I wanted to share a project with you guys that I am building. This was my first project using Lynxmotion products and I have to say that they are very nicely made and packaged.
Anyway…the project is a remotely driven r/c car with a twist. Instead looking at the car and driving it around from a distance like you would normally drive an r/c car, you get a view from inside the drivers seat!
But I know what you are thinking, “that’s been done before”. Well, not like this! A PC steering wheel and pedal system have been setup for control of the vehicle. All steering and pedal functions behave just as they would in a real car(automatic) except for the fact that I have the break pedal setup as reverse, but it is just temporary.
To take the project a step further I installed a Lynxmotion pan/tilt camera base on the car and I mounted an inexpensive wireless camera on it. Video currently is sent from the camera back into a video capture card on my PC because that’s where I have my wheel/pedals setup. It could be just as easily viewed on a big screen TV or any input with RCA connections though. The camera pan is also synchronized with the turning of the wheel. When the wheels turn right, the camera moves right also.
I have used a 4 channel Futaba 4YF radio to control the steering, throttle, pan, and tilt. I use a PCTx interface to tie the transmitter into my PC. The r/c car that I am using is a Tamiya Hummer if anyone was interested.
I know it is rude not to post pictures but I will make a video and take some pictures later tonight.
My setup does use a steering wheel and pedal system. That video is a demo when I had the car setup to steer and look via the mouse. I found the car 10x easier to drive with the wheel and pedals!
I’m using a PCTx interface to connect the transmitter into my PC via the trainer port. The software is my own creation, written in C++.
I just took this picture:
This is a previous setup that I tried. Driving with the camera towards the front was much easier. Looks cooler this way though…
I was planning to do that as an add-on feature later down the road. I thought it would be cool to drive around the house when I was away and check on things. This is where the mouse/keyboard driving would come in handy.
Only problem is battery life though. I’m not sure if the r/c battery pack would last a whole day even on standby. I’m also running the camera off a throw-away 9v battery so I would need to intergerate it into the car’s power source.
I guess I could make some type of charging base I could just drive up on, but I would need to turn the car off in order to charge it. Anyone have any ideas?
Here is a short video I made when the control was done via the mouse and keyboard and before I got my new aluminum pan/tilt from lynxmotion. I hope it is ok to post a link to this video here, if not let me know and I will remove it. endurance-rc.com/media/car.wmv
I’ll be designing something similar to get my bots to charge themselves.
AFAIK, MOSFETs are the way to go.
Pete can probably point you in the right direction better than I can, though…
The link to the car.wmv doesn’t seem to work for me. You could make a simple “refueling” probe on the front of the car. A low voltage drop diode and an LED on the end of the probe would indicate when the probe had been successfully connected to the jack, which would supply the charging power (you could see the LED light via the cam).
Hold on…
So you’re saying that a diode in series with the robot’s external battery input will seperate the battery charger’s input from the robot’s electronics?
Somehow that seems too good to be true…
I was thinking that something like a set of MOSFET switches would be necessary to detect an input volatge (i.e. the charger) and cut off the electronics supply.
Thanks! I have to admit that I’m really not farmilar with video editing and I could have had better lighting but I’ll get better
Stage 3 of the project will begin in a few days. I have ordered a set of video goggles from www.3dvisor.com. They will be used to display the video from the wireless camera and also move the pan/tilt base. I’m working on a robotics project using the RoboRealm software. I made the pan/tilt track a red ball. I’d like to get more projects based around the controller though. If anyone has any ideas lay em out!
Awesome job! Like the YouTube, I can view those since they are flash. Just to let you know, the videos are in some newer .wvms so I can’t play 'em directly (Linux) and other people (Mac, older versions of windows) might have issues too.
Nick, about the diode, I think its meant as a protection mechanism as I would imagine it would be easy to connect the power supply incorrectly if you were doing it by driving remotely.
Personally, I would love to see a magnetic charger like they have on the new Macs used here. It would guide itself into the charger making docking easier. A diode would still be a good idea for protection, even with this. I’m not sure if you’d want to use a real diode rated for the 1-2 A you probably will be charging at, or use the ideal diode circuit saipan59 brought up in the PROTEUS-RVI thread. My guess is just the diode, as it will be less efficient, but simpler and cheaper to make.
If the car has appropriate regulators inside and the charger can source enough amps (though the car really won’t be doing much when docked), it could just go in parallel with the pack, powering the car while plugged in and charging the battery. I’m assuming the charger could be connected a PC so you could start a charging cycle remotely (assuming web remote controlled). Alternatively you could rig it so there was some way the car could trigger the cycle to start (i.e. a big button that could be hit by the pan and tilt for example).
Just been looking at an RC toy car belonging to my nephew (yes i play with too). It has a charging dock which has a pair of sprung feelers that connect to the car via two flat plates under the chassis, theres a bridge rectifier on board to prevent reverse connection.
Tillin9:
I’m running SuSe Linux 10 (2.6), play my video on Xine with the win32 codec pack installed, plays everything.
Hmm… Last I recall the codecs were wrapped Windows x86 binaries. Being illegal to install (unless you have a legal copy of XP) Debian doesn’t package them with their distro. If this has changed upstream, great. However, the latest win32 codecs available for Debian Etch (my distro) used the wrapper trick, making them a no-no at work. Yes, at home since my machine dual boots, I do have them installed (actually pulling the .dlls from my NTFS partition). Even so, a good deal of people on Linux (i.e. they can’t - illegal or non-x86 - or don’t know how to install the codecs) or any Mac (I’m fairly sure there is no Mac support) can’t watch them.
Sorry if I seem to be making a big deal out of this, I’m really not. I actually was very glad BotHakr had been considerate enough to put the files on YouTube. However, when web publishing, picking a more universal video codec is usually a good idea.
Is it safe to charge an r/c battery pack when the car is still turned on? I’ve never tried this myslef. I know of people that charge their packs while they are still plugged in but not when the ESC is still powered on.
Yes, unless any of the more hardcore EE guys here see something I’m missing, I don’t see a reason why one can’t run a the bot off the charger while charging. I mean many devices, like Laptops do it all the time without ill effect. Basically there are a few things you need to think about when designing this:
The first is that you don’t overvolt anything. Since when charging there will be a 2-3V higher power source, you need to be sure your gear can handle that. For most motors and servos, this is a non-issue and they can safely run at higher voltage without any problems. For electronics, this can be an issue if they are not regulated. However, since electronics don’t run off 6.0 or 7.2V (they generally use 5 or 3.3V) they should have regulators in them already. Nearly all regulators can handle a wide range of inputs. You should check the documentation of your equipment as to what its tolerances are just to be safe, or better yet find the actual chips on the devices (for example common regulators are the 780x series) and look up the spec sheets. But I’m relatively confident that your equipment can handle it.
The second is current draw. The issue here is you might overload and damage your charger. So make sure you have more than enough leyway. I’m assuming your electronics might draw around 250 mA even at rest, and you might need to charge at 1C, with say a 1600 mAH capacity pack so as long as your charger could provide 1.85 A safely, you’d be fine. For comparison the Lynx LN-5014 can do 5 A. Most chargers can do at least 3 (and you probably never actually have to charge at 1C for NiMH), so you’d probably be fine without having to build/ buy anything.
The only other possible issue is that you could cause the total voltage to drop below what is necessary to charge the pack causing it not to charge or much more likely, charge more slowly. This last condition might be a bigger issue with Li-Pos since the exact nature of the V/I curves during the charging cycle is more important to prolonged battery health. However besides speed it really shouldn’t matter if you charge your NiMH pack at 1.0C, 0.9C, or even 0.5C for that matter.
I am assuming you are talking NiMH or NiCad chemistry batteries. The standard charger closely monitors the voltage to determine when it has peaked. This is done to prevent overcharging. Delta Peak chargers work like this. Apply a voltage to provide X amount of current, wait a bit, check the voltage again, did the voltage go up a little? if yes carry on, if not, we’re done here. That’s a simplification of course, but I see no way to charge these packs while they are being used. I suspect that laptop supplies are charging when in use, but also powering the thing separately. If you design something from the ground up I’m sure something could be made to work though. You just can’t use an off the shelf charger. Um, I think this is accurate.
I had not considered the peak charger. I will be using NiMH batteries in the r/c car.
Maybe we are going the wrong way with this by designing some kind of circuit. Maybe a mechanical solution would be easier?
For instance maybe the power switch for the ESC could be mounted somewhere on the car. Contacts running from the batter could be mounted somewhere on the front so that contact could be made when the vehicle drives up, like what Len_Fixit was describing. Once the car is in position and servo could flip the power switch, and a relay could activate the charger. When the charge is complete a servo could flip the switch and the car would be free to drive off. The charge could either be activated from the web or automatic. Good thing is you could drop in any charger or onboard r/c gear and not have to worry about it being compatible.
That video is a demo when I had the car setup to steer and look via the mouse. I found the car 10x easier to drive with the wheel and pedals!
or NiCad 
chemistry batteries. The standard charger closely monitors the voltage to determine when it has peaked. This is done to prevent overcharging. Delta Peak chargers work like this. Apply a voltage to provide X amount of current, wait a bit, check the voltage again, did the voltage go up a little? if yes carry on, if not, we’re done here. That’s a simplification of course, but I see no way to charge these packs while they are being used. I suspect that laptop supplies are charging when in use, but also powering the thing separately. If you design something from the ground up I’m sure something could be made to work though. You just can’t use an off the shelf charger. Um, I think this is accurate. 