Project Proposal
From Albatross
These are documents we submitted to the department while applying to do this project.
The goal of the project is to stabilize and control a small fixed wing aircraft through the design and implementation of an automatic pilot system.
This project has been under consideration for the later half of last year, and considerable research has been undertaken on our part, investigating possible system architectures and relevant control theory for unmanned aeronautic vehicles (UAVs).
Such UAVs find use in wide variety of fields, including environmental and meteorological monitoring, civil, military, border control, and search-and-rescue missions; and as a platform for further aeronautical research and development.
The system will consist of a single board computer, flying the plane in real time, in constant communication with a ground station for monitoring and command.
The single board computer (SBC) comprises a 400Mhz processor (Intel XScale), 400k gate FPGA, 64MB of ram, and 256MB of flash, and a number of additional useful features such as Ethernet interface and USB host; all in a small 70x40mm package.
The SBC will mount on a motherboard of our own design, containing primarily the sensors, but also additional necessary support circuitry such as the power supply, and communications. Also mounted on the motherboard will be an Ethernet, JTAG, and serial ports for debugging. The motherboard will also connect to a USB web cam, making possible machine vision applications.
Real time stabilization will be achieved with inertial sensors, namely the Analog Devices ‘MEMS’ gyroscopes and accelerometers. Pressure sensors will monitor barometric altitude and airspeed (with a Pitot tube). Higher level control such as path planning and navigation will achieved through the use of a GPS receiver and digital compass.
The airplane will communicate with the ground station over 867MHz license-free radio transceivers, which will send and receive command and status information.
While this may sound like an ambitious project we have both demonstrated to the department in the past that we are capable of such an undertaking, and through careful planning and research we are confident that the project will be completed successfully, and without compromising our other studies.
We have been fortunate enough to be offered the use of the aforementioned SBC for free, and also are confident that sponsorship may be obtained to support our studies.
There are several possibilities regarding the way this project may be split up between us, such as embedded software vs. control, navigation vs. stabilization, or sensory systems vs. control systems. We intend for these administrative details to be fleshed out at a later date in conjunction with (perhaps) another supervisor.
John Stowers, and Hugo Vincent
Individual Proposals
For the purposes of assessment, we had to split the project into two subprojects, and two budgets.
Hugo's Proposal
Systems Development for Unpiloted Aerial Vehicle (UAV) - Sensory Systems
The goal of the project is to stabilize and control a small fixed-wing aircraft, through the design and implementation of an automatic pilot and associated systems.
The system will consist of a single board computer (SBC) and associated hardware, responsible for flying the plane in real time, and a ground station for control and monitoring. The project has been split into two individual sub-projects, one concentrating on real time control (taken by John Stowers), and the other focusing on the sensory aspects (this project).
The Sensory Systems sub-project (this one) consists of data collection and data fusion, and involves aspects of hardware and FPGA design, algorithm design and analysis, and software development for use both on the ground, and aboard the plane.
Sensors will be included to measure vital flight parameters (position, attitude/heading, motion, etc.), along with appropriate signal conditioning circuitry. An estimate of the airplane’s state will be generated by combining data from the sensors with a state-space data fusion algorithm based around an extended Kalman filter.
In addition to data collection and processing, a system will be developed to allow bidirectional communication in real time between the aircraft and the ground, along with ground-based software for monitoring, visualization, and navigation.
John's Proposal
Systems Development for Unpiloted Aerial Vehicle (UAV) - Control Systems
The goal of the project is to stabilize and control a small fixed wing aircraft through the design and implementation of an automatic pilot system.
The system will consist of a single board computer (SBC) responsible for flying the plane in real time, and a ground station for control and monitoring. The project is to be split into two individual sub-projects, one concentrating on real time Control, and the other focusing on the Sensory aspects.
The Control Systems sub-project primary focus is real time aircraft control. This involves aspects of hardware design, software and simulation, on both the plane and ground station.
Different control algorithms will be investigated, and implemented to run in real time on the aircraft. Accompanying this, a flight simulator will be developed to allow systems testing and evaluation without full flight conditions.
In addition to real time control, support systems will be developed to aid in testing and debugging. These systems will include a system supervisor module, for resetting the SBC if it detects it has stopped functioning, and a data logging, error tracking, and testing infrastructure to speed development.
