Part 2: Artificial Intelligence and NASA’s First Robotic Lunar Rover

In our last post, we described how VIPER, NASA’s first robotic Moon rover, is using artificial intelligence to create several options for the VIPER team to plan the rover’s path during its mission to the lunar South Pole.

Today, we’ll share more about how AI also is used to help human operators drive VIPER and create highly accurate maps of the rover’s mission area on the Moon.

Like a self-driving car, VIPER has cameras that monitor the environment around the rover and software that detects hazardous locations where it shouldn’t go. However, unlike self-driving cars, this software isn’t on board the rover; it’s back on Earth, and presents its conclusions to the rover drivers who use this information, along with many other sources, to decide how the rover should move.

One reason AI isn’t completely given the reins to the VIPER mission, is that AI techniques require a lot of training data – and this is the first time NASA will be remotely driving a robotic rover on the Moon. Using AI while always keeping humans in the loop provides a balance of risk and reward by using innovative and efficient techniques while avoiding unnecessary risk.

“VIPER is using AI as a tool; we’re not giving it the keys to the car,” said Mark Shirley, who created the original deterministic planner for VIPER at NASA’s Ames Research Center in California’s Silicon Valley. “And for this science mission, we don’t have to – the Moon is close enough that we can monitor these systems that are still learning this new environment and watch everything, like how you’d want to watch over a new driver.”

We don’t know everything about the environment of the Moon, but we do know a lot – and we can use AI to help us fill in the blanks.

Learning the Terrain

Planning routes and sensing hazards aren’t the only ways VIPER is using artificial intelligence. Other AI techniques are helping generate very high-resolution terrain maps. Most of our data about the Moon comes from LRO (Lunar Reconnaissance Orbiter), including several hundred photographs of VIPER’s mission area and topographical data obtained by shooting a laser down at the lunar surface and seeing how long it took to bounce back up.

A subfield of AI, called computer vision, can determine what the local slope is at each pixel using points of altitude, images, and our knowledge of the lunar environment, including how lunar regolith reflects light, where the Sun is in relation to the Moon, what direction the camera is facing, and how bright each pixel is.

All those slopes can be combined to create a terrain model that helps the VIPER team know the shape of the lunar surface. This shape can be used to calculate how the shadows move as the sun moves, and these moving shadows inform SHERPA’s – short for the System Health Enabled Real-time Planning Advisor – route planning. It is especially important to know how the shadows move because VIPER runs on solar power. Being stuck in a shadow for too long could be deadly for the rover.

All these pieces fit together. The high-resolution terrain maps created from LRO data generate maps of moving shadows, which SHERPA accounts for planning VIPER’s route. Temporal constraint techniques help mesh activities on the ground with activities on board the rover. Finally, the hazards pointed out automatically from the rover’s camera images help the VIPER team navigate the minute-to-minute decisions that come up while exploring another world.

As AI continues to develop as a field, many of its methods will end up becoming part of the regular toolkit for engineers and scientists. VIPER uses some of the current well-trodden techniques, while also pushing the boundaries of AI’s applications. In the case of SHERPA, the cutting-edge techniques come from a subfield of AI called decision making under uncertainty. This will be the first time these techniques are used on a space mission, and if successful, could open the door to similar AI approaches being deployed on other missions to worlds beyond our own.

Follow us @NASAAmes for more details about how artificial intelligence supports NASA’s VIPER mission and efforts to explore the unknown in space for the benefit of humanity.

Part 1: Artificial Intelligence and NASA’s First Robotic Lunar Rover

When NASA’s VIPER (short for Volatiles Investigating Polar Exploration Rover) lands on the surface of the Moon on a mission to better understand the environment where NASA plans to send astronauts as part of the increasingly complex Artemis missions, its journey will be guided by the human ingenuity of its human team – and several key tools that use artificial intelligence. From helping the science team choose a landing site at the lunar mountain Mons Mouton, to planning out its path, the VIPER team has developed and used artificial intelligence algorithms to help assess risk and optimize decision making.

Artificial intelligence is a wide field, and the resulting techniques are still far from the self-aware robots of science fiction. Instead, the field contributes tools to help space missions deal with some of the uncertainties that come with planning and executing a real-time mission in a challenging, largely unexplored environment.

“AI allows VIPER to be more adaptable, flexible, resilient, and efficient,” said Edward Balaban, VIPER’s lead for strategic planning at NASA’s Ames Research Center in California’s Silicon Valley. “It’s a tool that allows us to use change as a strength.”

These tools don’t replace human input – NASA scientists design these systems in the first place, input the relevant data, and then use the AI’s outputs as a baseline for mission-related decisions. During VIPER mission operations, the team plans to use AI interactively to help map out various routes for the operations team members to choose from. This AI system is called SHERPA – the System Health Enabled Real-time Planning Advisor.

An artist’s concept of the completed design of NASA’s Volatiles Investigating Polar Exploration Rover, or VIPER. VIPER will get a close-up view of the location and concentration of ice and other resources at the Moon’s South Pole, bringing us a significant step closer to NASA’s ultimate goal of a long-term presence on the Moon – making it possible to eventually explore Mars and beyond.
An artist’s concept of the completed design of NASA’s Volatiles Investigating Polar Exploration Rover, or VIPER. VIPER will get a close-up view of the location and concentration of ice and other resources at the Moon’s South Pole, bringing us a significant step closer to NASA’s ultimate goal of a long-term presence on the Moon – making it possible to eventually explore Mars and beyond.

Traversing the Lunar Surface

The VIPER mission will run for about 100 days after landing on Mons Mouton near the lunar South Pole. Throughout its journey, VIPER will make many stops at several science stations – sites selected for their potential to achieve the mission’s science objectives. These objectives include understanding the factors that control the distribution of water on the surface of the Moon, understanding the delivery history of water to the Moon, determining the origin of lunar water and other , and determining how volatiles evolve over time after they are deposited on the surface. How the rover moves from one of these sites to the other, and where it can find a safe place, referred to as a “safe haven,” to pause while temporarily out of communications with Earth — without getting stuck in an extremely cold and dark shadow — is a complex question requiring analysis of vast amounts of data. Factors such as the Moon’s rugged terrain, VIPER’s needs and limits, and the potential of the various science stations all need to be considered.

SHERPA is able to process all these factors and present the VIPER team with several options while planning the rover’s traverse before mission operations. It can assess the various risks of different routes by running thousands of mission simulations, and even provide contingency branches for where to go if something changes or doesn’t go according to plan. But after launch, SHERPA’s work won’t be over – it’ll also be used for real-time, dynamic problem solving, giving the VIPER team potential solutions to adjust the rover’s traverse when it’s presented with new scientific or operational information.

A traverse from SHERPA isn’t just a one-and-done plan. The AI will provide a template that humans consider and revise. Any changes made are then run back through SHERPA to determine if it’s feasible or if there are any issues. Those revisions won’t be corrections in the traditional sense or enacted by default, but allow team members to make adjustments based on factors the AI may not be able to consider, such as constraints related to staffing for the team members driving the rover or operating the rover science instruments.

Another set of techniques from a subfield of AI known as temporal constraint planning helps VIPER make its to-do list, by essentially presenting an algorithm with the problem of scheduling a set of activities within a certain time.

Follow us here or @NASAAmes for a follow-up post with more details about how artificial intelligence supports NASA’s VIPER mission and efforts to explore the unknown in space for the benefit of humanity.