Back to search

Georgia Institute of Technology

Hazard Detection and Avoidance for Autonomous Spacecraft Landing

Abstract

dc:description.abstract

As the feasibility of autonomous spacecraft landing is increasingly demonstrated by space agencies and private space companies, the demand for even more advanced capabilities grows, with the aim of achieving anytime and anywhere global and safe landing capabilities across the solar system—a milestone articulated by NASA’s Precision Landing and Hazard Avoidance (PL&HA) program. A notable challenge in this endeavor is the need for real-time terrain mapping and processing, along with the appropriate guidance, navigation, and control (GNC) technologies to perform autonomous hazard detection and avoidance (AHDA). This dissertation presents novel mathematical formulations and methods for the design and development of next-generation AHDA technology. Specifically, this study focuses on terrain mapping and processing algorithms for onboard hazard detection (HD) and the guidance problem under safe landing site uncertainty. We begin by improving the state-of-the-art (SOTA) model-based HD algorithm by applying deep learning techniques. Given the safety-critical application of planetary landings, an uncertainty-aware learning-based HD algorithm is developed for improved reliability. Next, in Chapters 3 and 4, we address the landing safety evaluation under topographic uncertainty. Topographic uncertainty is often inevitable due to the hardware limitations of sensing instruments and challenging operational conditions, but in-depth studies and effective algorithmic solutions remain limited. Chapter 3 investigates the uncertainty quantification of landing safety under topographic uncertainty and proposes a semi-analytical evaluation method. In Chapter 4, novel real-time stochastic terrain mapping and processing algorithms are developed. We develop a real-time Gaussian digital elevation map (DEM) construction algorithm and a real-time stochastic HD algorithm compatible with the Gaussian DEM input. The novel stochastic terrain mapping and processing algorithm is shown to outperform the SOTA algorithm in terms of both prediction reliability and computational cost. Finally, Chapter 5 formulates the hazard detection and avoidance (HDA) guidance problem and presents novel solution approaches. The HDA guidance problem is unique, where the objective is to maximize the probability of a safe landing, with the major uncertainty being the landing safety over the terrain. This dissertation presents several illustrative examples as proof-of-concept to demonstrate the value of the proposed work.

Degree

thesis:*
Level thesis:degree_level
Doctoral
Department dc:contributor.department
Aerospace Engineering
Grantor dc:publisher
Georgia Institute of Technology
Year dc:date.issued
2024

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Tomita, Kento
Advisor dc:contributor.advisor
  • Ho, Koki
Committee members dc:contributor.committeemember
  • Tsiotras, Panagiotis
  • Christian, John A.
  • Lightsey, Glenn
  • Horchler, Andrew D.

Subjects

dc:subject × 3

Rights

Language dc:language.iso
en_US

Identifiers

dc:identifier.*
Handle dc:identifier.uri
https://hdl.handle.net/1853/76872

Chain of custody

source
Harvested from
Georgia Tech
Base URL
repository.gatech.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Tomita, Kento. Hazard Detection and Avoidance for Autonomous Spacecraft Landing. Doctoral thesis, Georgia Institute of Technology, 2024. https://hdl.handle.net/1853/76872