Hi,
I am building a program for steering my robot and mapping the environment.
My question is about drawing and storing the map.

What I want:

• 1cm resolution
• 10x10 meter map size (or dynamically expanding if possible)
• rotatable and zoomable

My approach would be to create a 1000x1000 byte array (store not scanned, clear, blocked).
If I want to rotate the map, I need to recalculate position of each pixel.

If I zoom I read only a smaller portion of the array and draw dots with arrayvalue*zoom in size.

I store the map in a csv (all array values). I could also store only scanned points. But as I would need store position, the amount of date would be equal or even more.

Is this the best or at least a good approach? Does someone has tips or a demo project?

Thanks
Robert

Hi,

I am trying to display an image I receive via serial. I submit different color depths.

The code is below reacts on:

g_grayscaleImage == 0     // 16 bit rgb
g_grayscaleImage == 1     // 8 bit grayscale
g_grayscaleImage == 2     // 4 bit grayscale

The code works fine for 16 and 8 bit. With 4 bit I don’t get the pixels to the right position. (so the part from "else if (g_grayscaleImage==2)" doesn't work)

The “best” result I got with the code below. It looks like this:
http://www3.picturepush.com/photo/a/11925461/img/11925461.jpg

Does someone see the mistake and can help me out?

Here the code:

1.     for (int row = imageh-1; row >= 0; --row)
2.     {
3.       for (int col = 0; col < imagew; ++col)
4.       {
5.         int loc = imagew*row+col;
6.         if (g_grayscaleImage==0) { 
7.           int msb = packedbuf[2*loc] & 0xff;
8.           int lsb = packedbuf[2*loc+1] & 0xff;
9.           int r = (lsb >>> 3);
10.           int g = ((lsb & 0x7) << 3);
11.           g |= (msb >>> 5);
12.           int b = (msb & 0x1f);
13.  
14.           r = (r << 3)|(r >>> 2);
15.           g = (g << 2)|(g >>> 4);
16.           b = (b << 3)|(b >>> 2);
17.           img.pixels[loc] = color(r, g, b);
18.         }
19.         else if (g_grayscaleImage==1) {
20.           img.pixels[loc] = color(packedbuf[loc], packedbuf[loc], packedbuf[loc]);
21.         }
22.         else if (g_grayscaleImage==2) {
23.           int LeftPixel = packedbuf[loc/2] & 0XF0;
24.           int RightPixel = packedbuf[loc/2] << 4;
25.  
26.           img.pixels[loc] = color(LeftPixel, LeftPixel, LeftPixel);
27.           img.pixels[loc+1] = color(RightPixel, RightPixel, RightPixel);
28.           ++col;
29.         }
30.       }
31.     }
32.  
33.   img.updatePixels();
34.   return img;
35. }

Thanks
Robert