Calibration-First Reward-Component Auditing for Reinforcement Learning Control in Smart Greenhouses

• URL: http://arxiv.org/abs/2607.11959v1
• Date: Sun, 12 Jul 2026 07:37:54 GMT
• Title: Calibration-First Reward-Component Auditing for Reinforcement Learning Control in Smart Greenhouses
• Abstract summary: Greenhouse reinforcement learning can test climate-control ideas at a speed and scale that is difficult to achieve with crop experiments alone.<n>For smart-greenhouse control, however, a single simulator return is not enough: a grower or control engineer also needs to know when the policy heats, enriches CO2, vents, manages humidity, deploys screens, or uses lamps.<n>We propose a reproducible calibration-first reward audit framework that keeps named greenhouse-control reward components comparable across simulator training, facility-adapted rollouts, logged Autonomous Greenhouse Challenge records, and actuator-rule distillation.
• Score: 4.103008228953626
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Greenhouse reinforcement learning can test climate-control ideas at a speed and scale that is difficult to achieve with crop experiments alone. For smart-greenhouse control, however, a single simulator return is not enough: a grower or control engineer also needs to know when the policy heats, enriches CO2, vents, manages humidity, deploys screens, or uses lamps.We propose a reproducible calibration-first reward audit framework that keeps named greenhouse-control reward components comparable across simulator training, facility-adapted rollouts, logged Autonomous Greenhouse Challenge records, and actuator-rule distillation. In GreenLight-Gym, the framework decomposes the scalar reward into conditional temperature, CO2, humidity and vapor-pressure-deficit, screen, and actuation-proxy terms; adapts GreenLight to the second Autonomous Greenhouse Challenge logged climate traces; and scores the same components on logged greenhouse data.

Related papers

• ADAPT: Physics-Aware Diffusion-based World Models for Adaptive Predictive Transferable HVAC Control [19.871939624912073]
  ADAPT is a physics-aware conditional diffusion indoor environmental world model for HVAC control.<n>It reduces HVAC energy consumption by 7.3% and occupant discomfort by 30.2% compared with state-of-the-art baselines under IID control.
  arXiv  Detail & Related papers  (2026-08-20T08:58:28Z)
• Contextual Quality-Diversity Evolutionary Reinforcement Learning for HVAC Control in Tropical Commercial Buildings [0.0]
  This paper proposes a contextual quality-diversity evolutionary reinforcement-learning controller, CQD-ERL.<n>The controller is trained on a two-tier reduced-order environment representing the latent load, cooling-tower approach and humidity constraints of a Singapore commercial building.
  arXiv  Detail & Related papers  (2026-08-11T18:16:59Z)
• WorldSimProbe: Diagnosing Simulator Faithfulness in Action-Conditioned World Models for Embodied Manipulation [85.15806404962193]
  Action-conditioned world models (ACWMs) promise to provide embodied AI with scalable predictive simulators for planning, policy evaluation, and data generation.<n>We evaluate ACWMs through the observable capabilities expected of physical simulators.<n>WorldSimProbe reveals systematic action-realization degradation across control variation, structured failures in interaction grounding and dynamics.
  arXiv  Detail & Related papers  (2026-08-10T08:48:06Z)
• Bringing GRACE to Recommendation: Fine-Tuning for Sustainable and Accurate Personalization [57.58261252242669]
  Growing concern about environmental sustainability has motivated green'' recommender systems that steer users toward more eco-friendly and healthier choices.<n>Many existing green recommendation approaches require training new models from scratch, incurring substantial computational and energy costs.<n>We propose GRACE (Green Recommendation via Adaptive Conflict-rEsolution), a fine-tuning framework that integrates item-level sustainability signals into pretrained recommendation models.
  arXiv  Detail & Related papers  (2026-07-24T14:17:37Z)
• A Unified Python Framework for Direct PPO-based Control of AHUs with Economizer Logic and CO2-Constrained Ventilation [0.0]
  This paper presents a new approach to optimizing HVAC systems through the use of Deep Reinforcement Learning (DRL) algorithms.<n>The DRL system uses a second order-capacitor thermal model and an integrated dynamic mass balance of CO2 to replicate the complex physics associated with buildings.<n>An enthalpy-based economiser is used to create free cooling from the outdoor environment.
  arXiv  Detail & Related papers  (2026-05-23T05:31:09Z)
• Physicochemical-Neural Fusion for Semi-Closed-Circuit Respiratory Autonomy in Extreme Environments [0.0]
  Galactic Bioware's Life Support System is a breathing apparatus designed for integration into a positive-pressure firefighting suit.<n>One-way exhaust valves maintain positive pressure while creating a semi-closed system in which outward venting gradually depletes the gas inventory.<n>Part I develops the physicochemical foundations from first principles, including state-consistent thermochemistry.<n>Part II introduces an AI control architecture that fuses three sensor tiers, external environmental sensing, internal suit atmosphere sensing.
  arXiv  Detail & Related papers  (2026-03-16T20:29:26Z)
• Explainable AI for Smart Greenhouse Control: Interpretability of Temporal Fusion Transformer in the Internet of Robotic Things [0.0]
  This study leverages the Temporal Fusion Transformer (TFT) model to automate actuator settings for optimal greenhouse management.<n>To enhance interpretability and trust in the model decision-making process, both local and global explanation techniques were employed.<n>The trained TFT model achieved a test accuracy of 95% on a class-imbalanced dataset for actuator control settings in an automated greenhouse environment.
  arXiv  Detail & Related papers  (2025-12-04T19:41:00Z)
• GreenLight-Gym: Reinforcement learning benchmark environment for control of greenhouse production systems [0.0]
  GreenLight-Gym is a fast, open-source benchmark environment for developing reinforcement learning methods in greenhouse crop production control.<n>It features a differentiable C++ implementation leveraging the CasADi framework for efficient numerical integration.<n>A modular Python environment wrapper enables flexible configuration of control tasks and RL-based controllers.
  arXiv  Detail & Related papers  (2024-10-06T18:25:23Z)
• Global Transformer Architecture for Indoor Room Temperature Forecasting [49.32130498861987]
  This work presents a global Transformer architecture for indoor temperature forecasting in multi-room buildings. It aims at optimizing energy consumption and reducing greenhouse gas emissions associated with HVAC systems. Notably, this study is the first to apply a Transformer architecture for indoor temperature forecasting in multi-room buildings.
  arXiv  Detail & Related papers  (2023-10-31T14:09:32Z)
• Autonomous Payload Thermal Control [55.2480439325792]
  In small satellites there is less room for heat control equipment, scientific instruments, and electronic components. An autonomous thermal control tool that uses deep reinforcement learning is proposed for learning the thermal control policy onboard. The proposed framework is able to learn to control the payload processing power to maintain the temperature under operational ranges.
  arXiv  Detail & Related papers  (2023-07-28T09:40:19Z)
• Data-driven control of micro-climate in buildings: an event-triggered reinforcement learning approach [56.22460188003505]
  We formulate the micro-climate control problem based on semi-Markov decision processes. We propose two learning algorithms for event-triggered control of micro-climate in buildings. We show the efficacy of our proposed approach via designing a smart learning thermostat.
  arXiv  Detail & Related papers  (2020-01-28T18:20:43Z)
• NeurOpt: Neural network based optimization for building energy management and climate control [58.06411999767069]
  We propose a data-driven control algorithm based on neural networks to reduce this cost of model identification. We validate our learning and control algorithms on a two-story building with ten independently controlled zones, located in Italy.
  arXiv  Detail & Related papers  (2020-01-22T00:51:03Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.