I need advice on a delivery drone build for a high school engineering project

For an engineering project for high school, I have chosen to build an automated delivery drone, capable of carrying 1 kg of cargo, in addition to the dropping system we are going to build. 

I need help optimizing this build for flight time, flight range, durability, and low cost (under $300 at most). 

I think the props need to be large, and motors at low kv, for decreased power draw as a trade off for maneuverability. However, I know little of how to select motors and props and I know there is a lot of complex factors in that choice. This is where I need the most guidance. 

I envision a quadcopter being a good choice, but it does not need to be a quadcopter. 

As for a transmitter, I will be buying a cheap one, if at all. The drone will be automated, so a controller will only be needed for the initial testing phase and will not be as important once it is up and flying. 


Heres my hobby king shopping list: 

Quantity     Stock     Product description                                                                     Price     Weight      
4     OK     171000436-0 - NTM Prop Drive 28-30 800KV / 270W Brushl..                    $60.20     320      
3     OK     017000065 - 12x4.5 SF Props 2pc Standard Rotation/2 ..                         $13.08     240      
5     OK     HKz30A - HobbyKing 30A BlueSeries Brushless Speed..                             $55.30     235      
1     OK     HKz-Prog - HobbyKing Programming card for BlueSerie..                             $3.28     17      
1     OK     HK-T6A-M1 - Hobby King 2.4Ghz 6Ch Tx & Rx V2 (Mode 1..                         $25.06     998      
1     OK     FS-L001 - Hobby King 2.4Ghz 6Ch Tx USB Cable                                         $2.85     60      
1     OK     015000095 - XT60 to 4 X 3.5mm bullet Multistar ESC P..                             $3.36     34      
4     OK     SHT007 - NTM Prop Drive 2830 Series Replacement Shaft                           $3.28     52      
4     OK     NTM28-ACC - NTM Prop Drive 28 Series Accessory Pack                             $7.40     88      

TOTAL: $173.81     2044g      


and Amazon: 

Sky Tools 127 Piece Heat Shrink Tubing Assortment                                                                $6.99   1 
Vktech 40x 3.5mm Gold-plated Bullet Banana Plug Connector RC Battery                                 $3.99 1 
SunFouder 2 pcs Ultrasonic Module HC-SR04 Distance Sensor                                                  $8.99 1 
NEEWER® Useful HJ MWC X-Mode Alien Multicopter Quadcopter Frame Kit Red/White               $16.99 1 
Hobbypower Top Pin APM2.6 External Compass APM Flight Controller w/ UBLOX NEO-6M GPS    $69.89 1 
FTDI Cable 5V VCC-3.3V I/O                                                                                                 $17.95 1 
Hobbypower Power Module V1.0 for Pixhawk Apm2.5 Apm2.6 Ardupilot Mega Flight Controller    $11.101 
Neewer Servo Tester CCPM Consistency Master Checker 3CH 4.8-6V                                         $6.95 1 

Subtotal (8 items): $142.85 

I am going to use a VEX robotics motor to drive the dropping mechanism, they are 
reliable geared motors and it will cost me nothing. I will wire voltage from the battery and 
servo signal from the APM, they never draw more that 5A so it shouldn't affect anything. 

FTDI cable is to set baud rate and name of bluetooth module (JY-MCU slave) 
    -Can the APM 2.6 take serial input coordinates and route to them? (I am willing to code it myself) 

I already have a battery and charger, so that saves ~$60 

   The Turnigy nano-tech 4500mah 3S 25~50C Lipo Pack is what I already have, and seems like a good choice to maximize flight time to increase delivery range, but is pretty heavy. 

I have a mix bag of those little plug wires, and 3 wire ones at the school I can use. 

The distance sensors are for collision avoidance, is there any reason not to use those? 
    -Can I add code to stop and rotate until it is clear of obstruction? We do that in vex with robots, any reason this would be hard on APM? 

The frame is inexpensive, if it does break, then we will build a H frame, but I want something to start with. 
        (also it looks cooler than any welded aluminum frame I could make) 

Anything I can cut cost on? $316 subtotal + ~$40 for hobbyking shipping (yikes!) is within the budget. 

Anything I forgot? 

Thank you for reading my post, if you have any knowledge that can help me succeed in this project, I would love to hear from you.

Views: 1529

Reply to This

Replies to This Discussion

you have to redo your whole setup, even if it flies it probably might just fly for a few minutes - it seems like you started with an idea and just went and purchased parts. 1kg payload delivery drone under $300 is not an easy thing to do if you want reasonable range/time.

Before we start, how about you tell us a few details:
1. What is the targeted flight time?
2. What is the targeted range?
3. How is the drop mechanism mounted? Rails or hard mount?

Essentially, suggest you work downwards from an objective-oriented approach. Let us know the objectives and we can give some suggestions.

Thank you for the insight! 

-Our minimum range is one kilometer. 

-We need enough flight time to make it that far comfortably, but I don't know what ground speed to expect.

-The drop mechanism is a motor with a hook,  bolted to the frame. A box with a metal eyelet is held by the hook, when the motor rotates, the hook slides out of the eyelet and drops the box. The motor we have weighs 60 grams.

What I need help with: build a drone that can fly 1 km, and back, using coordinates and a gps, and perform a dropping action at the coordinate.

What frame size is best for this?

What kind of motor and props will be most efficient for this?

What size battery?

Is the APM2.6 and UBLOX gps a good controller for this objective?

I have not built a drone before, and I understand this is not a good build to start with, but it's definitely feasible. 

    

 

That's not too hard. 2km round trip is very achievable. I was thinking more like 8km trip. Flying at 20km/hr means you need about 6min flight time back and forth. Good news is once you release the package your drone won't need so much energy to fly.

Going by this the target, you'll need to build a quad that can fly for about 8min, carry 1kg load for the first 3 min, then fly back. Let me work out some calculations and get back to you in this thread.
8min so you can have a 2min buffer. And also don't discharge your lipo too much.
As a side note this had similar objectives to one of a competition I helped a school with - the royal Thai air force uav challenge last year. Dropping a payload is not hard - what's hard is getting the payload to land accurately.
Finally, you are buying a lot of unnecessary stuff. And that frame you chose is overkill in terms of weight.

For electronics, all you need is an apm, a gps, a 5v reg, and data radio set. Forget the sonar they are next to useless. Invest in a better gps for higher accuracy if you wish.

The rest are extras you don't need. Even the bullets. Solder your connections.

I'll come up with a list next.

I did feel like I had come up with a lot of unnecessary parts but I wasn't sure what to strip.

The frame was also something I had trouble finding, I was trying to minimize radius but I realize that is probably pointless.

Thank you for all the help, the judges at our pre-project presentation seemed to think we were going to fail, most of the projects at my school are far simpler. I was worried that the constraints we had set were too hard for the budget, I'm glad to hear that they are doable.

I can't wait to see the list.

We have to do a lot of data collection on our projects, so I'll contribute all of our data here, and I'll get a copy of the presentation after we present in March.

I did some more research and came up with this:

flight computer: apm 2.6 with ublox gps

http://store.3drobotics.com/products/apm-2-6-kit-1

     $240

    

frame: Tarrot fy 650

http://www.helipal.com/tarot-fy650-quadricopter-frame-set.html

    $130

motors: Rctimer 5010/14 360KV

http://www.rctimer.com/product-575.html

   $73 for four

props: 16x5.5" Carbon Fiber

 http://rctimer.com/product-841.html

    $35 for four

battery: 4500 mah turnigy nano tech

Turnigy nano-tech 4500mah 3S 25~50C Lipo Pack 

    $0 I already have this

esc: 4 in one 20A Qbrain esc

http://www.hobbyking.com/hobbyking/store/__36674__q_brain_4_x_20a_b...

    $0 I already have this

optional:

    3DR telemetry kit for data collection

    $100, a bit too expensive to be justified for this project

apm, frame, motors, props, rxtx

240 + 130 + 73 + 35 + 25 =

$503

 

The APM is expensive, but for a good autonomous system it seems to be the best value

The frame is still expensive and heavy- but cheaper frames only seem to get heavier.

 

Is there an advantage of the expensive CF props, or are the wide plastic one comparable?

assuming I can find a cheap frame or make one out of pvc, and get cheaper props, that puts it at

apm, frame, motors, props, rxtx

240 + 40 + 73 + 15 + 25 =

$393 which is better but I don't know if that's feasible.

The computer is 60% of that cost, but any cheaper APM's turn out to be Chinese knockoffs. 

Thoughts on where to go from here?

you'll need a transmitter- I dont think flight without one is available with the firmware. Also be weary about going large- you'll have to fly in biigger places and flying near people/things becomes way more dangerous, and crashes can  be more costly

Re the frame if you can find someone with a 3d printer locally maybe they can print you a frame. You can get a real nice frame for not much cost in filament. Look on Thingverse for many options. 
Greg

The diy foam tube frame shown here: https://www.youtube.com/watch?v=5XinbTcQAAc

looks like a possible frame that would cost practically nothing.

Its very light, and the lack of hardness means it won't transfer vibrations to the electronics, but that also means it's flimsy.

What would be the advantages of something like that? If it's practical to use that kind of construction, the budget would relax a bit. Also, mounting the drop mechanism will be much easier with the H-style square frame.

We have a makerbot printer at my school, but to fit the large propellors, the frame has to have a 1 foot radius, and at that size I feel like it would get heavy; most 3d print frames are for smaller quads. Is there a particular print on Thingverse you think would work? 

The large prop size is for the sake of efficiency at the cost of maneuverability, which is optimal for carrying a load in a straight line over a few kilometers. Am I correct in thinking that is how prop size affects performance?

Reply to Discussion

RSS

Groups

Season Two of the Trust Time Trial (T3) Contest 
A list of all T3 contests is here. The current round, the Vertical Horizontal one, is here

© 2019   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service