Range-based relative navigation for femto-spacecraft swarms

Timmons, Thomas (2024) Range-based relative navigation for femto-spacecraft swarms. PhD thesis, University of Glasgow.

Abstract

Femto-spacecraft, characterised by a total mass of under 100 g, are the smallest category of active spacecraft that are currently classified. If this technology is scaled to use many devices for distributed sensing applications, there can be unique functional benefits over traditional space systems. Determining the location of femto-spacecraft within a networked swarm would add utility for scientific investigation and enable in-orbit navigation without relying on Earthbased tracking or on-board GPS. With extremely limited space and computing power on such platforms, this thesis proposes methods for relative navigation enabled by range estimates obtained via networking. This is experimentally demonstrated using received signal strength indication (RSSI) data as a coarse range metric within a highly interconnected network of devices.

The first part of this thesis is the development of a series of range-based relative positioning algorithms intended for centralised, decentralised, and distributed computation. These are tested in simulation by modelling varying levels of inaccuracy in the range estimates provided. Relative navigation techniques are then presented, modelling the scenario of a swarm of femtospacecraft deployed into low Earth orbit from a larger carrier spacecraft. Several ejection strategies are proposed using a model of the relative dynamics to propagate trajectories. These scenarios are simulated for real-time relative navigation using Kalman filtering techniques that use the relative positioning algorithms as partial linear state observers.

A network communications protocol enabling full interconnection within the swarm is then developed for experimental testing of these navigation strategies. Small development kits are used as femto-spacecraft proxies for embedded software development. Following this, a path loss model is experimentally developed to characterise the relationship between RSSI and range in an outdoor testing environment over a length-scale of 1-150 m. This model is then used to convert RSSI data into range estimates, used as inputs to the relative positioning algorithms. Relative navigation is demonstrated on an outdoor 120 m × 60 m sports pitch. This demonstration highlights the ability of the algorithms to fuse coarse proximity data and localise without any additional sensors or equipment

Finally, the application of distributed in-situ sensing is analysed in simulation to investigate the trade-offs of using a larger number of devices of lower positioning accuracy and sensor measurement accuracy, as a comparison between the use of femto-spacecraft swarms and traditional space systems.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Subjects: T Technology > T Technology (General); T Technology > TJ Mechanical engineering and machinery
• Colleges/Schools: College of Science and Engineering > School of Engineering
• Supervisor's Name: McInnes, Professor Colin
• Date of Award: 2024
• Depositing User: Theses Team
• Unique ID: glathesis:2024-84281
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 30 Apr 2024 15:29
• Last Modified: 14 May 2024 08:38
• Thesis DOI: 10.5525/gla.thesis.84281
• URI: https://theses.gla.ac.uk/id/eprint/84281

Related URLs

• Article DOI
• Enlighten Publications Record
• Enlighten Publications Record