a maze and sensors

what if vex was to create a a modifiable maze kit. it would involve a bunch of square panels, pegs with four slots at 90 degree angles that would all interlock to create a maze. you could combine kits to make lager mazes. the kit would come with three sensors that are a cross between ultra sonic range finder and line tacker. it would use infared lazers to follow the black edges of the wall. you could then program it to dive through the maze without programing moves. it would make a decision when it came to an intersection to go left, right or straight. and when it came to a corner it would turn that cornerand when it got out it would stop. you could program it with the add on disk for the programing kit that would come with the maze kit, or you could download it from the internet. you could then program it to follow a set of rules and see if it can find a way out. you could also use an optical shaft encoder to count the moves the robot makes. so with this kit (optical shaft encoder not included) you could make a maze running robot.

i like the idea of a preassembled maze, but there is no need for other sensors. all you need i one or two limit switches on one side and the fromt side. you just tell the robot to turn it’s left wheels forward until one of the limit switches hits is, then tell the right side to move forward. eventually it would get out of the maze.

u could also program

the sensors are just there to make following walls easier. and now that i think about it it would probaly need four of those sensors. it would abvously need programing, you would need to teach them the rules of the maze.

If you’re interested in maze robots there are many robotic competitions out there that involve mazes. Search the web.

The people voting “stupid” on this thread, no offence, probobly put that for the reason of either, no programming kit. If that isn’t the case it is because it is too much of a hassle. Many people could have their own reasons but I think everyone should just go with what they like.

I’m not voting though because it isn’t a stupid idea but its not amazing or “Wowing” me. It is not like it would be very hard to make that yourself. Also with First people are already occupied, but for the lazy people (no offense to anyone) it is “Awesome” because they don’t want to make it themselves (plus they probably don’t have enough experience to cut a piece of wood (no offense) or drill holes (no offense).) The average (every) person involved in vex though is NOT lazy because they made their robot, and doing that at all makes them classified under the category of NOT lazy. :smiley:

Why not make your own maze, and use ultrasonic rang finders, bumper switches, limit switches, line following kit, etc? Its doable.

If you made a simple maze yes, it may be doable.

Just restating, that a more complex one will take some time.

a maze is extremely easy to make with a bit of plywood, nails and a band saw, although kits are always nice because then you wouldn’t have to recut wood every time you wanted to make a maze or if you dont have the right tools in your arsenal

Unless the actual Vex robots got much smaller, the maze kit would be huge! Also if it were made out of anything but plastic, the shipping would be a lot of money. The maze kit probally would cost around $100-$200 because it is Vex. No, I think that it would be cheaper, more convinent, and a heck lot smarter to just build your own.

i think price is ok if you dont already have the tools but it is true that the maze would be huge:eek:

Technically it wouldn’t be really complex for the robot to be programmed to autonomously through a maze if you do it right. Any maze can be accomplished by sticking to either the left or right side of the maze.

Not true - The maze must satisfy certain restrictions before a pure left/right hand on the wall strategy will guarantee eventual success.

Blake

Check out this robot that I made that can go through a maze.

https://vexforum.com/attachment.php?attachmentid=184&d=1184439457

How does it work? Do you have rangefinders or something on it? Or does it just use bumper sensors?

The ultrasonic sensor scans left, right, and forward. Through the distance recorded, the robot decides which way it should turn and go. It will never hit a wall.

What restrictions? I have always been told that as long as you constantly make left turns (or right) you will always reach the end.

The restriction that comes to mind first is that for the left/right-hand-on-the-wall strategy to work, the goal mus be on an edge of the maze (a leaf node if you think of the paths through the maze as a graph of connected nodes).

Another restriction is that the maze must lie entirely within a plane (no paths that cross over other paths).

I understand that most of the time in casual conversations here, we take shortcuts and use the word “maze” to refer only to the 2-D entrance-to-exit sort of maze; but we all must careful when passing out advice/guidance. At that point in the conversation we need to be accurate and precise; so that we don’t give our listeners rules-of-thumb without also telling them the exceptions to look out for.

Knowing the exceptions is at least as important as knowing the rules of thumb.

Blake
PS:I’m not a maze mathematics expert; but I do remember that I had to unlearn the hand-on-the-wall rule when I encountered one of the exceptions…

i agree but remember the K.I.S.S. principle (keep it simple, stupid):smiley: :smiley: :smiley: