GET YOUR BEARINGS
Plan your next AI Learns To Drive session
Our independent guide walks through each core system of AI Learns To Drive, offering structured observations and comparison ideas for network architecture, input selection, evolutionary settings, track construction, and competitive racing.
01 / THE GUIDE
Layer Depth Test
Build one network with a single hidden layer and another with three hidden layers. Run identical training sessions on a straight track and note differences in learning speed and stability.
- Start with a minimal layer count to observe basic behavior before adding complexity.
- Combine positional inputs (wheel angle) with speed inputs for richer decision data.
02 / THE GUIDE
Input Variation Study
Train two identical networks, one receiving vision ray inputs and the other without. Compare lane‑keeping and obstacle avoidance to see how visual data influences behavior.
- Use high‑contrast colours to improve visibility during fast runs.
- Record the paint choice alongside network settings when sharing results.
03 / THE GUIDE
Reward Combination Trial
Create a reward that balances fast completion with a penalty for collisions. Observe how the AI adjusts its aggressiveness over several generations.
- Begin with a modest agent count to reduce compute time while observing early trends.
- Blend a speed‑focused reward with a safety penalty to promote balanced driving.
04 / THE GUIDE
Surface Transition Challenge
Design a track that switches from asphalt to ice halfway through. Test multiple network configurations to see which handle the traction change most effectively.
- Mix surface types within a single track to test AI adaptation to changing traction.
- Place a simple checkpoint challenge first, then add a more complex one to observe performance shifts.
05 / THE GUIDE
Campaign Benchmark Comparison
Complete a campaign challenge and record the star and coin rewards. Then attempt the same track with a different reward mix to compare performance outcomes.
- Use campaign challenges as a baseline before attempting community‑created races.
- Observe how different reward mixes affect your AI’s ranking on the leaderboards.
