Lunar Resources: Geosensing for Autonomous Exploration

Francis, B. (2022) Lunar Resources: Geosensing for Autonomous Exploration. [USQ Project]

Text (Project)

Download (5MB) | Preview


In 2017, National Aeronautics and Space Administration (NASA) announced the Artemis Mission, setting an ambitious goal for mankind to set foot on the South Pole of the Moon by 2025, for the first time in more than 50 years. One of the largest milestones, is the development for technologies for in-situ resource utilisation (ISRU) which harnesses in-place resources through various hardware and operations, increasing mission performance and sustainability, while decreasing lifecycle costs and risks.

There are several steps in the ISRU process, including resource assessment, acquisition, and processing. On Earth, resource assessment would generally be completed in the field by Geologists, creating detailed maps of the geology and the geomorphology. Given the harsh nature of the Lunar surface, a robotic platform is ideal to perform this task. The generated maps could suggest destinations for further human exploration, offer insights into the geology, or define unsafe areas. An inherent problem is working with the intensity and low level of the sun experienced at the Lunar Poles and developing methods for identification of the size, colour, or shape of objects that may be in permanently shadowed regions (PSR), ares on the surface that sunlight never reaches. Key features of the Lunar surface have been determined, to ascertain the features in scenes the system should be able to detect, now and in the future.

Here, we evaluate a novel, energy efficient approach that makes use of reflected light sources to enhance the generation of point clouds from a stereo camera system, when viewing a partially shadowed scene. The scenes used for evaluation are a “natural scene”, mimicking a crater on the Lunar surface, and a “comparison scene”, where cubes (with a known side length) are placed on a flat surface in an otherwise sparse location. The recorded point clouds and camera images are then evaluated alongside those from an Intel RealSense D435i depth camera, to critically characterise and compare the accuracy of reproduction of the evaluation scenes with respect to identification of size, colour and shape, to determine the enhancement this passive approach offers against an active approach. Additional considerations were made to the evaluation of the systems’ suitability for a micro rover platform, by analysing the size, weight and power consumption of the prototype sensor.

Results show that the introduction of reflected light on a scene significantly improves the colour reproduction of the system, which in turn can improve the accuracy of stereo matching algorithms. There were however complications with the implementation of the stereo matching algorithm used for generation of disparity images and in turn point clouds, which affected the ability to fully evaluate the system’s potential at determining the size and shape of objects. Evaluation of the size, weight and power consumption concluded that the size and weight of the system is feasible at this early stage, however significant improvements will have to be made to the power consumption for it to be viable.

Informing future development of the robotic platform, it was concluded that although reflected light has the potential to improve accurate detection of colour, shape and size of objects, for the system to function the reflector has to be correctly orientated to the Sun, which reduces the practicality of the proposed solution. As it was demonstrated to improve colour reproduction, it was concluded that the approach would be more suitable when used alongside other sensing technologies, in order to extend mission runtime and performance.

Statistics for USQ ePrint 51868
Statistics for this ePrint Item
Item Type: USQ Project
Item Status: Live Archive
Faculty/School / Institute/Centre: Current – Faculty of Health, Engineering and Sciences - School of Engineering (1 Jan 2022 -)
Supervisors: Lobsey, Craig; Hodgkinson, Jane
Qualification: Bachelor of Engineering (Electrical & Electronic)
Date Deposited: 19 Jun 2023 02:20
Last Modified: 20 Jun 2023 01:07
Uncontrolled Keywords: lunar resources; geosensing

Actions (login required)

View Item Archive Repository Staff Only