars to understand how life could have survived.”

NASA, meanwhile, has four goals in exploring Mars:

  1. Determine if life ever arose on Mars
  2. Characterise the climate of Mars
  3. Characterise the geology of Mars
  4. Prepare for the human exploration of Mars.

How has AI been used in NASA’s Mars rover missions?

Artificial intelligence has the potential to revolutionise space exploration, and it’s already playing a pivotal role on Mars where it enables rovers to function autonomously, utilising sophisticated algorithms and machine learning techniques that allow them to make real-time decisions based on the datasets collected during their missions. 

The Sojourner rover

Sojourner, NASA’s first Mars rover, landed on Mars in 1997. While its AI capabilities were relatively basic compared to modern rovers, it laid the foundation for advancements in both artificial intelligence and automation in future missions.

According to NASA, Sojourner’s intelligence was based on a mathematical model that emulated animal behaviour in avoiding or fleeing danger:

“For the first time, a ‘thinking’ robot, equipped with sophisticated laser eyes and automated programming, is ‘thinking’ and reacting to unplanned events on the surface of another planet,” NASA stated in a 1997 press release

The Spirit and Opportunity rovers

Launched in 2003, Spirit and Opportunity were twin rovers tasked with exploring different regions of Mars. They were equipped with enhanced AI that allowed them to analyse images and autonomously plan their routes, avoiding obstacles as they traversed the Martian surface.

According to NASA, it was the Spirit and Opportunity rovers that found “dramatic evidence” that Mars used to be wetter, and that conditions on Mars could have sustained microbial life.

The Curiosity rover

Launched in 2011, the Curiosity rover, managed by NASA’s Jet Propulsion Laboratory (JPL-Caltech), marked a significant leap in rover artificial intelligence.

Curiosity’s AI system – called AEGIS, or Autonomous Exploration for Gathering Increased Science – uses a number of AI technologies, including deep learning, computer vision, and neural networks. These have enabled it to conduct research, choose some of its targets, and assess the safety of its surroundings with minimal human intervention. 

The Curiosity rover is still active on Mars today.

The Perseverance rover

NASA’s Perseverance rover arrived on Mars in 2021 at its Jezero Crater landing site. Like the Curiosity rover before it, it is still active today, and is armed with the AEGIS AI system that enables it to identify and investigate high-priority geological targets without direct human input.

It also features technology such as the Mars 2020 Perseverance navigation camera and hazard avoidance cameras that allow for autonomous driving.

What are the benefits of using AI for Mars rover missions?

The integration of artificial intelligence in Mars rover space missions offers several significant advantages, including:

  • Real-time decision-making. Mars rovers operate in environments where communication with Earth can experience delays. AI allows them to make critical decisions on their own and reduce mission delays.
  • Efficient data collection. Machine learning algorithms enable rovers to identify scientifically valuable targets on the ground, ensuring the collection of high-resolution data and enhancing the overall scientific output of the missions.
  • Autonomous navigation. AI-driven navigation systems mean that rovers can avoid hazards and select safe paths, reducing the risk of mission failure due to unforeseen obstacles.

These benefits have collectively enhanced humanity’s knowledge of Mars, enabling more extensive explorations and helping to investigate a broad range of geological features.

The difference between AI and robotics in Mars rover missions

Artificial intelligence and robotics work together on Mars rover missions, but they are distinct specialisms.

In the context of Mars rovers, robotics refers to the physical construction and mechanics of the rover, which will typically be equipped with robotic arms, wheels, and sensors that facilitate movement and data collection

AI, meanwhile, is what enables the rover to operate autonomously. It complements the physical robotic capabilities by providing the rover with a degree of intelligence, enabling the machine to operate – and even learn – as it moves through its environment and gathers information.

Launch your career in artificial intelligence

Join a field characterised by its exponential growth and importance to modern industry – and society – with the 100%-online MSc Computer Science with Artificial Intelligence at the University of Wolverhampton. This flexible Master’s degree has been developed for forward-thinking individuals who may not have a background in computer science.

Studying this course, you will develop a critical understanding of the processes, techniques, methodologies, and hands-on applications of artificial intelligence. Topics of study include:

  • machine learning
  • AI technologies
  • applications of AI
  • intelligent agents
  • data mining
No. 1 Ranked in the UK For Teaching first generation students
85% of research 'world-leading' or 'internationally important' (latest REF)
Online Psychology Master’s accredited by the British Psychological Society (BPS)
22,000 A university of 22,000 students
1827 Providing education and opportunity since 1827