nickcaputo / lejosev3project

Where Chris and I put the leJOS EV3 Java code for the robot we built. Uses AI to analyze its surroundings and a depth and breath first search algorithm to solve a maze and find the most efficient way back to the start.

This is where Chris and I are putting code for DaveKillbot, our robot for the final project in the AI class we are taking.

Created February 7, 2016, 23:44

lejosev3project's Issues

Go home!

Help Dave get home!

This can be done by using the following:

• The history of cells visited (how do we eliminate bad path choices?)
• The data for each cell (wall positions, directions taken [maybe])

2 options I can think of:

1. The robot follows its steps back through the maze roughly the way it came. This can be done by eliminating cells from its list of visited cells where it went off course. (Like, look for duplicate cells like you were saying, npocaputo)
2. Use a new algorithm to look through its data on cells that it's stored and find the best path to take, not necessarily the one it took to get to the goal. This would not use a list of cells previously visited. Maybe some sort of heuristic?

I don't know which one will be better, so we can discuss this.

Draw more on the screen

Be able to write on the map the location of the goal, at least.

If we can make another screen and switch between them, the second screen can display:

• Current location
• Location of goal
• Other stuff (number of moves made?...)