So i plan to build a bot, my approaches regards this goal so far been some javascript JED and D2JSP bots, especialy than with mattlants botOS OOP pathfinding solution.
So i plan so far a robot who generates a 2D map genearted with sensor data. As the beOS project is offline i want to know what other recommend for pathfinding.
Cheers
You could ask some more specific questions. It's very unclear what you're asking for.
Dijkstra's algorithm is probably what you're looking for, it will find the shortest path through a weighted graph in polynomial time.
You could also look into A*, and if memory is an issue you could consider A* with iterative deepening or something.
Are you aiming to find a ready to go code? Or to implement this yourself?
Much more fun doing it yourlsef? At least I think so.
I would've used a finite state machine.
Im not sure yet i still sort things out and will start in the coming weeks with the construction - based on the roboduino MC.
So we have light sensors and basic autonomous movement.
But when it comes to the actual compute of the data, i dont know so i thought i start withwhat i know from my scripted bots.
And it is nice to read how other people would approach this or to learn about solutions.
For example i want to be able to configure how the bot handles the environment. I want to define constants of known objects from the bots sensor object data or ambient margins. So that i can let the bot know where my computer is and that this is a door.
Maybe it is possible to sense certain kind of objects through the sensor data, by meassuring the ambient data and size, contrast, surface ...
And than i can have functions in which i meassure distance from known objects ... well of course first i will follow the specific tutorials and test all the pathing solutions.