Author Topic: My Motorized Shoes Project  (Read 10215 times)

0 Members and 1 Guest are viewing this topic.

Offline yash101Topic starter

  • Contributor
  • Posts: 48
  • Country: us
  • Sup there!
    • My Land
My Motorized Shoes Project
« on: May 01, 2015, 03:19:24 am »
In my STEM class, a friend and I set out to venture upon some sort of gear to aid in transportation via foot. We came up with the idea of motorized shoes. The fundraiser is over, and we got 10% of what we wanted, but it's alright! We were able to shrink our budget to $150, which we spared from the funds as well as through fundraisers at school.

Essentially, we are creating a device to aid a user in walking through the form of placing motorized wheels at the bottom of the shoes. For the control, there is a high-resolution gyroscope. As the user tilts forward, motor power is increased in the forward direction through the use of an H-Bridge controller. This causes the device to lunge forward and straighten itself out. When the user leans backwards, the H-Bridge is first set to braking, converting all the momentum of the shoe to heat, dissipated by our MOSFETs. If the user is standing straight, the H-Bridge will be shut off.

The control systems are based off of Arduino Mini Pros. Yes, I've never used an Arduino and it will be new. I am used to the Parallax Propeller and BASIC Stamp architectures for basic things. For anything more complex, I prefer a Parallax Propeller working together with the Raspberry Pi.

If you guys are wishing to check out our fundraiser, here's a link.
https://www.indiegogo.com/projects/motorized-shoes-the-blue-lodha
It was our first croudfunding-based fundraiser and thus had many loose ends. This resulted in the failure to reach our goal. If I were to relaunch the campaign, I now understand how to make it much more successful ;)

That's the basic detailed description on how the device will work, and you all are probably bored to death by now! :)

I would like feedback towards our project -- it's impracticalities, etc.
I would also like some help so I can complete the build and code of the project.

The H-Bridge drivers are based out of N-Channel MOSFETs. These MOSFETs came with the issue that they are not TTL-rated. Thus, I designed in some 2N3904 transistors to take a 3.3v or 5.0v microcontroller signal and saturate the MOSFET.
We are using a LiPo battery for the first time so we are being a bit cautious. In our setup, for safety, we will have an LM34 temperature sensor in the battery holder to measure the battery temperature.

For overvoltage and undervoltage detection, we will have a resistor voltage divider so it can be measured by the ADC in the Arduinos with their 5v ADC reference.

A picture of the schematic of the H-Bridge driver is attached.
 

Offline SL4P

  • Super Contributor
  • ***
  • Posts: 2318
  • Country: au
  • There's more value if you figure it out yourself!
Re: My Motorized Shoes Project
« Reply #1 on: May 01, 2015, 08:03:46 am »
My first thought when I see projects like this are
'Did the developers take into account the mass of the wearer swinging above the shoes?'

An inverted pendulum comes to mind.
If you tackle problems like these - it will be a fun and rewarding project.  Good luck - and Have fun !
Don't ask a question if you aren't willing to listen to the answer.
 

Offline yash101Topic starter

  • Contributor
  • Posts: 48
  • Country: us
  • Sup there!
    • My Land
Re: My Motorized Shoes Project
« Reply #2 on: May 01, 2015, 02:58:14 pm »
Haha yeah! That's the purpose of the gyroscope. It's just like balancing a pole in your hand. When i starts tipping forward, you move your hand forward, etc.

Of course, it's 100% practical, but it's chugging along well. We're about done with much of our structural stuff. We're 3D printing much of the motor mounts out of ABS so it should be really strong.

The electronics are the fun part; THE H BRIDGES DONT LIKE TO WORK  |O |O |O

I tried simulating the thing in Multisim. It don't work. I built it in the breadboard. It kinda worked. When I plugged a microcontroller in, it died (p00f)  :palm:.

I connected the MCU pin to the base of the 2N3904.
 

Offline Fred27

  • Supporter
  • ****
  • Posts: 726
  • Country: gb
    • Fred's blog
Re: My Motorized Shoes Project
« Reply #3 on: May 01, 2015, 03:59:25 pm »
3D printed ABS is not very strong.
 

Offline yash101Topic starter

  • Contributor
  • Posts: 48
  • Country: us
  • Sup there!
    • My Land
Re: My Motorized Shoes Project
« Reply #4 on: May 01, 2015, 07:23:17 pm »
We have a very high quality 3D Dimension 1200ES printer. We printed the parts so the layers are flat, giving them a much greater strength. We have stepped on the 3D-printed parts and balanced on them, putting twice the typical weight. It handled the weight without much hesitation; there was no visible damage to the 3D-printed parts. They were not even printed solid!
 

Offline Dave

  • Super Contributor
  • ***
  • Posts: 1352
  • Country: si
  • I like to measure things.
Re: My Motorized Shoes Project
« Reply #5 on: May 02, 2015, 05:46:01 am »
That H-bridge obviously isn't going to work. Even the low side FETs won't get turned on with that circuit.
<fellbuendel> it's arduino, you're not supposed to know anything about what you're doing
<fellbuendel> if you knew, you wouldn't be using it
 

Offline yash101Topic starter

  • Contributor
  • Posts: 48
  • Country: us
  • Sup there!
    • My Land
Re: My Motorized Shoes Project
« Reply #6 on: May 02, 2015, 06:01:23 am »
Why is that? The 2N3904s are supposed to switch 12v into the MOSFETs from the MCU TTL voltage.
Basically, at the top, I have the drains of the N-Channel MOSFETs connected. The source is connected to the drain of the bottom transistor. The source of the bottom transistor is connected to ground. The gates are pulled low via a 100k ohm resistor. On each gate, there is an NPN transistor which has it's collector connected to +12v. The base is pulled low with a 100k ohm resistor. The base of the transistors are connected to the microcontrollers. There are two sets of these transistor setups. In the middle of both the FETs (where the source and drain meet) is the connection to one pole of the brushed motor.

I can't seem to see anything wrong with that circuit. It works fine on my breadboard without a microcontroller (still to test).
 

Offline Dave

  • Super Contributor
  • ***
  • Posts: 1352
  • Country: si
  • I like to measure things.
Re: My Motorized Shoes Project
« Reply #7 on: May 02, 2015, 09:56:48 am »
Either you don't have it wired like your schematic is drawn or you are using a much higher voltage to control this H-bridge.
I'll give you a hint: You have emitter followers in your circuit.
I suggest you look them up, hopefully you will see the problem.
<fellbuendel> it's arduino, you're not supposed to know anything about what you're doing
<fellbuendel> if you knew, you wouldn't be using it
 

Offline Phoenix

  • Frequent Contributor
  • **
  • Posts: 420
  • Country: au
Re: My Motorized Shoes Project
« Reply #8 on: May 02, 2015, 10:16:15 am »
Sorry to say, you might be biting off a lot more than you can chew. Both the electronics and the control system are going to be incredibly challenging.

On the electronics side:
The HBridge as drawn won't work properly. It might still do something..

The BJTs Q1, Q2, Q3, Q4 are configured as "common collector" http://en.wikipedia.org/wiki/Common_collector amplifier. This means their output will be roughly their input minus 0.7V. I.e. if you use a 5V micro they will output 4.3V, a 3.3V micro will output 2.6V. This is not satisfactory for driving power MOSFETs.
Also, the voltage applied to the gate of the high side MOSFETs needs to be with respect to their source, not the negative DC bus rail. The challenge here is the source pin is tied to the output of the Hbridge and is sometimes at 12V and other times at 0V - you need a voltage supply which is somewhat independent/floating.

Fortunately for you there are many many Hbridge packages and drivers available (why reinvent the wheel).

On the control side:
What you are describing is the beginnings of a feedback control system. Even worse, the system you are intending to control is inherently unstable to begin with (inverted pendulum). I.e. unless you make exactly the right amount of corrective action the pendulum (person) will inevitably topple. You will need to design a controller that will ensure stability - this involves some pretty complex maths!
http://en.wikipedia.org/wiki/Control_theory
 

Offline Bud

  • Super Contributor
  • ***
  • Posts: 6877
  • Country: ca
Re: My Motorized Shoes Project
« Reply #9 on: May 02, 2015, 01:26:51 pm »
What problem this solution is trying to solve?
Facebook-free life and Rigol-free shack.
 

Offline yash101Topic starter

  • Contributor
  • Posts: 48
  • Country: us
  • Sup there!
    • My Land
Re: My Motorized Shoes Project
« Reply #10 on: May 02, 2015, 04:54:45 pm »
Sorry to say, you might be biting off a lot more than you can chew. Both the electronics and the control system are going to be incredibly challenging.

On the electronics side:
The HBridge as drawn won't work properly. It might still do something..

The BJTs Q1, Q2, Q3, Q4 are configured as "common collector" http://en.wikipedia.org/wiki/Common_collector amplifier. This means their output will be roughly their input minus 0.7V. I.e. if you use a 5V micro they will output 4.3V, a 3.3V micro will output 2.6V. This is not satisfactory for driving power MOSFETs.
Also, the voltage applied to the gate of the high side MOSFETs needs to be with respect to their source, not the negative DC bus rail. The challenge here is the source pin is tied to the output of the Hbridge and is sometimes at 12V and other times at 0V - you need a voltage supply which is somewhat independent/floating.

Fortunately for you there are many many Hbridge packages and drivers available (why reinvent the wheel).

On the control side:
What you are describing is the beginnings of a feedback control system. Even worse, the system you are intending to control is inherently unstable to begin with (inverted pendulum). I.e. unless you make exactly the right amount of corrective action the pendulum (person) will inevitably topple. You will need to design a controller that will ensure stability - this involves some pretty complex maths!
http://en.wikipedia.org/wiki/Control_theory

Alright! I didn't know about that! Perhaps that might solve some issues I was having initially.
If that's the case, how do I use the transistors as switches? I want to put 3.3v at the base and switch on 5v.
 

Offline yash101Topic starter

  • Contributor
  • Posts: 48
  • Country: us
  • Sup there!
    • My Land
Re: My Motorized Shoes Project
« Reply #11 on: May 02, 2015, 05:20:45 pm »
J1 is power.
J2 is the motor output
J3 is the control. There's a ground pin so that the micro can use a separate source. However, this can be ommited.
In J3, there is one pin which goes to the base of each BJT so that the micro can switch it on and off.

Sorry if my question is n00bish. I haven't played with electronics in years and am still trying to catch back up.
 

Online Mechatrommer

  • Super Contributor
  • ***
  • Posts: 11537
  • Country: my
  • reassessing directives...
Re: My Motorized Shoes Project
« Reply #12 on: May 02, 2015, 05:41:55 pm »
ok i get it after concentrating on that, forget the R8.... you are not driving every npn into saturation, if you put 3V on the base, guess what on the emitter? 3V! give or take. but assuming the npn can take it to charge the mosfet to 3V, and assuming now the mosfets are working as you claimed its working on breadboard (the mosfet must be logic level otherwise they will OFF). but it goes poof when mcu'ed? so the suspect is in your code, where is your code? do you know you cant turn on q6 and q7 simultaneously? and q8 q9 too? whats certain, your high side n-mosfet will get hot in no time if your motor sized to bear a man as your project tells, because its no saturated. as other have pointed out, you miss alot of clue you must never learn to walk on a rollerblade, inline skate or a skateboard. you didnt calculate what torque it takes to restabilize a leaning man, what if they lean backward but there forefoot lean forward? you may kill an old man.
Nature: Evolution and the Illusion of Randomness (Stephen L. Talbott): Its now indisputable that... organisms “expertise” contextualizes its genome, and its nonsense to say that these powers are under the control of the genome being contextualized - Barbara McClintock
 

Offline Phoenix

  • Frequent Contributor
  • **
  • Posts: 420
  • Country: au
Re: My Motorized Shoes Project
« Reply #13 on: May 03, 2015, 01:10:03 am »
haven't played with electronics in years and am still trying to catch back up.

Let's start from the top then. What volts, amps, power do you need (you will need to have done the calculations how of much torque is required to move a person)?
 

Offline rob77

  • Super Contributor
  • ***
  • Posts: 2085
  • Country: sk
Re: My Motorized Shoes Project
« Reply #14 on: May 03, 2015, 12:56:51 pm »
i wish you the best with your project, but i'm afraid it's simply mission impossible for you.
as it was mentioned - your problem is like to solve the stabilization of reverse pendulum - but with a MAJOR TWIST ! the reverse pendulum is rigid , the human is not ! so placing the gyro inside the motorized shoe is useless - because the angle of the shoe tell's you nothing about the motion of the mass above (significant mass). Segway is solving this by making the human "rigid" - human is standing on a platform while holding a stick attached to that platform - that makes things easier because the significant mass of the body moves the same direction and at a same angle as the platform.
 

Offline SL4P

  • Super Contributor
  • ***
  • Posts: 2318
  • Country: au
  • There's more value if you figure it out yourself!
Re: My Motorized Shoes Project
« Reply #15 on: May 03, 2015, 01:03:32 pm »
Just while we're looking at variables - I'm always challenged by the use of 'too small' wheels in these projects.

When the chassis hits a small crack in the pavement (let alone a large one) - the balancing algorithms have to go beserk as the wheel stops momentarily and instantly apply massive power (kW!) in reverse - just to stop the rider falling... then if the wheel frees itself, or the load assists with the physical act of re-balancing (by shifting weight etc) to apply a reverse burst to avoid flinging the rider across the room !!!

Razer scooter wheels come to mind - with a MUCH simpler problem to solve!!
Don't ask a question if you aren't willing to listen to the answer.
 

Online Mechatrommer

  • Super Contributor
  • ***
  • Posts: 11537
  • Country: my
  • reassessing directives...
Re: My Motorized Shoes Project
« Reply #16 on: May 03, 2015, 03:16:54 pm »
you dont need stick to be stable, only good judgement and sound analysis...
http://www.tuvie.com/s-walker-board-combines-segway-skateboard-and-balance-board-in-one/

looking at this segway no stick style, the op objective maybe not impossible... with good analysis...
http://www.dailymail.co.uk/video/sciencetech/video-1148832/IO-Hawk-Segway-meets-skateboard-revealed-CES-2015.html
Nature: Evolution and the Illusion of Randomness (Stephen L. Talbott): Its now indisputable that... organisms “expertise” contextualizes its genome, and its nonsense to say that these powers are under the control of the genome being contextualized - Barbara McClintock
 

Offline rob77

  • Super Contributor
  • ***
  • Posts: 2085
  • Country: sk
Re: My Motorized Shoes Project
« Reply #17 on: May 03, 2015, 04:18:50 pm »
you dont need stick to be stable, only good judgement and sound analysis...
http://www.tuvie.com/s-walker-board-combines-segway-skateboard-and-balance-board-in-one/

looking at this segway no stick style, the op objective maybe not impossible... with good analysis...
http://www.dailymail.co.uk/video/sciencetech/video-1148832/IO-Hawk-Segway-meets-skateboard-revealed-CES-2015.html

no stick means it will be almost as hard to ride as riding a surf on waves ;) - totaly useless in "helping" everyday people to move around.
 

Offline rob77

  • Super Contributor
  • ***
  • Posts: 2085
  • Country: sk
Re: My Motorized Shoes Project
« Reply #18 on: May 03, 2015, 04:26:24 pm »
and let's not forget , motorized shoes - it means 2 independent drives trying to achieve the same goal... btw.. i really would like to have an accident on those... just imagine each shoe going to a different direction... ouch ! :D

so if motorized shoes , then definitely one driven and one free-wheeling shoe ! ;)
 

Offline yash101Topic starter

  • Contributor
  • Posts: 48
  • Country: us
  • Sup there!
    • My Land
Re: My Motorized Shoes Project
« Reply #19 on: May 04, 2015, 01:18:26 am »
Truthfully speaking, this project has been really poorly planned!

This was actually one of the first things which came to my mind. Thus, we haven't set a spot for the gyro yet. Maybe we might even strap it onto the leg of the person to be more accurate.
 

Offline yash101Topic starter

  • Contributor
  • Posts: 48
  • Country: us
  • Sup there!
    • My Land
Re: My Motorized Shoes Project
« Reply #20 on: May 04, 2015, 01:22:35 am »
and let's not forget , motorized shoes - it means 2 independent drives trying to achieve the same goal... btw.. i really would like to have an accident on those... just imagine each shoe going to a different direction... ouch ! :D

so if motorized shoes , then definitely one driven and one free-wheeling shoe ! ;)

One of the parts of this project is actually to see if it is possible. We have already figured that it may not be the most practical way of transporation. However, remember that we don't have access to everything. If we are able to get our prototype to move without falling, we  know it will work. If I had a nice lab with unlimited everything, I know that I might be able to make it work and be practical.
 

Offline SL4P

  • Super Contributor
  • ***
  • Posts: 2318
  • Country: au
  • There's more value if you figure it out yourself!
Re: My Motorized Shoes Project
« Reply #21 on: May 04, 2015, 01:31:00 am »
A good attitude, so you probably need to look at your original spec/ requirements.
Do the math and consider comments given here.
It has some serious challenges, but 'nothing is impossible' if you throw enough resources at it !
Don't ask a question if you aren't willing to listen to the answer.
 

Offline zerorisers

  • Regular Contributor
  • *
  • Posts: 132
  • Country: us
  • young noobie, my apologies for low knowledge base.
Re: My Motorized Shoes Project
« Reply #22 on: May 07, 2015, 03:43:45 pm »
I have read quite a few comments on this and they are based on people using the shoes a particular way. just like the heelies that have wheels on the heel you must learn to position yourself with them; even with a bike you may fall a few times but you will get the hang of it. how many wheels does each shoe have and where are they positioned? Stability can help with your design. how will it handle small bumps and cracks where the person is riding (such as a ghetto sidewalk, and other riding surfaces that may be difficult?
Code: [Select]
[img]https://cleanupjamaicaqueens.files.wordpress.com/2013/07/astoria-garbage.jpg[/img]
[img]http://fc02.deviantart.net/fs70/i/2010/218/3/e/Broken_Sidewalk_by_darlingstock.jpg[/img]
[img]https://photos.travelblog.org/Photos/48065/243567/f/2021174-Porto-Ghetto-0.jpg[/img]
 

Offline yash101Topic starter

  • Contributor
  • Posts: 48
  • Country: us
  • Sup there!
    • My Land
Re: My Motorized Shoes Project
« Reply #23 on: May 07, 2015, 06:34:21 pm »
Exactly. This project is designed for those who are flexible and ready to take risks of falling.
There is one wheel on each shoe under the main balance point. The motors balance you going backwards and forwards. You hold your ankles firmly to keep it up. It's just like roller blading on a ball.
 

Offline Delta

  • Super Contributor
  • ***
  • Posts: 1221
  • Country: gb
Re: My Motorized Shoes Project
« Reply #24 on: May 15, 2015, 12:06:05 pm »
Assuming this is a joke, I say well done.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf