Scientific Rationale

Autonomous lunar surface navigation is the ability for assets on the Moon to remotely travel long distances. This is particularly important on the lunar surface because of the limited capacity to send human operators and the propagation delay for Earth-to-Moon communication. Autonomous lunar surface navigation is both an enhancing and an enabling technology. It is enhancing for short duration lunar missions that require relatively little autonomous transport. However, it is enabling for long-term, sustained human operations on the Moon, as operations will need to continue regardless of in-the-moment human presence. As a majority of long-term infrastructure on the Moon will likely be robotic, the ability for robotics to navigate autonomously will allow scaling of lunar infrastructure without perpetual teleoperation or manual operation.

This area has a couple of challenges, namely in technological development. The first challenge is that this SKG requires the development of a couple of interconnected pieces of technology: sliding autonomy and localized hazard avoidance. While none of these technologies are new, rapid implementation and testing on the Moon could pose some challenges. In order to overcome this challenge, it will be important to address the following questions:

1. What range of autonomy will lunar robots need to accomplish?

2. What types of hazards will robots need to circumnavigate on the lunar surface and what are the tolerances for error?

3. How much distance will robots need to travel in how much time?

Additionally, a lunar GPS constellation may need to be deployed around the Moon to support autonomous navigation, which could incur additional costs and considerations. In order to understand the requirements of lunar GPS, the following questions will need to be addressed:

1. What resolution of GPS will be needed for lunar robots to autonomously perform tasks?

2. What areas of the Moon will require GPS coverage?

3. For what length of time will a robot need to be connected to lunar GPS?

Technology Analysis

Telerobotics

In 2013, NASA’s Intelligent Robotics Group completed a couple of surface telerobotics test sessions where an astronaut onboard the ISS successfully operated a K10 rover located at NASA Ames. This mission was meant to simulate the deployment of a radio antenna on the lunar surface. During the 100-minute real-time teleoperation mission, data on operator workload and situation awareness, data communications link utilization, robot user interface, and robot telemetry was collected. As this mission was primarily to test teleoperation capabilities, the design was not entirely applicable to the lunar environment. This test could serve as an important data point for the development of sliding autonomy.

Autonomous Navigation on MERs

Perhaps the best recent examples of autonomous navigation on an extra-Earth body are NASA’s Mars Exploration Rovers (MERs). MERs have three basic mobility commands: a) low-level motor control commands, b) directed driving along arcs, and c) autonomous path selection with a given goal. The first two modes allow operators a higher degree of robotic control. For example, directed driving is blind to all obstacles and does not actively avoid hazards.

Mission requirements for MERs autonomous navigation is the ability for MERs to safely drive tens of meters autonomously per day. For autonomous path selection, rovers are equipped stereo cameras, vision processing, and terrain traversability assessment software so that the rover can identify and avoid hazards. Stereo vision was specifically chosen for MERs because of its low power requirements and no moving parts. It is likely that lunar robots would use a similar technology. Stereo vision technology was developed at JPL and has been implemented on several robotic systems so is very well understood and tested. Selection of camera resolution must be tuned for the environment the rover will land in.

MERs’ autonomous navigation starts with terrain assessment and sensing the shape of the terrain. For example, Opportunity can measure 48,000 XYZ points. Geometric data is then processed by GESTALT, which creates a grid-based localized traversability map with a grid size of 20 cm. At this point, possible rover paths are projected onto the traversability map and then scored on that paths ability to navigate to the goal point. Path-based safety evaluations, goal location, and current steering direction are then used to determine optimal pathing. MERs AutoNav also has a several other features like Pose Update and Visual Odometry that enable safe and accurate navigation.

One outstanding inquiry for autonomous lunar navigation is about distance and processing speed. Due to bias towards safety and limitations in computing power, Perseverance can drive just 200 meters per day. Future lunar missions could require longer travel distances, so technological upgrades could be required.

Another outstanding inquiry for autonomous lunar navigation is accuracy. Spirit and Opportunity mission requirements were to maintain estimated position knowledge within an accuracy of 10% of integrated distance traveled. It is likely that future lunar operations will require a higher level of accuracy, which could also require technological upgrades, perhaps in the form of a lunar GPS.

Lunar GPS

Autonomous navigation could be aided by a lunar GPS. One proposal for lunar GPS is to use existing GPS satellites that orbit Earth. Upgrades to GPS equipment and a powerful receiving antenna could be required, but simulations show that this version of lunar GPS is feasible. The first demonstration of lunar GPS will come with Artemis-1 in late 2021. Orion’s GPS receivers will give a good indication as to the strength of GPS signals near the Moon. Whether or not ground based lunar rovers will be able to use these same GPS signals is uncertain.

Another proposal is to deploy a constellation around the Moon to serve networking and navigational use cases. LunaNet intends to use SmallSats to establish an infrastructure layer for communication between the Earth, the Moon, and any end users that might be on the lunar surface. Links between relay satellite and relay service users would be done in S-band, Ka-band, and optical.

Technology Readiness Level Assessment (as of 2015)

• Technology: Three-Dimensional (3D) Range Imaging Sensors for Above-Surface Mobility

  • Establish Figure of Merit: TRL 4

  • Target: TRL 6

• Technology: In-Situ Camera Geometric Calibration Diagnostics and Self-Calibration

  • Establish Figure of Merit: TRL 4

  • Target: TRL 6

• Technology: Vision-Based Aiding of Dead Reckoning for Navigation of Surface Vehicles

  • Establish Figure of Merit: TRL 4

  • Target: TRL 6

• Technology: Map-Based Position Estimation for Navigation of Surface Vehicles

  • Establish Figure of Merit: TRL 4

  • Target: TRL 6

• Technology Terrain Mapping for Surface Vehicles

  • Establish Figure of Merit: TRL 4

  • Target: TRL 6

• Technology: Landmark Mapping from Image Sequences and Other Navigation Data

  • Establish Figure of Merit: TRL 4

  • Target: TRL 6

• Technology: Adaptive Autonomous Surface Navigation

  • Establish Figure of Merit: TRL 6

  • Target: TRL 6

• Technology: Direct Teleoperation

  • Establish Figure of Merit: TRL 4

  • Target: TRL 8

• Technology: Supervisory Control

  • Establish Figure of Merit: TRL 5

  • Target: TRL 9

Works Cited

Hadhazy, Adam. Cosmic Gps. 28 Apr. 2020, aerospaceamerica.aiaa.org/features/cosmic-gps/.

Israel, David, Kendall Mauldin La Vida Cooper, and Katherine Schauer. "LunaNet: A Flexible and Extensible Lunar Exploration Communications and Navigation Infrastructure and the Inclusion of SmallSat Platforms." (2020).

Maimone, Mark, et al. “Surface Navigation and Mobility Intelligence on the Mars Exploration Rovers.” Intelligence for Space Robotics, TSI Press, 2006, pp. 45–69.

“NASA Technology Roadmaps - TA 4: Robotics and Autonomous Systems.” NASA, National Aeronautics and Space Administration, May 2015, www.nasa.gov/sites/default/files/atoms/files/2015_nasa_technology_roadmaps_ta_4_robotics_autonomous_systems.pdf.

“News - Surface Telerobotics Team Successfully Conducts Second K10 Rover Test with the International Space Station.” NASA, NASA, ti.arc.nasa.gov/news/iss-telerobo-test2/.

Oberhaus, Daniel. “How NASA Built a Self-Driving Car for Its Next Mars Mission.” Wired, Conde Nast, 21 July 2020, www.wired.com/story/how-nasa-built-a-self-driving-car-for-its-next-mars-mission/.