Publications
Explicit efficient constrained model predictive control
2016 | International Journal of Automation and Control
A more efficient model predictive control algorithm is algebraic model predictive control. The efficiency is improved by forming the prediction horizon with a non-uniform distribution of points. The most efficient case reduces the problem down to a single prediction point, defining the horizon length as well as removing any redundant control calculations. This paper shows that for this single prediction point case, the constrained control can be easily evaluated explicitly, thus making the control calculation deterministic and more efficient. We also show that because an explicit control can be determined for the system, the closed-loop constrained stability bounds can be evaluated a priori. Using the key constraint of the peak location of the response also allows the system to meet the constraints while still being explicit and deterministic. The proposed explicit algorithm is applied to two numerical simulation examples with the performance, computational efficiency and stability analysed.
Airborne Vision-Aided Navigation Using Road Intersection Features
2014 | Journal of Intelligent & Robotic Systems
Modern airborne navigation systems for manned and unmanned platforms usually rely on GPS measurements to update the location estimate of the platform. This reliance on GPS unfortunately limits the precision of the navigational estimate to that of the GPS. When the navigation system is operating in GPS-limited or denied areas, the accuracy of the solution can quickly cause it to become unreliable. This paper presents a vision-aided inertial navigation system that uses ground features (in this case road intersections) matched to a database to provide position measurements. This removes the dependence on GPS. Flight test results of the system demonstrate that the system can successfully produce accurate navigation estimates that are comparable or better than the accuracy of GPS.
Efficient Terrain-Aided Visual Horizon Based Attitude Estimation and Localization
2014 | Journal of Intelligent & Robotic Systems
Inertial Navigation Systems typically rely on aiding-sensors such as GPS (Global Positioning System) to estimate the location of the system. The navigational performance of the Inertial Navigation System can be severely degraded when the GPS measurements are inaccurate or unavailable. Terrain-Aided Navigation is another method of localizing the platform by correlating the measured terrain information with a Digital Terrain Model. This paper presents an efficient Terrain-Aided Navigation method of generating position measurements from the visual appearance of the horizon profile (and hence terrain) surrounding the platform. An optimization process is used to align the measured horizon profile to an off-line pre-generated terrain-aided reference profile which allows for efficient position and attitude estimation. Numerical simulations are presented to evaluate the effectiveness of the proposed method. These results show that precise real-time attitude and position estimation is achievable using visual horizon profile information.
Efficient constrained model predictive control
2014 | European Journal of Control
Algebraic model predictive control is a more efficient predictive control algorithm in which the optimal control problem can be reduced down to a single prediction point, thus removing any additional and redundant calculations. The drawback of the reduction is lost system information and hence degraded performance while the system is constrained. This paper introduces a method of improving the constraint handling performance of the system by introducing additional points at which the constraints are checked. The computational efficiency is retained by recognising that the key constraint is the peak location of the response, with a method shown for calculating its location. The proposed method is demonstrated on a numerical simulation of a second order system and linear aircraft model with performance and computational efficiency compared.
Airborne Vision-Based Attitude Estimation and Localisation
2012 | University of Sydney Dissertation
The development and implementation of a robust vision-based system which fuses inertial information with visual information in a probabilistic framework with the aim of aircraft navigation.
Horizon Profile Detection for Attitude Determination
2012 | Journal of Intelligent & Robotic Systems
The horizon appearance is a strong visual indication of the attitude of an aircraft, so a vision based system should be able to detect the horizon and use its appearance to extract attitude measurements. Past methods have made the assumption that the horizon is straight, this neglects possible navigational and attitude information. This paper outlines a horizon detection method which allows for the actual horizon profile shape to be extracted. This horizon profile is then used for visual attitude determination. Test results for a captured flight video are presented and the proposed method is compared and evaluated against other methods.
Flight Control System Design: Learning Enhancement through Motion Based Flight Simulation
2010 | Australasian Association for Engineering Education
In these modern times of flight automation, students of aerospace engineering need to learn the importance of flight control system design and stability. However few have any experience of the operation and dynamic performance of these systems. It is difficult for students to connect design procedures used in computer based design exercises with the practical behaviour of a system in flight. Experiential laboratory exercises have been designed using a motion based flight simulator to assist students in learning the impact of their design decisions by experiencing their control system responses ‘in-flight’. This paper discusses the curricular aspects of flight control system design and the design of a simulation based experiential learning laboratory. It presents the outcomes of a questionnaire based learning assessment that shows substantial learning improvements are achieved. Student feedback indicates students draw great learning benefits through exposure to practical operation of flight control interfaces and consequent motion-based flight responses.
Enhancement of Learning in Aircraft Handling Qualities through Variable Stability Flight Simulation
2009 | Australasian Association for Engineering Education
There is an inherent need for aerospace engineering students to understand the meaning and consequences of flight stability and response to pilot control inputs. For most, limited flying experience makes learning this connection difficult. A full-motion flight simulator has been specifically developed to demonstrate a range of aircraft flight responses. Experiential learning exercises give first-hand experience of various important aerodynamic parameters used in aircraft design, and the effects that they have on an aircraft handling qualities. This paper discusses the facility, the background to the problem, and the form of the experiments performed. It presents the methods of analysis used to assess the effectiveness of the experiments and the facility. Learning outcomes from before and after knowledge analysis show a substantial improvement. Student feedback also shows great enthusiasm for this teaching system and indicates that students draw great learning benefits through experiential learning and their exposure to motion-based flight responses.
Vision Systems for Flight Guidance
2007 | University of Sydney Undergraduate Thesis
The research and development of horizon and runway detection algorithms for the purpose of extracting guidance information from the image features. This guidance information is then used in a terminal predictive control guidance system for automated visual only landing.
Patents
Automatic pitch mounting compensation in an automatic steering system
US20190299966A1. October 2019
Vehicle Implement Control
US20190124819 A1. October 2018
Predicting terrain traversability for a vehicle
US20190129435A1. October 2018
Three-Dimensional Terrain Mapping
US20190128690A1. October 2018
Low Cost Implement Positioning
US2018202804A1. July 2018
Vehicle Control Optimization
US20190129430A1. May 2018
Using Optical Sensors to Resolve Vehicle Heading Issues
US20180095476A1. April 2018