Description: Multi-agent autonomy will be a key tool for future landed planetary exploration missions to maximize resource utilization to better enable systems and instruments operating within extreme mass, volume, time, and power constraints, while minimizing programmatic and technical risk. This includes autonomous robotic operations of navigation, communication, computation, perception, and decision-making. This increased capability enables individual robot autonomy and multi agent autonomy that allows the robots to collaborate as a team to accomplish system objectives. This platform capability will enable future exploration and science measurements using distributed instruments in better context with expected geo-spatial diversity and deployment precision. Cooperative Autonomous Distributed Robotic Exploration (CADRE) expands upon the capability to explore planetary bodies through autonomous robot team operations. The CADRE project will design, develop, and demonstrate (first on the earth and then on the lunar surface) a network of mobile robots that can autonomously navigate and survey the daytime Lunar environment to accomplish a shared goal of performing a distributed measurement. The system of solar powered rovers and single base station will autonomously plan (and replan, if necessary), schedule, and explore the lunar surface to generate a navigable map of a [20] meter x [20] meter area. Using that map the system will autonomously plan, schedule, and traverse the lunar surface in a line formation with [1] meter spacing for a distance of [20] meters taking a multi-static measurement every [0.2] meters with the Ground Penetrating Radar (GPR) on all rovers. To demonstrate resiliency to planned and unplanned attrition, autonomy will be turned off on the lead agent of the system forcing the re-election of a new leader.

Benefits: CADRE will demonstrate cooperative autonomous exploration by navigating, communicating, computing, perceiving, and decision-making without human interaction.Cooperation (error reduction in autonomous mapping due to cooperation) Given the same area, same algorithms, same sensors, a team of rovers produces a data product with less error than a single rover integrating the same set of measurements.Scalability (scalability of autonomous cooperation) The multi-agent autonomy algorithms are capable of coordinating up to N agents without modification (including change of compute or hardware).Resiliency (the resiliency of autonomous cooperation to attrition) The multi-agent autonomous system is made up of N agents and is capable of sustaining attrition up to some % and still achieve its goal.Human-robot Teaming (the % increase in operation time when increasing the number of autonomous agents) The cost (time) of operating autonomy required to achieve a specific task grows by a sublinear factor when adding more agents to the system. The lunar surface demo advances the technology from A-PUFFER (and Ingenuity) and demonstrates capability in hazardous and unknown terrain conditions.