Tool for Autonomous Terrain Exploration of Remote Space

NASA currently has multiple missions focused on exploration of extraterrestrial bodies in our solar system, both by autonomous platforms alone, in teams, and in collaboration with long-duration human habitation, all of which require advanced levels of autonomy. TATERS will develop mission planning and high level collaborative autonomy capabilities for swarms of surface and satellite vehicles that would allow NASA and other groups to be able to deploy autonomous teams of robots into previously unexplored spaces. TATERS Phase II will focus on maturation of the planning,analysis, and interface capabilities developed in Phase I for such missions. Phase I focused primarily on traversal analysis for individual and teams of robots in reference scenarios at the south pole of the moon that incorporated both scientific exploration and construction tasks. Phase II will expand and enhance the methods used in Phase I, increase fidelity of the testing environment by incorporating scientific data sources and models, integrate with existing NASA software products, and demonstrate transition feasibility through hardware-in-the-loop testing with UGVs. When planning complex missions with extended operation times, risk analysis requirements, expected autonomy requirements, and relatively high uncertainty of science objectives and mission execution, the ability to perturb any and all variables under consideration during planning is critical for identifying system failure modes. The traversal planning approach used by TATERS is novel in its ability to incorporate data across multiple resolutions, which enables pre-mission planning for high-level multi-agent coordination, rapid risk analysis of traversals through Monte Carlo testing, as well as integration into a navigation stack with real-time sensor stream and updates. Therefore, the TATERS approach demonstrates interoperability between human-in-the-loop navigation and control and autonomy stacks implemented on deployed platforms. Identification of Problem: • NASA's DRMs for the Moon, Mars, Ocean Worlds, and Small Bodies all require advanced levels of autonomy for small groups of robots (N<15) • Mission objectives are exploration, wide area sampling, and mapping of volatiles • A-priori information about target environments is limited and at coarse resolutions • Resource optimization is key for robotic systems • Communications to ground control are constrained and time delayed Significance of Innovation and Advancements: • Replace the human in the loop by adding high level autonomy to local search algorithms • Enabling of heterogenous teaming enabled by graph-based environment abstraction • Develop novel methods for dynamically updating nodes of belief roadmap (BRM) • Knowledge of how information compression affects graph based searching • Automatic selection of search algorithms based on local conditions • Autonomous goal selection and decentralized planning techniques Technical Objectives: 1. Demonstrate integrability of TATERS’ pre-mission planning and risk analysis methods into NASA’s simulation, mission planning, and systems operations ecosystem 2. Expand and enhance the mission planning, risk analysis, and demonstration capabilities of the TATERS toolsuite, including DRM-specific demonstrations, incorporation of mission-relevant datasets, and logistical constraints, and 3. Demonstrate hardware-in-the-loop integration with real-time local graph generation, integration, and communications to a human-in-the-loop operator. Deliverables: • Software modules for mission planning and robotic operations: multi-dimensional risk assessment, inter-agent communication, multi-dimensional path planning, graph merging and generation. • Software Enhancements to NASA simulation and testing tools: GIS raster alignment to mesh geometry, planning and interface GUI, parametric control of Monte Carlo simulations, automated report generation and analysis.
Benefits: TATERS focuses on transitioning technologies into the Moon to Mars mission such as coordination of in-situ autonomous robotic platforms and vehicles, the transition path into mission planning and autonomy stacks for platforms used in scientific, exploration, and site construction mission, and human mission support tasks. The collaborative autonomy enabled by TATERS is a necessary step towards development of vanguard robotic teams that can proceed ahead of human teams in missions in extreme environments. TATERS can be used in terrestrial environments, by enabling collaborative autonomy of field robotics in unstructured environments, such as firefighting, disaster response, maritime, and defense applications. The graph representation used to store environmental information was developed to be easily stored, modified, and shared, making them well-suited products for distributed low-SWAP systems.