My AI Project: Part 1

These are copies of an AI mind I've coded.
- There is the main BOT (spheres), then there is their direction/speed vector (short line). And finaly, the IR (infra-red) radar (long line).

It all started out when me and a friend decided to create an AI controlled robot in real life. So I started on my part on the PC, while my friend is working on the electronics.
So when I was creating these bots, I was keeping in mind the fact I'd have to use that AI on a real robot.
So for example, collision detection in real life will be done with an IR transmitter and receiver. So...in my digital version, I simulate a rotating IR transmitter and receiver. (transmitter is the long line, receiver is the collision detection algorithm)

In short: They go about randomly, and when they encounter a wall (red) they break then steer in order to avoid it. They will do this forever.

Details: The IR spins around the bot, checking each angle for a wall. When a wall is "touched" by the radar, the bot brakes. Once it is going slow enough, it steers away from the wall and pics a new direction. -- When not "seing" any wall, the bot randomly chooses a new direction to explore.
-- The walls are detected using polygon-line collision detection. So walls can be placed anywhere and still be detected.

Problems:
- After lots of time, they start to get stuck, because the IR is not spinning fast enough.
Must think about this because in real life, we can't have the IR spin too fast.

Next Part:
- Make the IR "see" other bots as well as walls;
- Create a bigger and tougher to navigate level (map);
- Add battery life, and a charging center, where the bot will go to recharge when low on battery;
- Design and implement a better AI:
+ Bots will rememeber walls once they encounter them;
+ Bots will be able to go from A to B watching for new walls and the walls in memory;