Remote controlled car robot

Hello all

I am new to arduino and I just ordered my mega board. One of the reasons that made me buy it is because of this useful forum. It seems that I posted this topic in the wrong section so I moved it to here, Sorry about that.

I just started working on a project that is about a skid-steering robot (20kg) that is controlled remotely over long ranges (1km). This robot will have live video streaming and by reading many discussions on this forum and on the internet it was advised to buy readymade wireless surveillance cameras. Therefore I am going to buy a camera, 1.2GHz AV transmitter and receiver. Where I live, anything below 2.4GHz is legal and no license is required

Also, this robot will have sensors (eg. GPS, temperature, vibration, noise, gas sensors) and the values should be sent to the operator of the robot wirelessly. This robot should have two way communications, one way is to control the movement of the robot and the 2nd is to send back the sensor values. In addition, in this project I need two motors and they need to be controlled separately (for skid steering). To control the robot movement and after doing a lot of research, it ended on two possibilities: -2 1600oz/in Nema34 stepper motors and two drivers -2 350W brushless dc motors (similar to the motors in E-scooters) and a Sabertooth dual 25A motor driver (one driver can control two motors) please keep in mind that I cannot get these motors and drivers where I live and I will have to order 3 motors and an extra driver just in case one of them failed and it takes about one month to be delivered. The total cost with the shipping was about 600$ for stepper motors and 480$ for DC motors.

My questions are: -Is buying the readymade live video streaming cam the best solution? -Should I use brushless motors or there are certain advantages for stepper motors? I know its more precise but its not really an issue in the project.

-I found this chip online: XBee-PRO® 868 OEM RF Modules. Will it cause any interference if my country uses 900MHz as its GSM frequency?

-Is it better to use two separate wireless chips if I want it to be fast?

-What is the best board to use in this case?

So please tell me what do you think about it and provide me with any ideas/comments. I never used GPS or wireless before so please, any help will really be appreciated.

Thanks in advance

[edit]NOTE: This was a reply I was editing for your -first- post, before you moved it; I understand why you moved it, but at the same time I could have lost my book-sized reply below, and that would've really ticked me off, as I am looking for an agent and publisher, you see...

;) [/edit]

Hello all

Hello.

I am new to arduino and I just ordered my mega board. One of the reasons that made me buy it is because of this useful forum.

Cool. And welcome. Glad you found the forums useful!

I just started working on a project that is about a skid-steering robot (20kg) that is controlled remotely over long ranges (1km).

What is your experience in mechatronics? This project is anything but trivial; I trust you understand that?

This robot will have live video streaming and by reading many discussions on this forum and on the internet it was adviced to buy readymade wireless surveilance cameras. Therefore I am going to buy a camera, 1.2GHz AV transmitter and receiver. Is that the best solution?

Its one possible solution.

In addition, in this project I need two motors and they need to be controlled seperately (for skid steering). To control the robot movement and after doing a lot of research, it ended on two possibilities:

-2 1600oz/in Nema34 stepper motors and two drivers -2 350W brushless dc motors (similar to the motors in E-scooters) and a Sabertooth dual 25A motor driver (one driver can control two motors) please keep in mind that I cannot get these motors and drivers where I live and I will have to order 3 motors and an extra driver just in case one of them failed and it takes about one month to be delivered. The total cost with the shipping was about 600$ for stepper motors and 480$ for DC motors. Should I use brushless motors or there are certain advantages for stepper motors? I know its more precise but its not really an issue in the project.

That's a lot of money! Those are also pretty powerful motors; you should know that you are going to need a fairly hefty power supply for that system (which is going add weight). Either motor system will probably be adequate, but you are likely going to need a gearing/transmission system to get the power to the wheels (you -do not- want to attach the wheels directly to the shafts without some form of indirect coupling; you are likely going to need at least a small amount of gear-down to provide the torque needed once the rest of the robot's weight is factored in anyway - besides, you need to keep the weight off the motor bearings, as they aren't designed for it).

Also, this robot will have sensors (eg. GPS, temperature, vibration, noise, gas sensors) and the values should be sent to the operator of the robot wirelessly. This robot should have two way communication, one way is to control the movement of the robot and the 2nd is to send back the sensor values. I found this chip online: XBee-PRO® 868 OEM RF Modules. Will it cause any interference if my country uses 900MHz as its GSM frequency? Is it better to use two seperate wireless chips if I want it to be fast?

I'm doing a similar project, however, I am using an existing motorized platform for my (beta) system that is already tried and tested to withstand the loading and abuse - a PowerWheels H2 ride-on toy:

http://www.phoenixgarage.org/show_article/114

Anyhow - in this project, my Arduino (ultimately, it will be a standalone unit in the final form - once I get there) will act as a controller for a custom steering servomechanism (unlike your design, this vehicle uses Ackerman steering), as well as for control of the drive motors. It will also relay commands for a pan/tilt unit controlled by a Pololu Serial Servo Controller.

An on-board Micro-ITX based PC, running Linux (Ubuntu, most likely) - will be used to send commands to the Arduino for steering, pan/tilt, and drive control:

http://www.phoenixgarage.org/show_article/78

This server also will handle the input and digitization (for machine vision as well as human control) of a USB camera, and communication with the system will be done via an encrypted (WEP and SSL), MAC-addressed locked, 802.11g link via a regular router. The server will run a LAMP (Linux, Apache, MySQL, PHP) stack, along with Python and OpenCV, and anything else that is needed (there's 2 GB of memory and 500 GB of hard drive space).

The Arduino will be custom programmed with a system to not only take in commands and such from the server, but also to include fail-safe mechanisms such that should something break and a "run-away", "unsafe", or "failure" mode occur, it will automatically detect this and shut the system down. That's the plan, anyhow. You should implement a similar system as well.

What's your chassis going to look like? Bolted or welded construction? Tubular steel? Aluminum? Something else?

I went with this chassis because, like I said, it was designed for abuse (ie, reckless 3-4 year olds!), it could carry the weight (I rode on it once, I weigh 240 lbs), it was "pre-built" (so no difficult design work coming up with a frame or such), and it was fairly cheap (bought used off of Craigslist). Plus there's already a large modder/hacking community for these ride-ons:

http://www.modifiedpowerwheels.com/

At a later point, if funds permit, I have a plan to go with either a custom framed differential steered system (like you're planning - I already have the motors needed, designed to run a large mobility chair), or maybe something like a Polaris off-road vehicle.

For your power source, you might look into a hybrid system, using a small camping generator wired to charge a deep-cycle battery (12 or 24 volts), then power your system off of that. Some of the small Honda gensets are extremely quiet (though I have heard good things about the smaller Harbor Freight gensets that work well after a few minor mods - plus they're cheaper!).

Finally - I sincerely hope this is not your first project of this nature; I really hope that you have experience in other areas that are similar to this nature, and have given a lot of thought to how such a robot can be constructed. 20 Kg doesn't sound like a lot, but once it gets moving...

Good luck, and let us know how things turn out!

:)

BTW - I hope the above doesn’t come across too much as “blowing my own horn”; that isn’t my intent. My intent is to potentially show you a method of approaching the project, perhaps there is something you can learn from (good or bad). I mean to share how I am doing things, potentially to help others, such as yourself.

:slight_smile:

WoW, I did not expect such an informative response this fast! Thank you cr0sh!

What is your experience in mechatronics? This project is anything but trivial; I trust you understand that?

I understand the difficulty of this project. To make things clear, I am working with a colleague who is a mechanical engineer. My colleague will be responsible of the mechanical configuration of the project.

I'm doing a similar project, however, I am using an existing motorized platform for my (beta) system that is already tried and tested to withstand the loading and abuse - a PowerWheels H2 ride-on toy:

Thank you for sharing this! I'm gonna read through this

What's your chassis going to look like? Bolted or welded construction? Tubular steel? Aluminum? Something else?

My chassis is an originally welded tubular steel four wheeled bicycle but it is reasonably durable.

I went with this chassis because, like I said, it was designed for abuse (ie, reckless 3-4 year olds!), it could carry the weight (I rode on it once, I weigh 240 lbs), it was "pre-built" (so no difficult design work coming up with a frame or such), and it was fairly cheap (bought used off of Craigslist). Plus there's already a large modder/hacking community for these ride-ons:

I would have done the same but such chassis are very expensive where I live. There are some with reasonable prices but are very fragile and break easily.

Finally - I sincerely hope this is not your first project of this nature; I really hope that you have experience in other areas that are similar to this nature, and have given a lot of thought to how such a robot can be constructed. 20 Kg doesn't sound like a lot, but once it gets moving

I have worked with small robots and motors before. I'm not really worried about the mechanical structure or the movement of the robot. My deepest worry is in the two-way wireless communication of the robot. It is my first time.

Thank you again for your response and I am going to go through your project. I really appreciate your help.

Before you spend $$$ you need to have tested all the aspects of your project. You might want to look into using wifi routers to do your communications. They can handle the wireless connection, including streaming video. I’m not sure about getting 1 km range, but they would be an inexpensive way to start testing, and some can have very long line of sight range. If gas powered riding lawn mowers are available where you live, you might consider modifying one for your platform (can be put back to origional and sold at the end of the project), as they can carry heavy loads and operate for many hours. An electric start one would have a battery/alternator on it. Below is a simple wireless arduino router bot test setup that is operated from a simple web page. Bottom is a high powered router that with hi-gain or directional antennas might make your desired range.

http://www.radiolabs.com/products/wireless/networking/1-watt-router.php

http://www.radiolabs.com/products/wireless/wireless-2.html

http://www.radiolabs.com/products/antennas/long-range-wifi-antenna.html

Thank you zoomkat for your response

If gas powered riding lawn mowers are available where you live, you might consider modifying one for your platform

I never thought of that.. Can the speed be controlled using an arduino? Please keep in mind that I need two of them (for skid/Ackerman steering) and the robot is approximately 1m long and 0.5m wide. From the pictures I saw their size is somehow big.. isn't it? Also, the use of gas powered engines might be an issue in this specific project becuase of some strict safety rules in certain areas (eg. Oil fields).

You might want to look into using wifi routers to do your communications. They can handle the wireless connection, including streaming video.

Thank you, this is a very good idea. However, this project might be used in remote areas and an internet connection might not be available (even on the controller's side), won't that be a problem? What about using two-way communication using XBee? Is it practical?

I'm still interested in using wifi routers. I went through the links you provided, I might do another prototype using this method. With the availability of internet, the robot then could be controlled from anywhere in the world. Unfortunately, I have no experience at all in such fields. Can they be used for two-way communication? Where can I find more technical details on using this method?

Thanks

I never thought of that… Can the speed be controlled using an arduino? Please keep in mind that I need two of them (for skid/Ackerman steering) and the robot is approximately 1m long and 0.5m wide. From the pictures I saw their size is somehow big… isn’t it? Also, the use of gas powered engines might be an issue in this specific project becuase of some strict safety rules in certain areas (eg. Oil fields).

From your description sounds like you need a smaller bot, which normally easier to make using electric motors. Do note that anything used in potentially explosive atmospheres can be very complex to design.

Thank you, this is a very good idea. However, this project might be used in remote areas and an internet connection might not be available (even on the controller’s side), won’t that be a problem? What about using two-way communication using XBee? Is it practical?

Routers in the bridging mode do not require any internet connection. The setup is usually pc <> router <> router <> bot. I’m not that familiar with the xbee capability, but I doubt that they can easily support streaming video that a router can using attached IP cams.

I’m still interested in using wifi routers. I went through the links you provided, I might do another prototype using this method. With the availability of internet, the robot then could be controlled from anywhere in the world. Unfortunately, I have no experience at all in such fields. Can they be used for two-way communication? Where can I find more technical details on using this method?

The router approach is pretty much off the shelf, plug ang go, no modifications required. In the past I tinkered with putting a wireless router up on an antenna mast to be able to use my laptop at extended distances outdoors. You can get the high powered routers and hi-gain antennas and do some experimentation. If they don’t work as desired, you can sell them and recover some of your expenses as they are not hacked or modified. For more info, Google is your friend!

http://www.unwiredadventures.com/unwire/2005/12/defcon_wifi_sho.html