NASA has taken a significant step in its Mars exploration efforts by utilizing artificial intelligence (AI) to plot a course for its Perseverance rover. The mission marked the first time a large language model, Claude, was used to pilot a car-sized robot on the Martian surface.
The Perseverance rover, which landed on Mars in 2021, has been making history with its numerous milestones, including sending back audio recordings from the planet. This latest feat sees the rover successfully navigating approximately 400 meters of terrain through a field of rocks mapped out by Claude between December 8 and 10.
The use of AI to plot the course was not without its challenges, however. NASA explained that routing the Perseverance rover requires meticulous planning to avoid accidents, such as the rover sliding or getting stuck. To achieve this, human operators painstakingly laid out a "breadcrumb trail" for the rover to follow, using a combination of images taken from space and onboard cameras.
To enable Claude to complete the task, NASA provided the AI model with extensive contextual data from the rover's years of operation. The model then methodically stringed together waypoints, which were later critiqued and refined by NASA engineers at the Jet Propulsion Laboratory (JPL).
Before deploying Claude's route, the JPL team ran a simulation to confirm the accuracy of the commands sent to the rover. As it turned out, only minor changes were required, with one tweak resulting from access to ground-level images that Claude hadn't seen during its planning process.
The success of this collaboration has significant implications for NASA's future Mars missions. By leveraging AI-powered autonomous systems, the agency hopes to increase the efficiency and consistency of its rover operations. According to NASA, using AI will cut route-planning time in half, allowing the rover's operators to fit in more drives, collect more scientific data, and conduct more analysis.
This marks a major milestone for Anthropic, the company behind Claude, which has come a long way since struggling to navigate a simple 8-bit Game Boy game. The success of this collaboration opens up new possibilities for future collaborations between NASA and AI-powered systems.
As NASA looks ahead to its ambitious plans, including returning to the Moon with reduced workforce, any tool that can enhance efficiency is welcome. With AI playing an increasingly crucial role in space exploration, it's clear that the agency will continue to harness the power of artificial intelligence to unlock more secrets about our solar system.
The Perseverance rover, which landed on Mars in 2021, has been making history with its numerous milestones, including sending back audio recordings from the planet. This latest feat sees the rover successfully navigating approximately 400 meters of terrain through a field of rocks mapped out by Claude between December 8 and 10.
The use of AI to plot the course was not without its challenges, however. NASA explained that routing the Perseverance rover requires meticulous planning to avoid accidents, such as the rover sliding or getting stuck. To achieve this, human operators painstakingly laid out a "breadcrumb trail" for the rover to follow, using a combination of images taken from space and onboard cameras.
To enable Claude to complete the task, NASA provided the AI model with extensive contextual data from the rover's years of operation. The model then methodically stringed together waypoints, which were later critiqued and refined by NASA engineers at the Jet Propulsion Laboratory (JPL).
Before deploying Claude's route, the JPL team ran a simulation to confirm the accuracy of the commands sent to the rover. As it turned out, only minor changes were required, with one tweak resulting from access to ground-level images that Claude hadn't seen during its planning process.
The success of this collaboration has significant implications for NASA's future Mars missions. By leveraging AI-powered autonomous systems, the agency hopes to increase the efficiency and consistency of its rover operations. According to NASA, using AI will cut route-planning time in half, allowing the rover's operators to fit in more drives, collect more scientific data, and conduct more analysis.
This marks a major milestone for Anthropic, the company behind Claude, which has come a long way since struggling to navigate a simple 8-bit Game Boy game. The success of this collaboration opens up new possibilities for future collaborations between NASA and AI-powered systems.
As NASA looks ahead to its ambitious plans, including returning to the Moon with reduced workforce, any tool that can enhance efficiency is welcome. With AI playing an increasingly crucial role in space exploration, it's clear that the agency will continue to harness the power of artificial intelligence to unlock more secrets about our solar system.