Where do I go? Robot Idea

Hello!

I'm looking to build my first big project. Granted, I haven't done more than mess around, but I'm up to the task. However, I don't know which direction to go. I was wondering if someone could point me in the right direction.

Hardware:
I have many Arduino Unos, one Mega 2560, a Mega prototype shield, Circuits@Home USB shield, Seeed Studio Relay Shield, and a Seeed Studio SD Card shield. I also have the servos (HXT900 9GR servos) that I need for prototyping.

Concept:
I want to make a robotic head on a transport device. The head will have 10 servos for the neck, ears, eyebrows, eyes (2 servos each), and mouth. The rest of the body will be mounted on either a tricycle or 4 wheeled toy car (think Power Wheels). I want to be able to control the robot from a distance (10-20 feet). I also would like to communicate through the robot with a microphone and speaker (to send and receive audio) and control one of the servos.

I would like to use a PS3 controller to manipulate the servos. I have successfully set up this and working on the code. Any tips here would be appreciated.

I have no idea what motors to get, or to control them (using the PS3 controller). I'm also stumped about sending and receiving audio, as well as controlling the servo with a received signal (like an analog decibel meter). Power is also something I'm not sure about. I would like to use a small 12 volt battery to power everything (with the appropriate voltage regulation).

As far as communication, I like the Bluetooth option. However, I think it may be limited long term. With the ability to transmit sound, I'm guessing that creating an intranet and sending the data over wifi might be a better option. If all goes well, I could see adding a webcam to the whole project down the road.

If anyone has any pointers, suggestions, comments, and flat out "you can't do that's", I would be greatly appreciative!

Thanks in advance!

Todd

Great idea for a project, ambitious but a rewarding venture. I think it would a good idea to break the project up into a number of steps, to keep the entire task seems less overwhelming.

Since you have already set up the PS3 controller, I think it would be good to complete this portion first i.e. determine which servo will be assigned to what on the PS3 controller then complete the code to verify that it works the way as planned.

In regards to motors, one thing to keep in mind is the weight of your robot. Ideally, you would want a motor that is at least powerful enough to carry your load. Therefore you would want a motor that has enough torque to carry the load and if you want speed, you would need a motor with high rpm. Its a compromise between the two, as motors which has high torque also has low rpm and vice versa. It would be a good starting point to first estimate the weight of your robot before you get your motors, to determine the minimum torque you would need to carry the load.

You would probably want to look at DC motors as they are common and cheap, you then have to decide on using brushed or brushless type motor. This then leads to the type of ESC (Electronic Speed Control) you would need, the brushed solution is a more economical one.

The weight of the robot and selection of motor then leads to the selection of you battery i.e. the voltage you need. You would need a battery that will give you enough volts to get you the speed you want, and the capacity to power the robot for the duration you want. You can look into LiPo batteries with high C rating as they are lighter in weight, as an option.

tdewitt274:
The rest of the body will be mounted on either a tricycle or 4 wheeled toy car (think Power Wheels).

From experience, I can tell you that using such a platform will be a challenge in itself; not because it is difficult to power such a machine - though you will need a fairly high-amperage - around 30A - h-bridge to control the drive motor(s).

The challenge is in steering it - or rather, coming up with a means to steer the wheels. That said, you can easily throw money at it and get something to work (ie - by using a high-power geared servo motor, like from Servo City - or a linear actuator from say Firgelli Automation); sometimes, even throwing money at it won't help, mainly because such ride-on toys weren't designed to be controlled in such a manner - finding mounting points for the steering actuator (or building and attaching a bracket system to hold such an actuator) can be a real brain-teasing exercise!

I can't offer you any specific advice, only a warning: Your best bet is to research how other people have solved the problem; for instance, if you decide to use a PowerWheels or similar ride-on toy, visit the Modified Power Wheels forum (http://www.modifiedpowerwheels.com/forum/) and research how people have modded theirs for R/C control and similar. There's a ton of different solutions and ideas out there; both in forums like the above, and independent designs. Because there are so many different ride-on toys, manufacturers, and designs, there generally isn't a perfect, one-size-fits-all solution to the problem (unless you are able to purchase an identical ride-on toy that someone else has modded in a similar fashion).

Lastly - and this applies mainly for any homebrew servo solution (where you build your own servomechanism and/or linear actuator) - be sure to install limit switches for end-of-travel signalling on your steering mechanism, so that the motor will stop turning once the end-stops are reached. In my own case, I managed to destroy the teeth on the output gear of one of my gearmotors by not doing this...

Thanks for the responses so far!

I agree with AiRgeek for continuing to code with the resources I have. The current setup will work, I'm looking for future expand-ability as well.

Regarding the motors, I'm looking at high torque with a relatively low speed (3-5 MPH max), able to propel 30-40 lbs (25 lbs for the "vehicle", maybe a little more with the battery), hopefully a 4 hour battery life (there will be a lot of start and stop, primary function is entertainment), and both forward and reverse capability.

Ideally, I would like to start with slow acceleration because the environment may have multiple people in the area and would avoid "high speed" collisions (especially at initial movement). I haven't worked with motors at all yet, but have some ideas. I would think that the microcontroller would help with acceleration if it's programmed correctly.

Cr0sh makes a good point on the vehicle. I think building the unit would be the best option. I'm basing the project off a previous version, so I have some ideas on how the tricycle option will work. And you're right, mounting the motors will be a challenge.

I am interested in the local internet option though so I can see if it's a good option. If anyone has any resources on this, I'd be quite interested.

I'm not totally opposed to "throwing money at it", within reason. The project alone has so many interesting pieces from electronics, sewing, fabrication, and probably a few more I haven't thought of yet! I consider it a challenge to my hobby.

As far as the motors go I would look into Ampflow or AME. Paralllax sells a kit with AME motors that gives you the wheels and mounts. As far as steering I would go for 4WD. It would make turning extremely easy.

However, I don't know which direction to go. I was wondering if someone could point me in the right direction.

Well, it will be difficult to point you in a direction beyond recommending you spend some time using google and youtube to find some bots similar to your needs and financial limits. I expect you want the most bang for your buck, but you need to state your limits and what the minimum robot performance items you can use.

tdewitt274:
I'm not totally opposed to "throwing money at it", within reason.

When I mentioned "throwing money at at" - I was meaning spending a couple of hundred dollars on a high-torque actuator (whether that was a large geared servo actuator from Servo City, or a beefy linear actuator from Firgelli); maybe that is in your budge, maybe not.

Either one could have solved my particular steering issue, but I didn't feel like spending such an amount for one sub-system in my design (which is based on an old "Little Tikes" H2 ride-on toy) - I ended up coming up with a solution using a drill motor gearbox and a short piece of all-thread to make my own linear actuator system (also as a part of that I created an optical-interruptor to count rotations for position information, as well as limit switches for travel).

Others have come up with different solutions (such as hooking an h-bridge to the outputs of a servo, to control a windshield wiper motor, which moves a lever connected to the servo control arm and potentiometer to complete the feedback loop - basically a hacked method of making a large "power servo").

These and other solutions are much cheaper, but in my case I had to think long and hard about the problem (which had actually been staring me in the face) until I found the solution that fit in the confines of the space I was working in.

solution that fit in the confines of the space I was working in

@tdewitt274:
This is a common problem and I suspect you might encounter this as well, this has been an issue plaguing me in most of my builds.....