Senior GNC Engineer, Anduril · Sydney

Dr. Steven Dumble

Aerospace Engineer & Software Developer

Let's Do Some Awesome Stuff.

Dr. Steven Dumble
Dr. Steven Dumble
PhD, BEng(Aero)(Hons)

About Me

Dr. Steven Dumble is a Senior Guidance, Navigation and Control Engineer at Anduril. An experienced PhD GNC engineer with over 15 years of experience, specialising in vision-aided inertial navigation systems as well as fixed-wing autopilot design and implementation.

Steven holds a PhD and BEng(Aero)(Hons) from the University of Sydney, where his doctoral research focused on developing robust vision-based navigation systems that fuse inertial information with visual cues for aircraft navigation in GPS-denied environments.

Beyond research, Steven is passionate about engineering education and has developed theoretical and practical resources for GNC engineers through gnctoolbox.com and a suite of popular online courses.

15+
Years Experience
9
Publications
7
Patents
3
Online Courses
Education
PhD Aerospace Engineering
University of Sydney, 2008–2012
BEng (Aero) Honours
University of Sydney

Experience

Senior GNC Engineer

Current
Anduril
Feb 2026 – Present · Sydney, Australia

Guidance, Navigation and Control engineering for advanced autonomous systems.

Lead GNC Engineer

PhantomWorks Global — Boeing
Jan 2021 – Feb 2026 · Australia

Guidance, Navigation and Control Team Lead for advanced defence programmes.

Software Engineer

PhantomWorks Global — Boeing
Nov 2019 – Jan 2021 · Brisbane, Australia

Autonomous teaming behaviour development for next-generation aerial platforms.

Senior Control Engineer

AgJunction
Dec 2014 – Nov 2019 · Brisbane, Australia

Designed and implemented control systems for autonomous agricultural platforms. Contributed to multiple patented technologies.

Maritime Strike Weapon Model Developer

YTEK Pty Ltd
Oct 2013 – Dec 2014 · Adelaide, Australia

Development and simulation of maritime strike weapon system models.

PhD Researcher & Academic

University of Sydney
Mar 2008 – Nov 2012 · Sydney, Australia

Research into vision-based navigation, horizon detection, and terrain-aided attitude determination. Teaching aerospace engineering students.

Projects

C++ Real-Time Systems Multi-threading Sensor Integration

Flight Data Acquisition System

University of Sydney — Jabiru Flight Test Laboratory

A full flight data and video real-time data acquisition system developed in C++ for the University of Sydney Jabiru Flight Test Laboratory Program. The FDAS records and processes inertial, air data, magnetometer, GPS, control, and camera measurements in real-time across a distributed computing system.

  • NovAtel ProPak-G2plus SPAN with HG1700 Tactical-Grade IMU at 100 Hz
  • 7 cameras: 5× Unibrain Fire-i Firewire (30 Hz) + 2× Prosilica GigE (25 Hz)
  • USB driver optimisations to achieve >500 Hz throughput
  • Double-buffered multi-threaded disk streaming for real-time data throughput
Computer Vision Navigation MATLAB C++

Horizon Detection for Attitude Determination

PhD Research — University of Sydney

Developed a real-time capable horizon detection method capable of extracting the actual horizon profile shape — going beyond the straight-line assumption used in prior methods. The horizon profile is then used for terrain-aided attitude determination and localisation, producing measurements at least an order of magnitude more precise than conventional horizon-line methods.

  • Terrain-aided attitude determination accurate to full attitude triplet (bank, pitch, yaw)
  • Efficient horizon profile generation from digital terrain maps
  • Real-time feasible position and attitude estimation via profile matching
  • Vision-based navigation independent of GPS
Navigation SLAM Computer Vision Sensor Fusion

Airborne Vision-Aided Navigation

PhD Research — University of Sydney

A completely passive, independent vision-based navigation system for aerial environments. The system combines SLAM for short-term precision, terrain-aided navigation using human-recognisable features (road intersections, rivers), and visual horizon-based attitude determination — mimicking VFR pilot navigation without reliance on GPS.

  • Road intersection detection algorithm validated in flight tests
  • Navigation accuracy comparable to or better than GPS
  • Feature-based TAN using GIS-mapped landmarks
  • Probabilistic sensor fusion of visual and inertial measurements

Publications & Patents

View on Google Scholar

Journal Papers & Theses

Explicit efficient constrained model predictive control

2016
International Journal of Automation and Control

Presents an explicit, deterministic constrained MPC algorithm based on a single prediction point, reducing computational overhead while maintaining closed-loop stability guarantees.

Airborne Vision-Aided Navigation Using Road Intersection Features

2014
Journal of Intelligent & Robotic Systems

Vision-aided inertial navigation system using ground road intersections matched to a GIS database as position measurements, removing dependence on GPS. Validated in flight tests.

Efficient Terrain-Aided Visual Horizon Based Attitude Estimation and Localization

2014
Journal of Intelligent & Robotic Systems

Efficient terrain-aided navigation method generating position measurements from the visual appearance of the horizon profile, enabling real-time attitude and position estimation.

Efficient constrained model predictive control

2014
European Journal of Control

Improves constraint handling in algebraic MPC by introducing additional constraint-check points at the peak response location, retaining computational efficiency.

Airborne Vision-Based Attitude Estimation and Localisation

2012
University of Sydney Dissertation

Doctoral thesis developing and implementing a robust vision-based system fusing inertial information with visual data in a probabilistic framework for aircraft navigation.

Horizon Profile Detection for Attitude Determination

2012
Journal of Intelligent & Robotic Systems

Horizon detection method that extracts the actual horizon profile shape (beyond the straight-line assumption) for visual attitude determination.

Flight Control System Design: Learning Enhancement through Motion Based Flight Simulation

2010
Australasian Association for Engineering Education

Experiential lab exercises using a motion-based flight simulator to improve student understanding of flight control system design. Demonstrated substantial learning improvements.

Enhancement of Learning in Aircraft Handling Qualities through Variable Stability Flight Simulation

2009
Australasian Association for Engineering Education

Full-motion flight simulator demonstrating a range of aircraft flight responses to teach aerospace students the connection between aerodynamic parameters and handling qualities.

Vision Systems for Flight Guidance

2007
University of Sydney Undergraduate Thesis

Research and development of horizon and runway detection algorithms for extracting guidance information used in a terminal predictive control guidance system for automated visual landing.

Patents

US20190299966A1

Automatic pitch mounting compensation in an automatic steering system

October 2019
US20190124819A1

Vehicle Implement Control

October 2018
US20190129435A1

Predicting terrain traversability for a vehicle

October 2018
US20190128690A1

Three-Dimensional Terrain Mapping

October 2018
US2018202804A1

Low Cost Implement Positioning

July 2018
US20190129430A1

Vehicle Control Optimization

May 2018
US20180095476A1

Using Optical Sensors to Resolve Vehicle Heading Issues

April 2018