Received DX6i Transmitter/Receiver, Dean’s connectors and bullet connectors. Purchased aluminum tube, sheet of aluminum and a handful of screws from Home Depot. Having difficulty cutting aluminum, purchased metal hacksaw. Tested Transmitter/Receiver combo with servo motors. Works as expected. The servos included with the transmitter are very small and light, and may prove useful during the construction of a gripping mechanism.
Used some spare female headers to create a power connection to the receiver. The Dean’s connector was soldered to the power input of the ESC, while the bullet connectors were used for the motor/ESC connection. The Dean’s connection proved to be a solid and strong connection. The bullet connectors also provide a strong connection, though has a greater risk of a circuit short. Heat shrinks were used as a protective cover.
For the receiver, power can be applied to any of the inputs on the receiver. At rest, the receiver operates on 5V and 30mAs. When controlling a servo motor, the current requirement depends on the number of servo motors attached and moving. Since the receiver will be sending signals only to the Arduino, current draw should be similar to when the receiver is at rest.
Read the output signals from the receiver using an oscilloscope. Despite the poor quality of the oscilloscope, the output signal is indeed a PWM signal; however, it does not have a fixed time slot. The order of the PWM signals is as follows: Aileron, Aux1, Elevator, Rudder, Throttle, Gear. There is no gap in between outputs, as soon as the pulse width with one output is done, the next one begins immediately. Due to the dynamic time interval, an interrupt system to read the receiver signals in necessary. In addition, the output of the Aux1 pin mimics the throttle, however the throttle signal is only outputted when the throttle hold on the transmitter is not on. Activating the throttle hold immediately disables the throttle.
Cut out and drilled the motor mount from a sheet of aluminum. Motor did not mount completely parallel, used cardboard to straighten. Not an ideal setup, but will suffices for a prototype. Final version will be better cut and drilled. Motor mount design will require extra work.

Figure 1: Motor Mount Drawing
Soldered the bullet connectors for the motor and the ESC. Soldered Dean’s connectors for the power for the ESC and to 18 gauge wire. Connected the receiver to the ESC and the ESC to the motor for a prototype test. Completed a successful test of the transmitter->receiver->ESC->motor chain. However, the current power supply is insufficient beyond 45% thrust; it cannot output enough current for the motor.


Images
Figure 2: Servo Motor
Figure 3: Receiver
Figure 4: Transmitter

Figure 5: Testing the Receiver
Figure 6: Bullet Connectors
Figure 7: Dean's Connector
Figure 8: Motor Mount Prototype

Figure 9: Motor Mounted
Figure 10: Motor Testing

Ordered More Parts

Ordered Arduino Mega, 5 DOF IMU and IDG500 gyro from Sparkfun Electronics.

Searched for the best way to connect the motor to the ESC. The most commonly used connection is the high amp rated bullet connectors, commonly found at RC retailers. A cursory search through Jameco, Mouser, Digikey did not find any suitable replacements. A similar issue with the connection from the battery to the ESC. The battery uses a bullet connector of undetermined size. To route power from one battery to four ESCs will require a separate adapter to be made. Suitable cable can be found in the back cabinets. A commonly used 2 pin connector used with RC equipment is the Dean’s connector. Ordered a pair of both from an online retailer, but will continue to search for cheaper alternatives.
Ran through the size calculations again. Maintaining a vehicle diameter of 750 mm. The rotor blades have a radius of 127 mm and a diameter of 254 mm. The wires from the motor have a length of 70 mm. The wires from the ESC to the motor are 114 mm. Placing the motor at a blades radius from the edge puts the motor at 248 mm from the center. The wires from the motor and the ESC places the ESC at 66 mm from the center of the vehicle. The wires from the ESC to a power source is 100 mm long, and the wires from the ESC to the control source is 215 mm long, both are sufficiently long enough that no extension cable is necessary for wiring.
Placed several orders for components: DX6i Transmitter/Receiver, 4x TowerPro Brushless Outrunner motors, 4x Turnigy Plush 12 A ESC, Zippy Flightmax 4000mAh 3S1P 20C, several connectors, and a set of metric tools. Components from Sparkfun Electronics currently out of stock, but expected to be in soon.

Yay! Real Work!

Received the small parts ordered on Sept 9th. Bought an 87 cent PVC pipe at Home Depot. Home Depot has a disturbingly low amount of supplies, lacking most egregiously metric rulers and small metric sized screws. This may require trips to craft stores or require ordering small parts online. Pictures of all received parts were taken. Notes on the parts: The base for the motor has three holes for 1.5mm screws. A drill bit of 1/16th of an inch is sufficiently close to drill holes for the 1.5mm screws. These three holes are not evenly spread out; one on one side, two on the other separated by about 1cm. The wires from the motor will need an adapter; I do not want to solder these directly to wire. As both the motor and the ESC will have up to 15 Amps through them, a high gauge wire will be necessary as well as high gauge adapters and switches. A female connector will be on the battery, so a male connector will be needed for the ESC. A three pin male connection will be attached on the ESC to the motor. Extra wire will be needed to ease connection. A cursory search on the ampacity vs wire gauge suggests that a minimum gauge of 14 (American) is required to handle the current. The propeller blade fits snuggly on the motor shaft, and interlocks well with the supplied locking nut (5mm). A 5mm wrench will be necessary to lock the nuts in place. The control signal from the ESC is a standard three pin connection. While not necessary, a separate breadboard may be used to keep the wires organized when connecting to the Arduino board. An old Arduino NG will be used for testing before the Arduino Mega arrives.

Hopefully I'll get the parts before the semester ends

Gave list of the components for Stage 1 to Prof Grodzinsky. Omitted the carbon fiber frame and the ultrasonic sensor because they’re both superfluous at the moment. Aluminum or PVC pipe will work as a cheaper and easier to obtain substitute for the frame, while the ultrasonic sensor is not necessary for manual flight. Halfway through the completion of Stage 1, I’ll order the sensor so that it will be available by Stage 2.

TIme to get serious

Drew Stage 1 wiring diagram. Began writing pseudocode for the Arduino Mega. Both of these elements borrow heavily from the AeroQuad project due to their similarity at this stage. Later stages will deviate from this project greatly.

Shhh... Don't tell Grodzinsky

Ordered some small parts: TowerPro Brushless Outrunner motor, EPP1045 Contra Rotating Blades, Turnigy Plush 12 A ESC. Once they arrive, work can begin on testing the motor thrust capacities and modeling the motor system.

School's Back

Read the Intelligence, Surveillance & Reconnaissance 2009 Report from Flight Insight. The report discussed the future of ISR vehicles, both manned and unmanned in the military sector. Majority of projects under review are large, long range vehicles. Notable exception is the Honeywell Tarantula-Hawk, a small close range ring-wing vehicle, designed for the US Army’s Future Combat System as a platoon drone.

Just being lazy

Continued SolidWorks rendering. Designed tethered rig for Quadrotor testing. Added sections to proposal on timelines and testing procedures.

The CLAW

Worked on Solidwork rendering. Started updating proposal with the 6th Mission criteria. Examined possible visual confirmation equipment. A self-contained camera system, with built in object tracking would be suitable. This system would ensure the correct orientation of the grip to obtain the flash drive as well as ensure that the fake flash drive was not picked by accident.

Minor Experimentations

Examined possible grabbing mechanisms. Potential actuator systems include microservo motors or ‘biowire.’ Another key consideration is designing the system to deposit one item and picking up another. This will undoubtedly require the Quadrotor to maintain a stable hover without X-Y movement. In addition, a nonscientific experiment was conducted on the effect of smoke on ultrasonic sensors. The smoke particles and the change in temperature does not affect the sensor in relation to the resolution of the output (approximately +/- 1.25 cm). Another unscientific experiment was performed to evaluate the effect of the rotor thrust on a stack of papers with a flash drive on top. The results of this experiment suggests that a helicopter cannot approach the stack of papers without disturbing it unless it flies above the stack at least 1 m. However, once directly above the stack of papers, it may descend to landing without disturbing the papers. This behavior will be incorporated into the flight path design during the depositing and acquisition of the flash drives.

The Future is now! Flying spybots

IARC released the 6th mission. The 6th mission follows the 5th mission by requiring navigation in a crowded indoor environment. In addition, the UAV must be able to pick up and deposit a small object without being detected by roaming video camera surveillance. Due to the similarities between the two missions, the major characteristics of the proposed Quadrotor design do not require any changes. The addition of a small grabbing mechanism is all that will be required.

Quadcopter Validation!

Read through the competitors research papers. IARC revealed that the MIT group won the competition. Which means my design is no longer valid for competing in the competition. The three finalists was the MIT group, GA Tech and Embry-Riddle. MIT used a Quadrotor design with the laser range finder, which very much resembled my design. They used an Intel board for stabilization, movement and obstacle avoidance, while the ground station laptop did all the mapping and object identification. The GA Tech group used a CoAxial helicopter design, controlled by an Atmega controller and used the XBee communication system. The Emby-Riddle used a monocopter design that used an Arduino for control.

IARC 5th Mission Completed

IARC revealed that the MIT group won the competition. Which means my design is no longer valid for competing in the competition. The three finalists was the MIT group, GA Tech and Embry-Riddle. MIT used a Quadrotor design with the laser range finder, which very much resembled my design. They used an Intel board for stabilization, movement and obstacle avoidance, while the ground station laptop did all the mapping and object identification. The GA Tech group used a CoAxial helicopter design, controlled by an Atmega controller and used the XBee communication system. The Emby-Riddle used a monocopter design that used an Arduino for control.

They will release the 6th mission on September 1st.

Meeting with Grodzinsky

Met with Grodzinsky today. he gave me some suggestions

1. Look into some real-life applications and what adjustments that need to be made: ex. Nuclear Radiation or Fire suppression
2. How does smoke effect ultra sonic sensors
3. Return policy on electronics
4. More detailed timeline

Back to work

And now I'm back to work. After being on the road for basically two months, I've found it very difficult to get back into any sort of rhythm. I'm slowly getting there though. Unfortunately, I accidently threw out all my notes that I made while I was at my parents house. They survived the entire 11,464 mile trip only to be thrown out when I cleaned my room. :(

Anyways, I've made some more progress on this proposal. Up to 40 pages now. Still feels like I'm missing a lot though. Tomorrow I'll flesh out the conclusion and then it should be fairly presentable. I'll of course continue to work on it.

The webforums have been fairly quiet while I was away. I guess everyone else also went on vacation. However, someone did find a place to get these vibration dampening grommets that look very promising. I'd like to test them in my motor thrust test-bench.

Starting tomorrow, I'll also get back to work on using SolidWorks. I picked up a book back in May, hopefully I'll be able to master it before the semester begins.

Road Trip

I did not finish the proposal. : ( And now it is time for the road trip. I'll be back mid July.

Productivity is proportional to the size of my desk

Having been stuck at my parents house, I have accomplished very little this past week. I've sketched out a couple more pages on the proposal, read a couple more papers, but very little substantial. I'll try to add what I've done to the proposal by the end of the week. I sorely miss my computer.

Sadly, I'm going to be even less productive over the next 5 weeks. From June 15th to July 15th, I'm going to be driving around the country, helping my parents move and visiting a bunch of national parks. It'll be an exciting adventure. Of course I'll regret it when I have to get back to work and realize how much stuff I need to get done before the school year begins.

Anyways, substantial info. I've fleshed out the thrust test rig section. This piece of equipment is going to be very critical to the success of the project. I have to give much thanks to Old Man Mike and 13brv3 for their excellent work with a similar thrust test rig. They've devoted countless months trying to find the best motor to use and how to decrease unwanted oscillation. While I won't be testing out other motors, I hope to be able to contribute to their great work. My test rig will mostly be dedicated to providing a detailed model of the motor system. One of the problems that they have been facing is the lack of detailed specifications. Hopefully in the course of my work, I'll be able to provide some details.

Hopefully I'll get another update soon

being lazy

I haven't been quite as productive as I would have liked. I ordered two new books, one on UAVs and another on SolidWorks. I figure I really should learn SolidWorks, and not just for this project. I'll do some light reading this week, but most of my major work will be next week. There's a lot of info in the RC groups forum that I need to sift through. I saw some nifty chassis ideas, including using one of those security camera domes as a protective top. Made it look really cool. I feel hesitant about the thin carbon fiber arms, but I know they're stronger than they look intuitively. I do like the screw locking method one person used, since it wasn't permanent, yet was strong. One person used a 4 tube interconnect as a base for the chasis. It looked a lot more stable than other methods.

In terms of my proposal, I wasn't entirely pleased with the format of the proposal so I'll probably try to reorganize it. I do need to add in a section on some mathematic models. I find it humorous that in Musial's book that he commented that several papers used the same model for controller design and simulation verification. Of course their simulation will work; they're predicting the data that they used. I notice Musial's observation to be too true. In my incomplete research survey, I've found no less than 7 such papers. Tisk tisk. I fully plan on doing more rigorous simulations for my mathematical models. Even if I'm just going to use a simple PID controller.

Anyways, back to my 'vacation'

PPT

Decided to make a powerpoint of the proposal. You know, for people who don't feel like reading a 35 page document. You can find it to the right.

so close

no introduction yet, but everything else is done. Ish. Its all a rough draft. A 34 page rough draft. It still needs work. sad face

Hungry

Up to 25 pages now. Getting closer. Not really. I'm almost done writing the components I need. But I haven't even started working on the introduction yet though. The construction process should be fairly simple though.

Slowly getting done

Still no introduction, no review of existing designs, and nothing yet on the details of my plan, but I'm almost done picking out my components. I know which motor and blades I want and what I'm going to build it out of. I just to write it up. Still thinking about the stabilization sensors. I might just use the MikroQuads set up.

Up to 18 pages now. Of course, now its starting to get populated by images and graphs. But there's still plenty of text and even more still to write.

No PDF yet, my lab computer is being screwy with the PDF maker. I'll probably have it up tonight.

Solid Works

Ok, so Solid Works isn't exactly the easiest program to figure out. All I want to do is make a very simple drawing of my quadcopter design... grumble. I may skip this section till another day when I have the time to figure this out.

Back to writing

Proposal Update

Updated the proposal. Added more details on my project plan, both for my thesis and for the quadcopter design. Up to about 13 pages right now. I am now a supporter of Word 2007 and its auto labeling, auto table of contents, auto works cited stuff. Makes everything a lot easier. Well except now that I have more things to fool around with. Started work in Solid Works so I can model the quadcopter. I should have a rough model done by Friday to add into the report.

Name?!

The planning phases of the proposal are coming along, albeit slowly. However, I've come across a major dilema. I don't have a name for it yet. This is critical. A name is everything. A rose by any other name would not smell as sweet. Its all about marketing.

Thesis Proposal

Ok, so I figured I'll always keep a copy of my Proposal up here for everyone to take a look at. Provided I figure out a way to do this easily. Its giving me fits thus far. So here's a pdf

Its a work in progress. Seriously. It really is.

Quadcopter UAV

So its time that I start working on putting my Master's Thesis into action.

• Step 1: Start telling everyone about it.
• Step 2: Start a blog so that everyone can keep track of my progress
• Step 3: Write the proposal
• Step 4: Forget to update Blog, Twitter, Facebook with my current progress
• Step 5: Make my quadcopter UAV
• Step 6: Finish Thesis

I'm now at Step 2. Work on Step 3 has begun. Expect Step 4 to start before Step 3 is completed. Who knows when I'll even reach Step 5. My due date to finish Step 6 is in less than a year. We'll see how that goes.