RoboRealm Forum

Java API [2]

Sat, 13 Mar 2010 11:03:51 EST

I am using your API for java... and i am not able to connect it to roborealm its throws exception could not connect to local host 5050 !... can u give some suggestion on how to write initial connection code using RR_API.java and XML.java

Image Processing [6]

Thu, 11 Mar 2010 08:03:43 EST

This is great! what can i say? THANK YOU! works great!!!

Image Processing [5]

Thu, 11 Mar 2010 02:03:00 EST

Most of this can be done using the Blob Filter.

We use the auto-threshold to ensure that we are working with only binary object (something you may not need to do if the above images are a result of other filters).

Then in the blob filter we select objects larger than 100 pixels, whose height is about at or above 20% of image height and whose perimiter / area ratio is above 30% (the last one removes large rounder objects).

This set seems to work on the images you sent ... naturally we would recommend testing on many more images to ensure that those attributes are the right ones.

Robofile attached ...

STeven.

Image Processing [4]

Wed, 10 Mar 2010 20:03:29 EST

Thanks very much STeven, here are the frames (5 pictures 160x120 pixels). The very first on the list is for illustration of what I want to achieve. The regions I mark with the red line across are the parts I don't want. Perhaps I will be enlightened when reply.

Image Processing [3]

Wed, 10 Mar 2010 19:03:42 EST

Christian,

Yes, that's basically what RoboRealm does. However, you can't just tell it "Please remove the clouds" but instead tell it remove all white objects that are in this approximate location in the image. Or more specifically look for light objects that appear straight are well defined and are located in the lower part of the image ... and remove everything else. (I.e. focus on the white lines). There are always more than one way to accomplish the task.

If you attach the image here we can send back a robofile that will accomplish this task for you to play with.

STeven.

Image Processing [2]

Wed, 10 Mar 2010 05:03:32 EST

Ok! i have a frame captured by my robot (this frame is a road picture of black & white binary pixels).

My question is, is there a tool in RR that let you get rid of the parts i am not interested in on the frame? something like cloud romoval, or just cutting off some part of the frame? i basically just want to see the white lines of the road.

Any help will be well appreciated.

Christian

Version 2.18 [7]

Wed, 10 Mar 2010 05:03:19 EST

That corrected the problem, this is awesome software, thank you

Version 2.18 [6]

Tue, 9 Mar 2010 20:03:19 EST

Robbie,

Can you download ver 2.18.1 and see if that fixes the issue?

Seems that an interface ID was changed that caused a load issue on the GUI interface which was incorrectly interpreted as a DLL error.

Thanks for your note!
STeven.

Version 2.18 [5]

Tue, 9 Mar 2010 20:03:01 EST

It is a generic error, also the 888 was working with version 2.17, I have phidget21 in my system directory

Version 2.18 [4]

Tue, 9 Mar 2010 20:03:27 EST

Opps, ignore that last post (except the part about keeping the code up-to-date). It looks like we have a small issue that may be creating the problem. We'll get that addressed shortly ... please wait for the next post.

STeven.

Version 2.18 [3]

Tue, 9 Mar 2010 20:03:50 EST

Which function did it say was missing? Or was it just a generic error?

Also, could you look in your

c:\\windows\\system32\\

folder and see if a phidget21.dll exists in there?

If you only have a Phidget.dll or a Phidget20.dll you should download the latest driver code from Phidgets and install. That may fix the issue.

STeven.

Version 2.18 [2]

Tue, 9 Mar 2010 19:03:50 EST

I downloaded the latest version, and my Phidget 888 quit working, It said i did not have correct Dll loaded

A question about the LAB_Channel module [4]

Tue, 9 Mar 2010 09:03:42 EST

Thanks STeven--that's exactly what I needed. As for the ANN stuff, for the moment I am just using a Perceptron with the Delta Rule but I have Backprop waiting in the wings in case I need a multi-layered network. I've been using the open source AForge.NET package from http://www.aforgenet.com/ which is really nicely written and easy to use (as you may already know!).

--patrick

Wall socket detection and data. [3]

Mon, 8 Mar 2010 17:03:55 EST

Replying to my own post!
I've been looking at getting the socket recognised as a Fiducial which works very well but only if I draw a black line round the socket. I can't seam to explain to the software that i'm looking for the three holes. The contrast of the holes is really good though so they are very easy to pick out. The software wants to analyse each whole as a fiducial and draws a nice purple rectangle round each one. Looks promising though...

RR API [6]

Mon, 8 Mar 2010 15:03:03 EST

sorted! this where to put the ip of the remote machine
% connect to RoboRealm (assumes API is on and RR is running)
rr.connect('192.???.?.???');

Thanks

RR API [5]

Mon, 8 Mar 2010 15:03:17 EST

No edit option??

Thanks again STeven,

Matlab seems to complain on call to

RoboRealm('connect', 'ip_address_of_remote_machine', 6060); it throws an error ??? Undefined function or method 'RoboRealm' for input
arguments of type 'char'.

Error in ==> RoboRealm_test at 9
RoboRealm('open', '192.???. ?. ???', 6060);

what should i do? please

RR API [4]

Mon, 8 Mar 2010 15:03:47 EST

thanks again STeven, matlab seam not to complain on call to

RoboRealm('connect', 'ip_address_of_remote_machine', 6060); it throws an error ??? Undefined function or method 'RoboRealm' for input
arguments of type 'char'.

Error in ==> RoboRealm_test at 9
RoboRealm('open', '192.???. ?. ???', 6060);

Wall socket detection and data. [2]

Mon, 8 Mar 2010 13:03:54 EST

Hi,
This is a rather general question. I need some software that is capable of detecting a UK 3pin wall socket and providing distance and angle data. The idea is that my robot will be able to plug itself in for charging. I'm considering having a dedicated camera just for this purpose. The camera will possibly be mounted in the plug to aid alignment. The wall socket is a unique pattern of three dark rectangles, the socket will always be upright and vertically aligned with the camera. What I want ideally is software that will take my video stream and just produce a few values: Yes/No is a socket visible. Approx range/scale. Object offset from camera centre. Approx angle in relation to camera.
I am working my way through some of the roborelm tutorials but I wonder if anyone on here has done anything similar?

robot control using controller with 1 analog stick [3]

Mon, 8 Mar 2010 05:03:30 EST

Paddy,

I think we may have this in some of the tutorials but basically it will go something like this

speed = 140 ' this is the forward speed that keeps it moving
max_turn = 128 ' this is the max turn speed

turn_speed = (max_turn * (joy_x / 1000))
motor_left = speed - turn_speed
motor_right = speed + turn_speed

assuming you get the joy_x value from your joystick that ranges from -1000 to 1000.

The idea is to scale the amount of joy_x into the max turn value. So when fully left or right the turn speed will basically set one wheel to 0 speed which indicates a maximum

You could use joy_y for the actual speed value (instead of the hardcoded 140) using the same technique.

STeven.

A question about the LAB_Channel module [3]

Mon, 8 Mar 2010 05:03:47 EST

The a and b values are scaled to a range of 255 and moved to 128. I.e if the value you get from a or b channel is 128 that is actually 0, 120 would then be -8 and 135 would be +7. Thus you can just do a - 128 to get the negative value back.

What kind of ANN are you using? Backprop?

STeven.

RR API [3]

Mon, 8 Mar 2010 05:03:09 EST

Try

RoboRealm('connect', 'ip_address_of_remote_machine', 6060);

instead of the 'open' function which tries to launch RR on the current local machine. The connect assumes RR is running on the remote machine as it is not able to actually start it on the remote machine ...

STeven.

RR API [2]

Mon, 8 Mar 2010 04:03:48 EST

Hi, I got the rr API working. I am working with matlab and RR on an AI final year project. Currently the API lets you connect to RR locally within one computer when RR is running, but is it posible to connect to RR via API from another machine? I want to be able to do this from matlab running on another computer?

The reason is that the computer on the robot has a very low processing power(intended to be so due to its low weight) therefore slows really down when matlab is executing. So i decided to use it just for image processing with RR API running and listeneing on the defualt port. I therefore want to be able to get image from RR on this low computer to matlb running on my main pc.
Any help will be appreciated.

Christian

A question about the LAB_Channel module [2]

Sat, 6 Mar 2010 20:03:53 EST

Hello,

I need to use the L, a and b values from the LAB_Channel module as inputs to an artificial neural network. As I understand it, the a and b values can take on both negative and positive values. Does the RoboRealm API return the negative values if I do a getImage() after applying the LAB_Channel module? Putting it another way, how can I get the negative values as well as the positive values?

Thanks!
patrick

how to convert pixel data from pipe into opencv IplImage [4]

Sat, 6 Mar 2010 11:03:54 EST

Ooops.  That first method should have returned a byte array.  The correction is:

public byte[] getImageBytes(string marker)
{
   int width = 1280;
   int height = 960;

   byte[] imageData = new byte[width * height];

   Dimension d = null;

   while (d == null)
   {
      d = RoboRealm.getImage(marker, imageData, width * height);
      RoboRealm.waitImage(1);
   }

   return imageData;
}

how to convert pixel data from pipe into opencv IplImage [3]

Sat, 6 Mar 2010 11:03:21 EST

Hi Scott,

I don't know if this will help, but I got the following pair of methods many moons ago from this forum and they work for me in C#.  It converts the image byte array obtained from the RoboRealm API to a bitmap.  Perhaps from there you could convert it into the OpenCV format (but I'm not familiar with that).  This is in C# but you can probably easily translated it into C++.

--patrick

        // This method grabs the image bytes for one frame from RoboRealm
        // using the getImage() method and the marker name.
        public void getImageBytes(string marker)
        {
            int width = 1280;
            int height = 960;

            byte[] imageData = new byte[width * height];

            Dimension d = null;

            while (d == null)
            {
                d =RoboRealm.getImage(marker, imageData, width * height);
                RoboRealm.waitImage(1);
            }
        }

        // Create RGB24 bitmap from Byte array
        public Bitmap BitsToBitmapRGB24(Byte[] bytes, int width, int height)
        {
            //swap RGB to BGR
            Byte tmp;
            for (int x = 3; x < bytes.GetLength(0); x += 3)
            {
                tmp = bytes[(x + 2)];
                bytes[x + 2] = bytes[x];
                bytes[x] = tmp;
            }

            if (bytes.GetLength(0) < width * height * 3)
            {
                return null;
            }

            Bitmap bmp = new Bitmap(width, height, PixelFormat.Format24bppRgb);
            int i;

            BitmapData data = bmp.LockBits(new Rectangle(0, 0, bmp.Width, bmp.Height), ImageLockMode.WriteOnly, bmp.PixelFormat);

            if (data.Stride == width * 3)
            {
                Marshal.Copy(bytes, 0, data.Scan0, width * height * 3);
            }
            else
            {
                for (i = 0; i < bmp.Height; i++)
                {
                    IntPtr p = new IntPtr(data.Scan0.ToInt32() + data.Stride * i);
                    Marshal.Copy(bytes, i * bmp.Width * 3, p, bmp.Width * 3);
                }
            }

            bmp.UnlockBits(data);

            return bmp;
        }

robot control using controller with 1 analog stick [2]

Sat, 6 Mar 2010 11:03:17 EST

i need help trying to control a robot with a controller that has only 1 analog stick, if the controller is pushed to the right the robots motors have to turn it right and the further right its pushed the faster it turns, the same goes when your turing left forward and reverse. the module has to send out signals from 0-255 to control each of the servos that control the movement of the robot. i have to use vbscript to do this n i know its only simple maths to do this but i just cant figure it out

how to convert pixel data from pipe into opencv IplImage [2]

Fri, 5 Mar 2010 02:03:14 EST

Hi, this is my first post. I am successfully piping image data from from my wireless linksys internet cam into you pipe program example in C++. I am then trying to run face recongition software on the stream, but to do so I need to convert the pixel stream into an opencv IplImage. Do you have any example code for how to convert the piped pixel stream into an IplImage type in opencv?
Thanks in advance for the help with my first post!
Best Regards,
-Scott

COG_BOX_SIZE = 0 when box display = 100% [4]

Wed, 3 Mar 2010 13:03:32 EST

Hi STeven,

Thanks for the quick fix! Also, when using the new RR 2.16, I can no longer reproduce the other issue regarding Display_Variables so I think it is safe to ignore for now.

--patrick

Robotic Eye-Hand Coordination using RoboRealm [2]

Tue, 2 Mar 2010 17:03:41 EST

Hello,

I just finished up some work on using RoboRealm to guide my robot as it reaches toward a target object. The ultimate goal is for the robot to be able to pick up the object from a random location or take it from someone's hands. For now, I simply wanted to work out the coordinate transformations from visual space to arm space to get the two hands to point in the right direction as the target is moved about. The following video shows the results so far:

http://www.youtube.com/watch?v=m8GthdGApz8

I don't have a full write-up yet on how I did this but it basically just uses 3-d coordinate transformations from the head angles and distance to the target (as measured by sonar and IR sensors mounted near the camera lens) to a frame of reference attached to each shoulder joint. The Dynamixel AX-12 servos are nice for this application since they can be queried for their current position info. The distance to the balloon as measured by the sonar and IR sensors is a little hit and miss and I think I'd get better performance using stereo vision instead.

--patrick

http://www.pirobot.org

COG_BOX_SIZE = 0 when box display = 100% [3]

Mon, 1 Mar 2010 01:03:00 EST

Patrick,

Thanks ... the first has been fixed in 2.16 just uploaded.

Can you send a robofile with the Display_variable that indicates the behavior you mention? We realize that it will probably work for us since it will have been saved but we were unable to replicate the issue and having the robofile will help.

Thanks!
STeven.

Blob filter [5]

Sun, 28 Feb 2010 22:02:32 EST

Can u suggest me similar type of software available for mac or any of the java api's available which performs the same function, if you know any of them... thanks

Nachiket

Blob filter [4]

Sun, 28 Feb 2010 22:02:18 EST

Hi steven please find the robo file which i was trying to create and the image file representing how the image will look like when it has white checkers piece on blue square... i want to find all the blobs which has same look...

Camera oscillation [2]

Sun, 28 Feb 2010 20:02:01 EST

I've got a servo mounted webcam following a blue object. To get a quicker response I set the 'Min Processing FPS' to 30. Now my camera oscillates back and forth when the object is in the center of frame. Is there a way to dampen the signal around center?

Thanks,

Dave

Blob filter [3]

Sun, 28 Feb 2010 19:02:00 EST

Can you post the image that you are testing with? And perhaps the robofile that you tried but did not work?

We can then test things out on our side and offer a possible solution. Its hard to do without a sample image.

Thanks,
STeven.

COG_BOX_SIZE = 0 when box display = 100% [2]

Sun, 28 Feb 2010 10:02:28 EST

I just noticed a bug in RR 2.14.15. If you use the COG module and set the displayed box size to 100%, then the COG_BOX_SIZE variable is set to 0 no matter what the actual size is. Setting the box size to any lower value (e.g. 90%) restores COG_BOX_SIZE to an appropriate number.

A lesser bug: if you add the Display_Variables module to an existing pipeline, the chosen variables don't display until after you close and restart RoboRealm.

Thanks!
patrick

Blob filter [2]

Sat, 27 Feb 2010 14:02:51 EST

I am trying to locate circular checker piece located inside a square or box on checkers board. I am trying to use blob filter but its not working... do you guys suggest any other method or something else... i am using "within" and "around" along with specified object say "object 1"

Future support for Minoru 3D webcam? [7]

Thu, 25 Feb 2010 13:02:15 EST

Hi STeven,

Thanks for the update. I look forward to the new Stereo module when the time comes!

--patrick

Using "Eliminate Global Flow" in the Optical_Flow Module [12]

Thu, 25 Feb 2010 11:02:10 EST

Hi Prof Mason,

Thanks to your tip about the robosavvy group, I finally solved my long standing timeout issue regarding the AX-12's.   A group member there suggested I catch the timeout, flush the serial buffer, then resend the instruction.   It took me awhile to figure out where to do this in the Forest Moon Dynamixel source code but I finally made it work.  I can even unplug a cable and have the robot recover gracefully when I plug it back in.  In case anyone else is using the Forest Moon library, here is the modification I made to the ReadData routine:

protected byte[] ReadData(int id, Register startAddress, int count)
{
   List<byte> cmd = new List<byte>();
   byte[] returnPacket = null;
   bool retry = true;

   // the start address and count form the parameters for the command
   // packet
   cmd.Add((byte)startAddress);
   cmd.Add((byte)count);
   while (retry)
   {
       WriteInstruction(id, Instruction.ReadData, cmd);
       try
      {
         returnPacket = ReadPacket(id, count);
         retry = false;
      }
      catch (System.TimeoutException)
      {
          Console.WriteLine("TIME OUT accessing ID: " + id);
          Stream.Flush();
      }
   }
   return returnPacket;
}

--patrick

http://www.pirobot.org

DLink DCS-6620 cam [7]

Thu, 25 Feb 2010 11:02:53 EST

Follow-up:

I just downloaded & tested the new 2.15.3 with the DCS-6620 PTZ camera and it now works as expected.
Thanks again STeven!

DLink DCS-6620 cam [6]

Wed, 24 Feb 2010 20:02:42 EST

Ah yes - the contact form. I'm ready and will do that right now!

Laser spot processing [8]

Wed, 24 Feb 2010 17:02:56 EST

Yes, keeping the brightness to a min will help too.

Camera bleed is when a really colorfull pixel influences those around it to also become the same color even though they are not that actual color. Seems like 30 would be about right for the saturation. It just needs to be set to something where the image looks ok.

Pattern recognition? What pattern, all you have is a spot? So if you mean that spot detection is a pattern then yes, that will help.

Color is never a reliable form of detection as it changes based on what lighting is being used. Plus, as mentioned before the laser light center will appear white. You could check for a bright spot surrounded but a somewhat redder area but I would not rely on color any more than that. But there again, this is your project so feel free to experiment and see what works best!

STeven.

DLink DCS-6620 cam [5]

Wed, 24 Feb 2010 16:02:45 EST

Jeff,

When you're ready to go just use the contact form on the website to send us the IP address. That sends an email directly to us and will not expose the information.

Thanks!
STeven.

Liste Module error [4]

Wed, 24 Feb 2010 09:02:21 EST

I reinstalled the Microsoft SDK version 5 and the Listen Module is now working. Thanks for the help.

DLink DCS-6620 cam [4]

Wed, 24 Feb 2010 08:02:38 EST

Steven:
Absolutely! I would be elated to do what I can to contribute to RR. I will transport the 6620 tonight to a network that I can make accessible to you.
Could you PM me an email address that I can send you my IP/port info outside this forum?
Thanks!!!

Is this the way to go? Obstacle avoidance [9]

Wed, 24 Feb 2010 05:02:23 EST

The mask is working well. But I am new to Roborealm and a bit overwhelmed by all the modules. That is why I thaught it to be a good idea to verify my solution within a more experienced community.

Thanks everyone for the response.

Laser spot processing [7]

Wed, 24 Feb 2010 05:02:12 EST

Hi STeven,

Thanks a lot!!

I'm not sure if luvcview is actually changing the camera parameters or processing on the image. Am trying to get an answer.

And regarding the shutter speed, I can only minimize the gain and take shots. The brightness setting also should be kept at minimum, right?

I didn't understand the term camera bleed if saturation is increased. Can I keep this value at 30? For the two snaps I'd sent, the values were 100 for saturation and 10,10 for gain, brightness.

I'll again try using a more suitable set of values for camera settings. I think some pattern identification alongwith brightness would give a more reliable spot detection.

Why do you say that colour detection doesn't work well?
Brightness plus colour can detect a spot more reliably, right?

Thanks & regards,
Lakshmi

Button Interface [5]

Wed, 24 Feb 2010 03:02:48 EST

Ben,

I'd probably use one of the joystick buttons to perform those tasks ... that makes it a much more usable interface than even the buttons. Most joysticks come with quite a few so you shouldn't have any issues on running out.

We're not sure about the vex wifi but are planning on supporting it when we get around to purchasing that kit. While it may not work today I would not worry about purchasing the unit as the old vex system is quite old and we've been meaning to update that.

STeven.

Future support for Minoru 3D webcam? [6]

Wed, 24 Feb 2010 03:02:45 EST

Patrick,

Yup, I agree that stereo would help in that task. The current module that does Stereo would probably not give very nice results for this purpose. The newer one should do much better but still needs some optimization and work before we can release it (otherwise it works worse than the current one). We'll hopefully get that out sometime soon but unfortunately not immediately. Its a big module and will be of tremendous use to many so we want to be sure to get it right!

In terms of stereo systems with the right software the Minoru would probably work ok for your task ... but only with the right software!

Is this the way to go? Obstacle avoidance [8]

Wed, 24 Feb 2010 03:02:08 EST

Aswin,

I think the mask is probably the best and easiest way to go. Was there something in particular that you found is not working well? Or is the mask causing some unwanted issue?

STeven.

Liste Module error [3]

Wed, 24 Feb 2010 03:02:46 EST

This is the first call to the Microsoft SDK that should create a dictation context. While this is somewhat meaningless to most it basically means that the current install of MS Speech is not working in your case.

It seems that the module is detecting that MS Speech is installed but then immediately fails. This most likely happens when you have a previous version or one that has become corrupted.

Do you have version 5 of the sdk installed? Or a previous version?

STeven.

DLink DCS-6620 cam [3]

Wed, 24 Feb 2010 03:02:40 EST

Jeff,

Unfortunately we do not own a 6620 (just too expensive) and the demo version we were using over the web to test is now no longer active.

Is there any chance you can place that cam on the internet for a short period for us to test to see what may be going wrong?

We tested the 950 which seems to work fine on Vista but uses a very different communication technique than the 6620.

STeven.

Laser spot processing [6]

Wed, 24 Feb 2010 02:02:04 EST

The YUYV format would probably not help you much in this case unless you just focus on the Y (intensity) channel as previously mentioned. Most people think that using color to detect laser spots with a webcam is the way to go ... from what we have seen it doesn't work very well. A better technique is to rely on the laser being the strongest light spot in the image. But in order to ensure that it is you have to tell the webcam to snap images really quickly so that not too much light gets into the image. When this happens the laser will normally be the lightest spot. If it isn't, then you can try changing the shutter or gain on the webcam, or get a brighter laser (higher power or diff color) or a different frequency (perhaps IR?).

As Baghdad_boy mentioned before, please include images. Perhaps an image with normal colors with a high shutter speed or low gain.

STeven.

im new to vision robotics plz help me in making a ball follower with atmega32 [6]

Wed, 24 Feb 2010 02:02:24 EST

Yes, you can find that in the tutorials section of the website.

Basically, if the X coordinate of the COG is to the left of the center of the screen them move right, if it is to the right then move left. I.e. try to keep the object as best you can in the middle of the screen.

STeven.

Is this the way to go? Obstacle avoidance [7]

Tue, 23 Feb 2010 13:02:49 EST

Hi Aswin,

One thought that came to mind is to try the Prewitt Edge detector followed by the Auto Threshold module rather than using the Canny Edge detector. I find this produces much more stable edge maps than the Canny filter.

--patrick

im new to vision robotics plz help me in making a ball follower with atmega32 [5]

Tue, 23 Feb 2010 12:02:19 EST

i was sucessfull in finding cog,and send it to atmega,after that i was stuck up how to navigate robot towards the object l.e red ball or something by atmega(algorithm).. could u plz help in using the cog values and navigating robot towards object using atmega32

im new to vision robotics plz help me in making a ball follower with atmega32 [4]

Tue, 23 Feb 2010 12:02:10 EST

thank u...

Laser spot processing [5]

Mon, 22 Feb 2010 23:02:10 EST

Sir,

Thank you for your valuable information.

I have been playing only with the MJPEG format and then a jpeg to ppm conversion is done(using a shell script which uses a convert function in IMAGEMAGICK).

And the laser source module is of less than 5mw only.

I'm considering the RGB values for finding the brightness. Do you mean to say that the better approach is to play with YUYV format for spot detection?

Awaiting for your valuable suggestions,

Thanks & Regards,
Lakshmi

Using "Eliminate Global Flow" in the Optical_Flow Module [11]

Mon, 22 Feb 2010 22:02:25 EST

Hi Prof Mason,

This is very helpful information! I don't think I can use asynchronous mode since I get timeouts immediately as soon as I try that. The problem is that I have more than one thread setting goal positions and movement speeds, so to keep them from clobbering each other I use the synchronized mode on the servos and send a sync write periodically in its own thread. Also, I can't turn off the AX-12 reply since I am reading back position information from the servos.

Fortunately, I have the source code for the Forest Moon library and so I can now at least catch the timeout exception. However, I haven't figured out how I can then recover the serial port without a power cycle. I will also check out the robosavvy group.

Thanks again,
patrick

DLink DCS-6620 cam [2]

Mon, 22 Feb 2010 22:02:04 EST

Hi Steven:
I just received a DCS-6620 IP camera to use with RR 2.15.2, but am having some trouble with the DLink_Internet_Camera module.

PC#1 Vista 32b hardwired to cam via crossover cable:
select DCS-6620, put in IP, port=80, username/password as configured in cam, hit Start - short pause then Windows dialog: "RoboRealm has stopped working" then close program.

PC#2 XPsp3 hardwired to cam via switch:
select DCS-6620, put in IP, port=80, username/password as configured in cam, hit Start - nothing happens. Hit Stop then Ok or Cancel then can resume RR UI.

Both PCs have full control of the cam via web browser interface.
Any advice please?

Jeff Sumey
California University of PA

From RoboRealm to C++ program [3]

Mon, 22 Feb 2010 14:02:53 EST

Sorry, there is no real tutorial as yet.  I meant to write a tutorial on using python and the API, but am still buried in getting robots running!

First try to get Steven's example code running.  He goes through using the GetVariable and SetVariable commands as well as loading programs etc.  I find that if I setup the program in roborealm and then save it out as a robo(XML) file, I can then load that XML file into my program. (I use python but the ideas are the same)  Once you have Steven's example working, try loading in your own programs.  If you have the registered version, you can run Roborealm in faceless mode which means the user just sees the camera image OR nothing at all.  I typically use roborealm for doing all the visual processing up until I get some numbers out (moments, centers of gravity etc)  I then work with those in my program.    I also find that I tend to write back to roborealm to display vectors on my image etc.  As long as I am at it, I can use roborealm as my mouse handler and even a keyboard handler(a bit clunky)

You might also look at the work that Rud did in wrapping up the API a bit at:
http://mysticlakesoftware.blogspot.com/2010/02/roborealm-vision-processing-wrappers.html

If you want to distribute Roborealm as part of your application, you need to do some sort of OEM agreement with roborealm.(Not the expert here. this is something for Steven to clarify)   I have always just run it on one machine at a time.  I find that roborealm is an AWESOME prototyping tool and is tremendously helpful in developing vision techniques, even if you don't plan to use it in your final code release.

im new to vision robotics plz help me in making a ball follower with atmega32 [3]

Mon, 22 Feb 2010 14:02:39 EST

Your best bet is to run a small laptop like an EEE pc on the robot or a wireless camera so that you can do your processing in roborealm. Use roborealm to find the center of the ball relative to the center of the screen and then pass that to the ATMEGA via a serial connection. Your atmega will handle driving the motors based on the error between the center of the ball and the center of the screen. Take a look at Steven's project on the color tracking cannon to get some ideas.

good luck!

Laser spot processing [4]

Mon, 22 Feb 2010 14:02:57 EST

LUVCVIEW uses the SDL library and supports either MJPEG or YUYV encoding. I have been fighting with this a bit as YUYV seems to encode the color information in a packed format. IE it is a 422 format (intensity for every pixel but each pixel is only sampled in one color) I imagine you know a lot more about this. The SDL library also has a YUYV overlay that then does the interpolation to display a color image. I have been just downsampling my images to get color information for every pixel by just doing 4 -> 1 and averaging the two color values to get an unpacked image that I can work with.

For laser tracking I have found that using a 100mW laser and just cutting on intensity channel work fine. (I can split the intensity channel off the YUYV image very quickly and work with that) You can do this in roborealm just as well and it should give good results.

hope this helps!

Using "Eliminate Global Flow" in the Optical_Flow Module [10]

Mon, 22 Feb 2010 14:02:41 EST

I am writing code to run directly on the CM5 using winAVR.  I started with the sample AX12 code from robotis and went from there.  I do have an RS232 protocol that I use to talk to the CM5 that I wrote.   I have not seen your exact problem but I have had problems (undiagnosed really) with synchronous mode timeouts.  Switching over to asynchronous seems to solve the problem on those particular robots.    It seems like some of the servos will randomly or due to thermal shutdown not reply within the specified time frame.    I don't think it is a CM5 problem.  (I have swapped CM5s on the same robot and it didn't fix it)    Disabling the AX-12 reply seems to fix the problem and swapping servos also seems to fix the problem.  (In the case of the humanoid, the knee servos often overheat and go into thermal shutdown which takes them off the bus and causes  the problems you describe)  The servos will start responding again on a power cycle.

The robosavvy bioloid group is pretty active and might be a good place for extra help.

Roborealm related bit.  Folks should try the new python script module!  This lets you do all the things you would have been doing in vbscript but in python which has numerous examples and is well documented on the web!  I think vbscript has always been frustrating and this is a GREAT alternative.

good luck!
mmason

Using "Eliminate Global Flow" in the Optical_Flow Module [9]

Mon, 22 Feb 2010 11:02:01 EST

Hi Prof Mason,

You mentioned you are using AX-12+ Dynamixels as am I. I was wondering if you experience the same problem I do which is driving me nuts. I have 10 servos connected to a CM-5 which in turn is connected to my PC by way of a Serial-USB converter. I power the CM-5 with the supplied 12V brick. As I said in an earlier post, I am controlling the Dynamixels using the .NET library from Forest Moon Productions. Everything works extremely well EXCEPT that at random intervals I get an "Unhandled I/O Timeout Exception" on the serial port connected to the CM-5. I am polling the Dynamixels in Synchronized mode using a dedicated thread with 15 ms between each query. When the timeout occurs, I have to power-cycle the CM-5 (which also power-cycles the servos) to regain control. The time it takes before I get a timeout can be anywhere between 10 seconds and 20 minutes but it's a show stopper if trying to do a demonstration.

Other forums have suggested a cable problem but I have replaced and/or tweaked all the servo cables, the Serial-USB converter and even the CM-5. I have also tried the USB2Dynamixel controller in RS-232 mode (the Forest Moon library does not support the direct TTL connection).

Any thoughts would be greatly appreciated!

--patrick

Liste Module error [2]

Mon, 22 Feb 2010 09:02:52 EST

When I click the Listen Module in RR, I get the error "Listen Module failed at CreateRecoContext!". I have no idea what this means and have tried searching for it but I cannot come across anything. Help would be appreciated!

Listen Module error

Mon, 22 Feb 2010 09:02:54 EST

Ive been having an error with RoboRealm when trying to access the Listen Module. When I click the Listen Module in RR a box pops up with the error "Listen Module Failed at CreateRecoContext!". I was just wondering if anyone knows why this error is occuring. Ive tried searching for it but have no idea what it means.

Laser spot processing [3]

Mon, 22 Feb 2010 01:02:26 EST

Lakshmi.

Do you know if luvcview is actually changing the camera values or just changing the image in software?

Don't worry about saturation, just keep that at a normal value otherwise the camera may bleed areas together. Further decrease the gain until the laser spot is the only visible light in the image. It should be the brightest thing in the image.

You can also try switching to grayscale (on the camera) as that may produce a better lighted image ( you don't really need color at this point since the idea is to detect the brightest spot in the image).

STeven.

im new to vision robotics plz help me in making a ball follower with atmega32 [2]

Sun, 21 Feb 2010 02:02:27 EST

hi im new with vision robotics but very much familiar with mcu projects,i find robo realm using some kit modules,i want to do a ball follower with mcu,dc motors,rs 232 comunication,normal webcam,so i request u to plz help in these area

Is this the way to go? Obstacle avoidance [6]

Sat, 20 Feb 2010 09:02:39 EST

Hi all,

Thanks for the replies. Indeed I did read the tutorial about obstacle avoidence. It was very usefull. Unlike the tutorial however I want my robot to go straight ahead as long as possible, it should only turn if it is no longer possible to go straight ahead. For this I need to find the nearest obstacle straight in front of the robot. I should ignore all other objects that are not on collision course. My question does not involve finding objects or paths, that I already learned from the tutorial. My question is how to ignore objects outside collision course.

What I did is to mask out the left and right side of the image so that only a triangle of the image stays visible. This shape takes perspective into respect. I treat this part of the image just like in the obstacle avoidance tutorial, only the side fill fills from above. This way I get a point corresponding to the closest object in the robots path. This technique gives me the information I want.
However, I was wandering if the trick with the mask is the best way to ignore objects outside collision course. Are there alternative ways that you used?

Attached is my program and two images, the snap1 has the raw input and snap2 shows the triangle and point with the closest object on collision course.

changing values on a webpage [4]

Fri, 19 Feb 2010 13:02:50 EST

That works great, exactly what I needed. Thanks

changing values on a webpage [3]

Fri, 19 Feb 2010 12:02:46 EST

spacehermit,

If you download 2.15.2 we included a short example in the index.html file that shows how to use the get_variable to update the image count. Note that this example uses a 1 sec polling to get that information. Hopefully you can see how to extend this to your usage.

Note that the returned variables are in Javascript format and executed such that a new variable with the appropriate value is created. This is a nice way to avoid having to parse the variable information on return.

Note that you should backup your index.html before re-installing RR over your current install otherwise you may lose any changes you have made to the index.html file.

STeven.

From RoboRealm to C++ program [2]

Thu, 18 Feb 2010 22:02:03 EST

I am very very new to using RoboRealm. I have been working with a group on a project for visually impaired users. We are building a robot that will use a camera to recognize objects and speak to the user.

We have been working with RoboRealm and changing settings and making it speak to us when it recognizes certain shapes.

My question is... How do you go about turning this into code? I have been going over an example API program trying to figure it out but I am not sure yet. Is there a tutorial that discusses how to turn these settings and options into a C++ program? We got it to output to a web server but we need to start coding.

Any help is appreciated. Positive feedback only please.

Future support for Minoru 3D webcam? [5]

Thu, 18 Feb 2010 13:02:52 EST

Hi STeven,

Thanks for your reply. I don't actually own one of the Minoru cameras at the moment and from what you are saying, it doesn't sound like it's a good way to go for stereo vision. Other options I know of are the Surveyor SVS system, the Videre cameras and those from Point Grey Research. The latter two are almost certainly outside my budget constraints. Do you have experience with any of these other stereo setups?

The reason I'm interested in the disparity map at this point is that I am developing routines to have my two-armed robot pick up various objects using RoboRealm to visually center the target, and a pair of sonar/IR sensors mounted beside the camera lens to estimate the distance to the target. I then use a bunch of trigonometry to position the robot's hands to either side of the object in preparation for picking it up. If I could stereo vision to get a fuller distance map over the object, I could more precisely determine the left and right sides of the target for grasping.

--patrick

Button Interface [4]

Thu, 18 Feb 2010 06:02:14 EST

Right now I am making a mobile robot. It has a lot of features like pan/tilt camera, laser pointer, and encoders on the wheel. What I wanted to use the button interface for is to turn on and off the laser. Also I am planning on controlling the robot with a joystick. While driving around, the value of the motors would be read by the encoders and stored into an array. Then when I press a button on the button interface, I will be able to automatically do the entire path I just drove, in reverse. I would just use the values from the array as the new motor values.

This brings me to my second question. I am building this entire robot with the VEX Robotics kit. I don't know if you know, but the VEX kit is coming out with new wireless WIFI control. Do you think the VEX RoboRealm module will be able to work with this WIFI system? I know the VEX RoboRealm module is kind of old...

Thanks for your help,

Ben

Future support for Minoru 3D webcam? [4]

Thu, 18 Feb 2010 02:02:07 EST

If you are curious how it works post two images here and we'll tweak the params on that module to see what the best disparity it can produce ... an interesting experiment and one that would generate more tests for the newer version.

STeven.

Future support for Minoru 3D webcam? [3]

Thu, 18 Feb 2010 02:02:48 EST

Patrick,

Yea, we picked up one of those a while back. While a compact way to get two images from a single webcam the camera does not compute a disparity map nor does it really align the images in prep for such a process. It is basically two webcams stuck together in a compact uncalibrated way.

So it is compatible with RR at least in getting the images (you can use the 3D_Viewer module to see the anaglyph) but we are still working on the stereo module to get the actual disparity map ... for this and all other cameras. The issue we are currently dealing with in that module are speed (as would be expected) and planar objects (think walls) that cause a lack of features to match the two images on. We think we will have a interesting module for this but can't really estimate the eta on launch for that one since it is still in active research.

The current stereo module should create a disparity map but will do so slowly and not very accurately ... thus the reason it has not really been recommended nor announced. It was added mainly for testing purposes on our side (it is probably not even documented either!).

STeven.

Variable Values on VBScript [15]

Thu, 18 Feb 2010 01:02:58 EST

Vnik,

I would probably use the Matlab API functions to communicate between Matlab and RoboRealm. While writing a text file will work there is a small change that both RR and Matlab will try to access the file at the same time and cause an error. Using the API will prevent this issue from happening and also be much more efficient than writing a text file. See the Matlab sample in

http://www.roborealm.com/downloads/API.zip

which has a matlab example that works quite well.

For the detection of multiple objects in the Blob Filter, yes, just switch on the create BLOBS array checkbox in that interface and an array will be created that contains the COG of all the object. Arrays are a little different than other variables but can also be referenced in other modules using

variable:X

where X is a number from 0 to the size of the generated array.

Or in VB/Python you can use GetArrayVariable to get access to this array.

Also note that Shape_Match module does a similar thing in creating an array of results.

STeven.

Traxster II Module + Serial Communication [7]

Thu, 18 Feb 2010 01:02:05 EST

Scott,

Interesting thought ... the provided dropdown is just for a variable (as you have determined) as we figured you would probably be better off using a variable configured from VB or other API application as apposed to a static text string. As a lot of folks do not know the [variable] substitution trick we often settle on a variable selection. Glad you got that working!

STeven.

Button Interface [3]

Thu, 18 Feb 2010 01:02:54 EST

Ben,

Unfortunately not. The full screen interface only allows graphics to be draw and no UI windows are accessible. What you may do instead is use the keyboard read to program in some keystrokes that would perform the same actions as the buttons ... the other possibility is to use the mouse in some way ... perhaps movement or even the mouse buttons as ways to get input into the application.

What are you using the buttons for during full screen view?

STeven.

setParameter for ButtonInterface [2]

Wed, 17 Feb 2010 22:02:49 EST

I am having trouble implementing a dynamic button interface. I have an array of values that correspond to csv files that are loaded using READ_VARIABLES.   I want to loop through the array and set the buttons vertically. The results are unpredictable. Some settings like name, color, height, toggle_group are not set for all buttons. Sometimes the screen flashes repeatedly

Are there limitations to setParameter? I have tried adding a delay between each setParameter statement using sleep and msgBox but it does not help.

Am I simply better off integrating an IE html form to provide this control?

The vbScript module does all the work:

v_jobs = "A,B,C,D,E,F,G"

IF NOT(ISARRAY(AR_JOBS)) THEN
  ar_jobs = split(v_jobs, ",")
  setParameter "BUTTON_INTERFACE", 1, "dialog_width", 160
  setParameter "BUTTON_INTERFACE", 1, "dialog_height", ubound(ar_jobs)*50+90
  setParameter "BUTTON_INTERFACE", 1, "dialog_x", 70
  setParameter "BUTTON_INTERFACE", 1, "dialog_y", 20
  for x = 0 to ubound(ar_jobs)
    setButton x, ar_jobs(x)
  next
END IF

sub setButton (b,n)
     vx = cstr(b+1)
     setBut vx&"_name", n
     setBut vx&"_height", 46
     setBut vx&"_width", 110
     setBut vx&"_x_position", 20
     setBut vx&"_y_position", 6+(b*50)
     setBut vx&"_color", "#330099"
     setBut vx&"_text_color", "#FFFFFF"
     setBut vx&"_is_toggle", "TRUE"
     setBut vx&"_toggle_group", "ALL"
     setBut vx&"_clear_on_release", "FALSE"
     setBut vx&"_variable_name", "V_CSV_X"
     setBut vx&"_variable_value", b
end sub

sub setBut (p, v)
  setParameter "BUTTON_INTERFACE", 1, p, v
  'write p&":"&v&"|"
end sub

changing values on a webpage [2]

Wed, 17 Feb 2010 21:02:00 EST

So I'm trying to get my parallax servo controller card to interface with a webpage.
I can get the webpage to change variable values and move the servos, with getvariable and setvariable. But I can't get the servo position to go to the webpage. For example moving a slider to show the position of the servo. Is there a vbscript command that I'm missing that will change a value on the webpage? Or what would you recommend I do so that I can have roborealm feed variable values to the webpage that aren't only in the video feed screen? Thanks

Using "Eliminate Global Flow" in the Optical_Flow Module [8]

Wed, 17 Feb 2010 13:02:10 EST

Great Solution. I have had good luck with the movement module for tracking moving objects in the frame. There might be something to this idea of using the movement module to determine the global flow and then subtract that from the flow vectors. I really want to do optical flow balancing for obstacle avoidance. This means I really need to correct the optical flow vectors for the global motion. I still haven't figured out any answers for this. Unfortunately, I am currently distracted with trying to get gcc to compile using theNEON fpu on the ARM8 board I am using for the embedded vision processing.

Using "Eliminate Global Flow" in the Optical_Flow Module [7]

Wed, 17 Feb 2010 11:02:03 EST

Hi Rud, Prof Mason and STeven,

I finally had a chance to come back to this problem and I think I found something.
As I mentioned earlier, I was hoping that the "Eliminate Global Flow" option in the Optical Flow module would do the trick but it seems to have minimal effect on the block intensities.  However, I just discovered that the same option in the Movement module (there called "Reduce global movement") seems to work very well!  Using the test video I posted at:

http://www.pirobot.org/videos/optical-flow/test6.avi

I can easily isolate the moving hand even when the camera is panning by using the Movement module followed by the Population Threshold module.  I've attached the .robo file I found to work.

STeven, I'm wondering if the Eliminate Global Flow option in the Optical Flow Module is working correctly.  When I toggle this check box on and off, I see the effect in the overall movement vector but not in the block intensities.  In contrast, the Reduce global movement check box in the Movement module has a clear effect on pixel intensities when the display is set to Grayscale.

--patrick

Pololu Micro Maestro [8]

Wed, 17 Feb 2010 09:02:45 EST

Nice work STeven! A new module for the Pololu Micro Maestro in less than a week! I wish we could get vendors at work to respond like that. Two thumbs up for RoboRealm.

Thank You
Doug
ArgonautJr.com

interfacing ps3 controller with rr [3]

Wed, 17 Feb 2010 09:02:19 EST

PS3 controllers are not directly plug and play compatible with Microsoft Windows applications, so you'll need to find a 3rd party driver which will allow you to use it with RR. I did a quick google for "ps3 controller on PC" and found a bunch of them.

interfacing ps3 controller with rr [2]

Wed, 17 Feb 2010 07:02:23 EST

the joystick module detects the playstation3 controller but when i move the d-pad or any of the analog sticks nothing happens the buttons dont work either, i have interfaced the logictech 3 controller with my project and it works great but now i want to go wireless so im trying to get the ps3 controller to work, im using the provided usb cable to interface it first before i try do it wirelessly. i would be very grateful for any help and thanks to every1 who has helped me so far

Laser spot processing [2]

Wed, 17 Feb 2010 04:02:21 EST

Hi Baghdad_boy,

I'm sending two snapshots which was taken with no filter, camera settings: brightness(b) = 10,saturation(s) = 100,gain (g)=10. The RGB values on the info window correspond to one of the brightest spots on the plastic bag, which are comparable to the laser spot values, (which is towards the top left corner of the bag).

Pls try with this. I've taken several shots with the b, s and g settings changed. I think the best combination is reduced b & g values and increased s value. Not tried with white balance & contrast. I'm doing all these through 'luvcview' on Linux and the webcam is logitech quickcam pro9000.

Thanks & Regards,
Lakshmi.

Traxster II Module + Serial Communication [6]

Wed, 17 Feb 2010 00:02:37 EST

STeven,

Yeah!  Perfecto!  This works great, thank you so much for the addition.

At first, I had variables for the six color channels in the SetVariable module, like redL, grnL, bluL, redR, grnR, bluR and would send it like "i2c 16 [redL] [grnL] [bluL] [redR] [grnR] [bluR]<cr>.

For the "serial variable" box for the Traxster, I tried putting this in but it didn't work, so I tried to create a variable using the SetVariable module, with the contents of the "i2c" string as from above.  I thought that putting the desired variables in the brackets like that would put in the values (works in the serial module but not in the SetVariable module).

Instead, I had to create a variable to be sent that had the i2c header and all of the relevant values already in it.  So, this is working out fine :)  Thanks again.

--Scott

Traxster II Module + Serial Communication [5]

Tue, 16 Feb 2010 20:02:24 EST

Scott,

not sure if you noticed (nor if this is relevant after a month) but the generic i2c textbox was added to the Traxster interface.

STeven.

programatically moving line profile [6]

Tue, 16 Feb 2010 17:02:21 EST

Thank you!

This is VERY helpful, I am genuinely appreciative. I continue to be impressed with this product an the support from fellow users. I hope I can return the favor.

The help, the code, and the video were over the top! That's awesome.

Steve

Button Interface [2]

Tue, 16 Feb 2010 17:02:15 EST

I was wondering if it is possible to use the button interface while the camera is set to full screen mode. I have version 2.14.8 and when I Go to -Zoom -Full Screen my camera obviously takes up the entire screen. But then I can't use the buttons anymore. Is there anyway I can access the buttons while in full screen mode?

Thanks,

-Ben

Is this the way to go? Obstacle avoidance [5]

Tue, 16 Feb 2010 12:02:42 EST

Rud,

One technique I forgot to mention last week was using a visible line laser and detecting the lowest (ie nearest) point on the laser and moving to avoid that. It works quite well indoors (don't even bother with it outdoors as few visible line lasers can compete with the sun) and has proven to be one of the more reliable obstacle avoidance techniques.

http://www.roborealm.com/help/Laser_Line.php

STeven.

Sony Camcorder 1394 Firewire [7]

Tue, 16 Feb 2010 11:02:39 EST

David,

Can you mention the error code here? It will mention something like XXXX format is not supported. If we know that format we can see about adding that directly to RoboRealm which will avoid those errors.

Another possibility is to click the Format button that you should see in the Video Tab and select the RGB24 as the format to use once the dialog becomes visible. Normally most cams have several formats that they can produce. Selecting one that RR supports would do the trick ... if not we can probably add in one that would work for you.

STeven.

Sony Camcorder 1394 Firewire [6]

Tue, 16 Feb 2010 09:02:22 EST

I see it as a Microsoft DV Camera. When I select it I get the RGB error (I apologize I don't have the exact error code). This is an error code I have seen in the forums before.

Parsing GPS NMEA strings? [4]

Tue, 16 Feb 2010 05:02:02 EST

Thanks Steven. Thanks for the link and good to hear you're developing a GPS module. What would be even better would be a general NMEA parser module, that way it could be used with any NMEA compliant device (GPS, digital compass, radar, sonar, etc.).

Sony Camcorder 1394 Firewire [5]

Tue, 16 Feb 2010 04:02:26 EST

David,

It is rare when opencv can grab an image that RR cannot. What error do you see in RR specifically?

Do you see the cam in the dropdown in Options->Video tab? What happens when you select it?

STeven.

Parsing GPS NMEA strings? [3]

Tue, 16 Feb 2010 04:02:48 EST

We do have a gps module in the works but before that gets released you should have a look at the guts of the GCS that Millswood has created. It is written in RR. Some very cool stuff ...

http://www.millswoodeng.com.au/downloads.html

Click on the Attopilot GCS version 1.25 to download the RR code.

STeven.

Wrapper classes for reading variables from RR_API [5]

Tue, 16 Feb 2010 04:02:25 EST

Yes, if you do make modifications you can email me/us directly and/or post your update summary to keep others informed.

Thanks!
STeven.

Pololu Micro Maestro [7]

Tue, 16 Feb 2010 03:02:42 EST

Doug and Cpt. Paul,

Seems that the SSC module would kind of work if the Maestro was set to USB dual port ... but probably not well enough to trust. Instead we created a new module specifically for that board that uses direct USB commands to control it. Download the latest version and see if that works for you ....

STeven.

Laser spot processing [5]

Tue, 16 Feb 2010 02:02:49 EST

Hey Lakshmi,

I have an exposure option on my webcam by going into Options -> Video Capture and then choosing the Camera Control tab. As you said this controls the exposure time. Maybe you don't have this on your settings.

Can you please send me a screenshot of the video you are working on with no filters applied so I can have a play around with it and see what works best.

Cheers
Baghdad_boy