A TinyML Reinforcement Learning Approach for Energy-Efficient Light Control in Low-Cost Greenhouse Systems

TL;DR

This paper introduces a lightweight reinforcement learning method using Q-learning for adaptive, energy-efficient lighting control in low-cost greenhouse systems, demonstrating effective stabilization at multiple light levels with minimal overshoot.

Contribution

It presents a novel low-power RL control strategy for greenhouse lighting, implementing a model-free Q-learning algorithm on microcontrollers for real-time adaptive regulation.

Findings

Effective stabilization at 13 light levels

Minimal overshoot and smooth convergence

Feasibility of on-device RL in resource-constrained environments

Abstract

This study presents a reinforcement learning (RL)-based control strategy for adaptive lighting regulation in controlled environments using a low-power microcontroller. A model-free Q-learning algorithm was implemented to dynamically adjust the brightness of a Light-Emitting Diode (LED) based on real-time feedback from a light-dependent resistor (LDR) sensor. The system was trained to stabilize at 13 distinct light intensity levels (L1 to L13), with each target corresponding to a specific range within the 64-state space derived from LDR readings. A total of 130 trials were conducted, covering all target levels with 10 episodes each. Performance was evaluated in terms of convergence speed, steps taken, and time required to reach target states. Box plots and histograms were generated to analyze the distribution of training time and learning efficiency across targets. Experimental validation demonstrated that the agent could effectively learn to stabilize at varying light levels with minimal overshooting and smooth convergence, even in the presence of environmental perturbations. This work highlights the feasibility of lightweight, on-device RL for energy-efficient lighting control and sets the groundwork for multi-modal environmental control applications in resource-constrained agricultural systems.