GardenRL
GardenRL teaches AI agents to grow hydroponic lettuce through 30-day episodes with delayed rewards.
Core trick
Verifiable harvest weight in grams serves as the GRPO reward for a 30-day hydroponic simulator where early pH errors surface days later.
Description
GardenRL teaches AI agents to grow hydroponic lettuce through 30-day episodes with delayed rewards. A pH mistake on day 5 causes calcium lockout, brown leaf tips by day 9, and 60% yield loss by day 30 — requiring genuine long-horizon planning and hidden-state inference.
Grounded in the HydroGrowNet dataset (390K real images), rewards are verifiable harvest weight in grams — no LLM judge. We trained Llama-3.1-8B with GRPO (OpenPipe ART) for 50 steps: reward +473%, eval harvest +11%, success rate 50%→60% on held-out seeds.
Addresses Problem Statement 2 (Long-Horizon Planning) and 3.1 (World Modeling) and the Mercor sub-bounty.