Documents
Project Overview
The main deliverable for this project is a robust quad-copter platform. We will be responsible for providing an integrated hardware and software solution that provides a flexible API for future students to modify, enhance, and extend the abilities of the quad-copter. The quad-copter will be used initially to navigate a three dimensional maze which will provide the basis for a “flying Roborodentia” competition. Our work on this project will benefit many people, as well as the school itself.
Clients
Our client for this project is Dr. Khosmood of California Polytechnic University, San Luis Obispo (Cal Poly). He would like to develop a new competition similar to Roborodentia, an annual autonomous table-top-based robotic competition, while adding flight and a 3d maze.
The direct beneficiaries of our work will be the students of Cal Poly. They will be able to gain experience with embedded programming and managing airborne autonomous robotics. Companies who hire from Cal Poly will receive the indirect benefit of being able to hire more experienced students.
Dr. Khosmood will receive our prototype quad-copter, all of our source code, quad-copter designs, and 3D-maze at the end of this project. While Dr. Khosmood is our direct client, other persons and entities will have a stake in our project.
Stakeholders
Stakeholders for this project include but are not limited to: Professor Lupo, Brandon Bussjaeger, the CPE department, Cal Poly University, and any other universities which might end up competing using our quad-copter as a basis.
Professor Lupo will ultimately be assigning grades for this project and as such has a vested interest in monitoring our progress and results. He will also be our main liaison to the CPE department in order to coordinate project demos for events such as Cal Poly’s Open House.
The CPE department has interest in this project chiefly because the concept of an autonomous quad-copter is seen as a great way to interest and bring into the department prospective high school students.
Open house is a big attraction point, and the autonomous quad copter would get a lot of attention and praise. For the quad copter to be at open house, it has to be robust enough and simple enough to set up and it must be safe.
Cal Poly University benefits indirectly from the success of this project. Our quad-copter would become another example of “Learn By Doing” to attract new students and to advertise the school across the nation.
Other universities may benefit from the success of this project if the concept of the “flying Roborodentia” competition spreads beyond just Cal Poly. In this case, our prototype quad-copter, associated code, and designs may provide a starting point for students from other schools to compete in this competition.
Brandon Bussjaeger is a Computer Science major who is working on his Master’s Thesis. He will be using our API to control a quad copter.
Need Statement
A few top engineering universities are investing in quad-copter systems, but most of these systems tend to be expensive and specialized. Nonetheless, quad-copters are a good platform for robotic research in vision processing, path finding, and multi-robotic systems. Currently, Cal Poly lacks any quad-copter program. At the end of this project, we plan to create a quad-copter system for Cal Poly that will be durable and extendable. This will allow current students and professors access to a new emerging field of research in computer science. It will also give Cal Poly the ability to attract talented new students who are interested in robotic systems.
Goals and Objectives
Our ultimate goal is to build a robust quad-copter design that can be used in a “flying Roborodentia” competition. The goal of our software is to provide an API for future students to be able to control the quad-copter without needing to touch the low level hardware controls. Our hardware design should be rugged, and relatively easy to build.
Our objectives are as follows:
- Select components (suitable sensors, motors, propellers, batteries, a processors, and a frame) for the quad-copter
- Interface the processor to the sensors and motors
- Interface and test network hardware and implement some kind of ground controller.
- Create and test a basic set of movement functions for the quad-copter.
- An example set of simple commands would be “hover”, “go up”, “go down”.
- Implement proximity detection to avoid obstacles.
- Use a 3D-maze to test more complicated movement for the quad-copter.
- Program the quad-copter to explore and solve the 3D-maze.
- Provide comprehensive documentation on our design decisions and implementation details.
Using our objectives, we have developed a set of outcomes and a final deliverable.
Project Duration
This project began on October 1, 2012 with our first client meeting. The duration of this project shall last until the end of the Winter 2013 quarter, March 22, 2013.
By the end of the Fall quarter, we will have a hovering prototype quad-copter, which will mean that all the hardware layout and construction will be complete. During winter quarter, the artificial intelligence, object avoidance, networking, and ground station will be completed. With these deadlines in mind, we have developed a mission statement and set of objectives.
Team Mission and Team Objectives
With our goals, outcomes, deliverables, and needs in mind, we came up with a definitive mission statement:
Develop a robust quad-copter system that can serve as an example for a possible future autonomous field-based aerial robotics competition at Cal Poly.
A key objective of our team is for the system should be affordable and easy to recreate for future students, while providing abstract API for them to use as well.
Another goal for the team is to gain useful real-life experience by designing from start to finish a robust embedded system. By the end of this class we should be more confident in being able to pick parts for a specific set of requirements and subsequently design software to make the hardware achieve those requirements. Building a system such as ours requires organization and planning.
Team Membership and Roles
There are five members on our team with varying skills and past experience. Doug Gallatin, Tim Peters, and Michael Chamoures prefer working with software while Drew Bentz and Will Budney prefer hardware. Even so, every member of the team will be helping in both areas when needed.
Using a nomination scheme, the following general team positions were determined:
Project Manager: Tim
Procurement: Will
Financial Officer: Michael
Similarly, the technical positions were assigned as follows:
System Architect: Doug
Software Developers: Tim, Drew, Doug, Michael, Will
Hardware Layout and Integration: Michael, Will, Drew