AUTONOMOUS
DRONE FLIGHT FOR PRACTICAL SOLUTIONS
HOW IT WORKS
01
PREPARE
Download our public source code along with Unity. Our code works with the Bitcraze CrazyFlie 2.0 Drones with an attached Flow Deck Module.
02
PLOT A COURSE
Within Unity a user can plot in various points with an x,y,z coordinates. The user can also enter in a time variable which will tell the drone how long to stay at a given location. The source code can also be entered in manually and run through Python.
03
FLY
Once your radio controller is connected, your CrazyFlie drone is properly configured and you have downloaded and installed all required software, simply run the software within unity and the drone will take off from it's origin to the specified positions.
LEARN MORE ABOUT OUR PROJECT
Deakin Symbiosis
MEET THE TEAM
Paul Hammond
Joshua Pujol
Thomas Guilfoyle
Justin Taylor
Wai Kit Lam
Rui Li
Sherry He
ABOUT SYMBIOSIS
OUR STORY
Comprising of seven randomly selected Deakin University Undergraduate school of IT students, the Symbiosis project aims to deliver upon our clients needs, an cutting-edge autonomous drone program that utilises various forms of software, and that has the ability to be tracked on the fly by motion capture cameras. Not much was known about the nature of the project by any of the members when we embarked upon this project, however each member conducted copious amounts of research in order to fully understand the details of how we would deliver such as vision. Through perseverance the group managed to deliver a working solution to the client by the end of the year that was well received by our client.
OUR VISION
The Symbiosis group was successful in deploying a working solution and we are excited to see where this project can be taken in terms of expanded functionality, this is why we have chosen to have our source-code open and free for the public to use, as-well as any other Deakin students embarking upon a similar project in the future. We believe what we have achieved lays a great foundation to expand the capability of the project to new heights.
TECHNOLOGY
The CrazyFlie 2.0 systems run on a Python code base. We identified early on we could utilise a cheap piece of hardware called a "Flow Deck" that utilises a low resolution lateral camera to help the stability of the drone. With the flow deck we could pulse in data to the drone such as velocity, yaw, and height. With this we wrote software that allows users to simply enter in an x,y,z coordinates and our software would use trigonometric equations to execute a specified flight path. Then expanded to a Unity interface which is what our clients work with.
OUR
CLIENTS
Prof. Stefan Greuter
Dr. Jordan Vincent
Our clients at the Deakin Motion-Capture Lab (DML) currently makes use of coloured tape on the lab floor space as a visual marker for an actor to identify that location as an "object" they are interacting with (such as a tree or a chair). DML have identified a major problem with this method, as it often results in the actor's eye line shifting direction away from the "object's" natural viewpoint to the marker's location on the floor. This eye change is being captured by the motion capture cameras, requiring considerable post-production work to fix.
​
Our clients have suggested a drone could be used as a physical, visual reference in place of floor tape, able to be flown into desired position that would help the actor achieve a more natural interaction and eliminating the need for post-production work, saving both time and money.
​
This is where our team comes in. Our team consisting of 7 final year Deakin students have created a Unity-based software application that enables a user to control the flight and positioning of a CrazyFlie drone within the motion capture lab space. In addition to flight control, the software is easy to use, with an intuitive, and user-friendly interface.
