Machine-Learning-Racer/notes.txt at main · SamuelSaylor/Machine-Learning-Racer · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
Positive Reinforcements
- Go fast, checkpoint, time without touching boundaries, stay on track

Negative Reinforcements
- off course, time limit, elapsed time

Packages:
-pygame, stable baseline, gymnasium, numpy, matplotlib

Proposed project structure:
data/
-> data.csv (or whatever we get)

Driver.py <-- this is the car, acceleration etc (attributes)
CarPhysicsEngine.py (imported into driver)
MachineLearning.py
charts.png
recources/
-> assets

Physics:
acceleration, velocity, orientation

Budapest Track for training.


Prince: Physics stuff - give sam the knowledge that the car needs for how real life physics works

Samuel: make/train model

Samantha: pygame: get display running, make sure object detection setup, make sure layering has priority (racer cannot go under track)
understand ratecaseting/object mapping


Start of hackathon: work on everything
After lunch: start training/debug
making plots/presentation: before the thing is due