Carnegie Mellon University
Browse

Planning and Incremental Modeling for Autonomous Microrover Exploration of Planetary Pits

Download (49.22 MB)
thesis
posted on 2023-07-20, 19:51 authored by Jordan FordJordan Ford

Planetary pits are unique portals to the subsurface of the Moon, Mars, and other solid-surface bodies. They provide opportunities for naturally protected underground habitats, and they harbor unique insights into the geologic processes that formed the Solar System. Though orbiters have revealed hundreds of pits, orbital cameras cannot acquire the close-range, high-resolution, long-exposure imagery needed to further our scientific understanding of planetary pits. Near-term surface rover exploration is needed to search pit floors for habitable alcoves and study the geologic timeline exposed by pit walls.

Fast, solar, autonomous microrovers could explore pit rims, image pit interiors, and return detailed 3D models to Earth. From orders-of-magnitude closer range, surface rovers could capture the extreme high-resolution, long-exposure imagery needed to study pit walls and search pit floors for habitable alcoves. This thesis advances key roving, mapping, planning, and incremental modeling technologies needed for autonomous rovers to circumnavigate immense planetary pits, construct accurate, high-resolution three-dimensional models, and return the models to Earth for study.

There are four principal contributions of this research. The first is a novel autonomous microrover with the sensing, computing, and mobility necessary for pit exploration. The second is an exploration planning algorithm that selects favorable viewpoints on a pit rim and plans exploration routes that maximize overlapping view coverage while satisfying the mobility and communications constraints of small rovers. The third contribution of this research is an incremental pit modeling method that constructs and efficiently improves full-coverage, accurate, high-resolution, high-dynamic range three-dimensional models of planetary pits. The final contribution is a comprehensive analysis of known lunar pits and their amenability to small-rover exploration and modeling. Together, these technologies are developed, integrated, and evaluated for efficiency, coverage, and accuracy around a terrestrial analog pit. The work developed in this thesis enables robust autonomous exploration of planetary pits with immense science and exploration return.

History

Date

2023-05-01

Degree Type

  • Dissertation

Department

  • Electrical and Computer Engineering

Degree Name

  • Doctor of Philosophy (PhD)

Advisor(s)

William L. Whittaker

Usage metrics

    Licence

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC