Project 1.5: Planetary Surface Analyser (PSA)

An environmental perception framework that converts raw planetary data into formal engineering constraints for autonomous infrastructure planning.

Problem

Current off-world mission planning relies on static, manual interpretation of environmental data. This creates a dangerous gap between raw sensor imagery and the actual physics-based limits of colonial hardware.

The PSA matters because it automates the extraction of affordances—turning a map into a set of engineering boundaries. This prevents total system loss caused by environmental miscalculations in long-duration autonomous missions.

Solution

Acting as the Sensory Cortex of the STC, the PSA provides the validated environment models required for all downstream simulations.

  • Consumes: Topographic maps, multi-spectral imagery, and NASA PDS data.
  • Produces: Creates and modifies all_environments.yaml yaml files and schemas.
  • Interfaces: Feeds the Simulation Engine and MILP Optimizer.

Method

*In Development and Concept Stage of Planning*

    Tools & Technologies

    Python, PyYAML, GDAL (Geospatial Data Abstraction Library), NumPy, NASA Planetary Data System (PDS)

    Diagrams / Visuals

    [Architecture Diagram: Raw Data → Extraction Engine → STC Environment Schema + STC Environment yaml data]

    Results & Outcomes

    ⚠️ STATUS: IN DEVELOPMENT

    Currently, the framework successfully handles structural validation and manual environment selection. Phase 2 of the STC successfully shows that any environment that meets schema validation can be used for the system.

    Current Credibility: Schema and validation layers are fully implemented; simple solar flux logic has been implimented

    Next Steps

    • Implementation of computer vision to analyse datasets and provide geospatial data.
    • Create autonmated engine to generate environment and sub environment yaml.