3D maze in browser controlled by Small Basic

[litdev]

I update the 3D browser code so that as an easier challenge it will pass back the current cell location, possible movement directions and you can the ask the robot to just animate to one of the available neighbor cells, ie. the robot has more internal awareness and less for the AI controller to do.

This is a very simple sample code to show the new data and commands: GRWW331.000 - any questions or problems welcome.

I do think that the original challlenge is much more interesting, trying to move and navigate just using the original methods provided.

EDIT

Fixed an issue.

[AbsoluteBeginner]

...
I do think that the original challlenge is much more interesting, trying to move and navigate just using the original methods provided.
...

I also think the original challlenge is more interesting. 

[stevantosic]

The simpler, the better. The newer one is excellent.

[AbsoluteBeginner]

The simpler, the better. The newer one is excellent.

I wonder if climbers agree with this idea. 

[litdev]

Haha AB,

We are all coming at this with different objectives, me and you mainly for our entertainment and I think Stevan maybe wants to use as a teaching aid as well.

[AbsoluteBeginner]

Yes, it is possible. 

I was able to do test runs of my code.
Now I have the following question. Can we change the format of the commands that the SB program sends to the browser engine?
Can we make the "move", "left" and "right" commands send not a logical 0 or 1, but a digital value of the step by which the browser engine should change the rotation angle or position of the robot in one iteration of the calculation? 

[AbsoluteBeginner]

Hello everyone. 

This morning I assumed that when my "turn right" command is executed, the engine in the browser performs this rotation by changing the robot's angle by 30 degrees per iteration.
So when my SB program receives information that the robot has reached the 90 degrees angle I wanted, and my SB program sends a "stop turning" command, the engine has time to perform one extra step of rotation.

This means that the SB program should send the "stop" command to the engine ONE STEP BEFORE the SB program receives information that the robot has reached the desired position. And we do not need to change the command format that I wrote about in the previous message.

What do you think about this?

[AbsoluteBeginner]

Now I tried to make the robot take just one step.
I did it. 
My next goal is to rotate the robot by one minimum angle.

[litdev]

AB,

I think you are on he right lines - I don't want to suggest anything extra, its a puzzle how to get it to move reasonably - it will never be fast or exact, lots of little movements towards a goal is probably best.

I could add other control options like rotate by a certain amount or move forwards a set distance (if these were larger amounts then an amination would look better and this is pretty much what I added with Stevan in mind).  Again I do think controlling something that is 'awkward' is interesting in itself.

[AbsoluteBeginner]

Once we learn to take small steps to move and turn, we will have the capabilities of step-by-step control technology, which is well known to fans of technical modeling. 

Our project will be able to develop easily and interestingly. 

Hooray!