Gyrocopter Project Concept

[Shadolance]

Hello everyone, my name is Shadolance (new to the forums here, this is my first post!), it's a pleasure to meet you! I was reading through a bunch of topics on here and it seems a lot of people have been getting great help starting projects, and I was wondering if I could get some help with my latest idea as well.

Simply put, the title is correct: I want to build a gyrocopter. The basic idea is to have a pendulum-style body, bottom-weighted where the hardware will almost entirely reside. Power supply, microcontroller, motor controller, etc. This "bulb" will be connected to two counter-rotating blades at the top, creating a gyro. Think of an umbrella shape almost, where the blades act as the cover and the "handle" is where all of the hardware is located. The end goal is to make a mock-aircraft model where in theory a person could ride inside the "shaft" part of the umbrella. As for scale size, preferably no taller than a foot and a half, and the blade can extend as far as they need to to provide appropriate lift.

Now, for what I need help on is primarily choosing parts, but not so much because of not knowing which parts are appropriate to do the job. The biggest issue for me is determining weight and the amount of power necessary to lift the device off the ground. More powerful motors are larger and heavier, as are batteries to power them. However, this needs to scale to the amount of force provided by lift. This scaling is hard for me to get my head around.

As for my current plans, I intend to use a Teensy 3.1 unless given strong reason otherwise. I am very familiar with all of the necessary code that will go into it, that is the least of my concern. I need help picking out appropriate devices for the following feature I intend to include:
- Auto-stabilization
- obstacle avoidance (sonar sensor preferably, 3 foot range minimum)
- Remote control (I've only ever built autonomous robotics in the past, this is new to me. Bluetooth perhaps?)
- Full lateral movement (meaning the blade will have to be able to angle themselves appropriately)

Thank you all for reading this if you made it this far, I greatly appreciate the time! Like I mentioned before, I'm new to this forum so if I did something taboo, I apologize and will fix what is necessary. I'm a fairly experienced electrical engineering student at university, and I'm confident my coding ability will be enough to get the majority of this project done (should I need help I look forward to asking for a hand in the future!)

Above all, this is a learning experience for me. I'm super stoked to build this thing!

Cheers,
~Shadolance

PS - Assume for the sake of the project that there are no financial constraints.

 

[onehorse]

Well, you'll want an altimeter, meaning a pressure sensor. I recommend either the MS5637 or the smaller BMP280. These are good to ~15 cm in relative height discrimination with good signal-to-noise. They are also small, light, and cheap.

You'll need an IMU, I recommend the MPU6500 (accel/gyro) or MPU9250 (MPU6500 + AK8963C magnetometer). It is small, light and cheap as well as accurate with proper calibration.

I am also thinking about using the VL6180X myself but more likely its successor in my flying devices. This is a proximity sensor (for collision avoidance and constant altitude applications) that uses time of flight to measure absolute distance currently out to ~25 cm, but the successor (which I saw demoed last week and is available early next year) can go out to 1 - 1.5 m. The advantage of this technology is that it is only weakly dependent on the reflectivity of the surface being measured, unlike proximity sensors that measure reflected signal strength. And since it uses an 850 nm (940 nm in the successor) laser, it has discrimination down to a small number of mm. Ultrasonic sensors can't do as well. The other benefit is it is a small, light-weight, and relatively inexpensive part.

I am using Nordic's nRF51 which has an embedded blue tooth engine along with the Cortex M0 processor (nRF52 will have an M4F) for a variety of applications but the blue tooth engine takes precedence over other processor tasks so there can be a latency problem, especially if you need to stream and process sensor data at high rates. You could add a bluetooth module to the Teensy, but for your application an nRF24L01+ 2.4 GHz radio is probably sufficient for remote control, and they can be very small, compact, and light.

You could consider an EM7180 sensor hub to offload sensor fusion from the host controller leaving it free to manage the motors.

For motor control, I would recommend TI's NexFets if you are using a FET/resistor network to drive brushed motors, these have very low on-resistance and are very small and light; they come in a chip-scale package so maybe a bit of a challenge to solder but I do it routinely with hot air and solder paste. I also like TI's H-bridges for some applications since you can run the motors forward or backward. The on-resistance is several times larger though compared to the NexFet solution and the size might be a tad larger depending.

If you are going to use brushless motors, I have a very small ESC (10 x 10 mm) that runs from a single cell (1s) LiPo battery and can drive up to 3A of current; 12 W per motor should be enough for your project, although most people want to go with a 2S or 3S solution to be able to use massive motors. I would expect that even with the 3 g BLDC motors you can buy from Hobby King for $15 you could easily get 20 g of thrust per motor if you choose a sensible prop with these ESCs. I would recommend you scale your device down so it can run from a 300 mAH battery, which probably means 50 - 60 g total weight. If this is not feasible, you will be in the 2 or 3S category with a large jump in weight and cost.

You might also need a boost converter to keep the voltage on the Teensy and peripherals above ~3.3 V or so at maximum thrust.

 

[joe_prince]

There was a Kickstarter campaign a few months back for something similar to what you're trying to do: https://www.kickstarter.com/project...le-rugged-totally-different-small/description

You might be able to pull some inspiration from that to help guide you. I know that UAV was capable of autonomous flight, but I'm not sure if it was using sonar, lasers, optical flow sensor, etc.

As far as sizing the components goes, first figure out what your expected/target weight should be. A good rule of thumb is to provide around twice as much lift (thrust) as the entire weight of the system. So, for example if the total weight is 1000g, the motor should be capable of outputting at least 2000g of thrust. If you have two motors, then each motor needs to put out 1000g of thrust, and so on. If the thrust output of the motor is too close to the weight of the entire system then even if you get off the ground it's not going to get very far.

A lot of BLDC motors that you find online will have a thrust chart with them for given prop sizes and pitches, but if they don't then you can perform thrust tests yourself. All you need to do is build a simple support for the motor, and attach the opposing end to a digital scale. Apply power to the motors at various percentages and see what the thrust curve looks like (you're probably only interested in the max thrust though at 100% power). See here for more information on such a test stand: http://blog.oscarliang.net/diy-thrust-station-measure-motor/

The weight of the motors and electronic speed controllers (ESCs) aren't that much compared to the battery, so the bulk of the weight will be in that. In addition to having a battery capable of a high enough voltage for the required thrust (e.g., 1S, 2S, 3S, etc.), you'll also want to consider flight time. A larger battery will give you more time in the air, but the tradeoff is the extra weight. One thing I have found helpful in the past is to go online and find a UAV, or quadcopter, kit that somebody sells, and see what size motors, ESCs, batteries, etc. that they have selected for the given size and weight. This is usually a good starting point when you're not quite sure on the size of each component.

For auto-stabilization you'll likely want an IMU and possibly a GPS, or some other form of barometric pressure sensor. For obstacle avoidance sonar would probably work fine, and you can also consider a laser range finder (LRF) which might give you more precise measurements, or additionally an optical flow sensor. The LIDAR Lite is a cheap solution for an LRF: https://www.sparkfun.com/products/13680

There are many solutions for the wireless control, but it largely depends on the range you need. Bluetooth would be adequate for short distances (maybe a few hundred feet at best), but for anything further you might want to look into the xBee RF modules, or even a dedicated off-the-shelf RF receiver.

 

[Shadolance]

Thank you both for your help so far! I'll assume the pieces are all compatible with each other. I'll definitely look into these parts over the next week and see what looks good, and come back with any further questions!

As for the kickstarter page that Joe linked me to, My mental image of the device is pretty similar to that, but the goal of my project is less about durability, and I don't intend to attach a camera to it (at least not in the first iteration). I want to make a super-stable remote-control aerial device, mostly for fun but also for personal recreational use.

Anyways, thank you again, if anyone else has input I'd love to hear from other sources as well! I'll get working on this ASAP

Cheers,
~Shadolance

 

[JacksonWillis]

Gyrocopter Project Concept is innovations, and best our own one of way work for the plan of this forced gyroplane to give a perfect safety.