arXiv subjects

Eldert van Henten

Publications and source records attributed to Eldert van Henten.

8 recordsLinked to original sources

arXiv2026

LettuceVisSim: A Simulator That Generates Lettuce Image Time-series for Vision-Based Reinforcement Learning

Vision-based reinforcement learning holds strong potential for decision-making in controlled environment agriculture (CEA). However, its development is hindered by the scarcity of labelled crop images. To address this gap, LettuceVisSim, a lettuce growth simulator that generates labelled time series of crop images, was developed and validated. The simulator contains a process-based model (PBM) for shoot dry weight dynamics, a canopy layout algorithm for deriving canopy layout representations from shoot dry weight, and a Unity rendering engine for image generation. Five findings support the simulator. First, the PBM reproduced shoot dry weight under dynamic plant-density management with $\mathrm{R}^{2}=0.84$. Second, a piecewise cubic regression mapped shoot dry weight to potential projected area with $\mathrm{R}^{2}=0.94$. Third, the canopy layout representation was validated using 12 experimental datasets each having different dynamic environmental and spacing conditions. It reproduced the ground coverage ratio dynamics observed in measured images, achieving $\mathrm{R}^{2}=0.84$ when driven by measured shoot dry weight and $\mathrm{R}^{2}=0.40$ (0.76 excluding one outlier) when driven by PBM-simulated values. Fourth, the Unity rendering engine converted canopy layout representations into RGB and segmentation images at less than 10~ms. Fifth, a demonstration showed that a lighting-control policy can be learned and applied by observing only crop images that were generated with LettuceVisSim, providing a proof of concept of vision-based reinforcement learning in CEA using LettuceVisSim.

cs.CV↗

arXiv2026

A drone that learns to efficiently find non-uniformly distributed objects in agricultural fields: from simulation to the real world

Drones are promising for data collection in precision agriculture but are limited by battery capacity. Drone paths are usually planned using full coverage planners, even though this is not always required. This paper presents a drone path planner trained with Reinforcement Learning (RL) to detect as many objects as possible with a minimal flight path length. The agent uses low-quality prior knowledge derived from a high-altitude full coverage flight as guidance. The agent was trained in simulation, modeling object distributions, drone movement, field geometry, detection errors, and uncertain prior knowledge. Combined with a flight controller and object-detection network, it controls flight direction, terminates flights, and can be deployed on a real drone. It was evaluated across six levels of realism, from pure simulation to real-world drone flights, to quantify the simulation-to-reality gap. The agent achieved a 57% shorter flight path than a full coverage planner in simulation (13% lower recall) and a 38% shorter flight path on real-world orthomosaic data (21% lower recall). In real-world drone flights, the agent found 73% and 23% of the objects in trials 1 and 2, respectively; the lower real-world performance was mainly attributed to prior knowledge quality. Although framed as a weed-detection task, the approach is expected to generalize to other agricultural tasks with non-uniformly distributed objects and tolerance to false negatives, though further research is needed before practical deployment.

cs.RO↗

arXiv2025

A Teacher-Student MPC-PPO Coupled Reinforcement Learning Framework for Winter Temperature Control of Solar Greenhouses in Northern China

Solar greenhouses are crucial infrastructure of modern agricultural production in northern China. However, highly fluctuating temperature in winter season results in poor greenhouse temperature control, which affects crop growth and increases energy consumption. To tackle these challenges, an advanced control system that can efficiently optimize multiple objectives under dramatic climate conditions is essential. Therefore, this study propose a model predictive control-coupled proximal policy optimization (MPC-PPO) control framework. A teacher-student control framework is constructed in which the MPC generating high-quality control experiences to guide the PPO agent's learning process. An adaptive dynamic weighting mechanism is employed to balance the influence of MPC experiences during PPO training. Evaluation conducted in solar greenhouses across three provinces in northern China (Beijing, Hebei, and Shandong) demonstrates that: (1) the MPC-PPO method achieved the highest temperature control performance (96.31 on a 100-point scale), with a 5.46-point improvement compared to the non-experience integration baseline, when reduced standard deviation by nearly half and enhanced exploration efficiency; (2) the MPC-PPO method achieved a ventilation control reward of 99.19, optimizing ventilation window operations with intelligent time-differentiated strategies that reduced energy loss during non-optimal hours; (3) feature analysis reveals that historical window opening, air temperature, and historical temperature are the most influential features for effective control, i.e., SHAP values of 7.449, 4.905, and 4.747 respectively; and (4) cross-regional tests indicated that MPC-PPO performs best in all test regions, confirming generalization of the method.

math.OC↗

arXiv2025

UAV-based path planning for efficient localization of non-uniformly distributed weeds using prior knowledge: A reinforcement-learning approach

UAVs are becoming popular in agriculture, however, they usually use time-consuming row-by-row flight paths. This paper presents a deep-reinforcement-learning-based approach for path planning to efficiently localize weeds in agricultural fields using UAVs with minimal flight-path length. The method combines prior knowledge about the field containing uncertain, low-resolution weed locations with in-flight weed detections. The search policy was learned using deep Q-learning. We trained the agent in simulation, allowing a thorough evaluation of the weed distribution, typical errors in the perception system, prior knowledge, and different stopping criteria on the planner's performance. When weeds were non-uniformly distributed over the field, the agent found them faster than a row-by-row path, showing its capability to learn and exploit the weed distribution. Detection errors and prior knowledge quality had a minor effect on the performance, indicating that the learned search policy was robust to detection errors and did not need detailed prior knowledge. The agent also learned to terminate the search. To test the transferability of the learned policy to a real-world scenario, the planner was tested on real-world image data without further training, which showed a 66% shorter path compared to a row-by-row path at the cost of a 10% lower percentage of found weeds. Strengths and weaknesses of the planner for practical application are comprehensively discussed, and directions for further development are provided. Overall, it is concluded that the learned search policy can improve the efficiency of finding non-uniformly distributed weeds using a UAV and shows potential for use in agricultural practice.

cs.RO↗

arXiv2025

Adaptive path planning for efficient object search by UAVs in agricultural fields

This paper presents an adaptive path planner for object search in agricultural fields using UAVs. The path planner uses a high-altitude coverage flight path and plans additional low-altitude inspections when the detection network is uncertain. The path planner was evaluated in an offline simulation environment containing real-world images. We trained a YOLOv8 detection network to detect artificial plants placed in grass fields to showcase the potential of our path planner. We evaluated the effect of different detection certainty measures, optimized the path planning parameters, investigated the effects of localization errors, and different numbers of objects in the field. The YOLOv8 detection confidence worked best to differentiate between true and false positive detections and was therefore used in the adaptive planner. The optimal parameters of the path planner depended on the distribution of objects in the field. When the objects were uniformly distributed, more low-altitude inspections were needed compared to a non-uniform distribution of objects, resulting in a longer path length. The adaptive planner proved to be robust against localization uncertainty. When increasing the number of objects, the flight path length increased, especially when the objects were uniformly distributed. When the objects were non-uniformly distributed, the adaptive path planner yielded a shorter path than a low-altitude coverage path, even with a high number of objects. Overall, the presented adaptive path planner allowed finding non-uniformly distributed objects in a field faster than a coverage path planner and resulted in a compatible detection accuracy. The path planner is made available at https://github.com/wur-abe/uav_adaptive_planner.

cs.RO↗

arXiv2025

Safe Stabilization using Nonsmooth Control Lyapunov Barrier Function

This paper addresses the challenge of safe stabilization, ensuring the system state reach the origin while avoiding unsafe regions. Existing approaches relying on smooth Lyapunov barrier functions often fail to guarantee a feasible controller. To overcome this limitation, we introduce the nonsmooth Control Lyapunov Barrier Function (NCLBF), which ensures the existence of a safe and stabilizing controller. We provide a systematic framework for designing NCLBF and feedback control strategies to achieve safe stabilization in the presence of multiple bounded unsafe regions. Theoretical analysis and simulations of both linear and nonlinear systems demonstrate the effectiveness and superiority of our approach compared to the existing smooth functions method.

eess.SY↗

arXiv2023

Reinforcement Learning Versus Model Predictive Control on Greenhouse Climate Control

Greenhouse is an important protected horticulture system for feeding the world with enough fresh food. However, to maintain an ideal growing climate in a greenhouse requires resources and operational costs. In order to achieve economical and sustainable crop growth, efficient climate control of greenhouse production becomes essential. Model Predictive Control (MPC) is the most commonly used approach in the scientific literature for greenhouse climate control. However, with the developments of sensing and computing techniques, reinforcement learning (RL) is getting increasing attention recently. With each control method having its own way to state the control problem, define control goals, and seek for optimal control actions, MPC and RL are representatives of model-based and learning-based control approaches, respectively. Although researchers have applied certain forms of MPC and RL to control the greenhouse climate, very few effort has been allocated to analyze connections, differences, pros and cons between MPC and RL either from a mathematical or performance perspective. Therefore, this paper will 1) propose MPC and RL approaches for greenhouse climate control in an unified framework; 2) analyze connections and differences between MPC and RL from a mathematical perspective; 3) compare performance of MPC and RL in a simulation study and afterwards present and interpret comparative results into insights for the application of the different control approaches in different scenarios.

math.OC↗

arXiv2018

TrimBot2020: an outdoor robot for automatic gardening

Robots are increasingly present in modern industry and also in everyday life. Their applications range from health-related situations, for assistance to elderly people or in surgical operations, to automatic and driver-less vehicles (on wheels or flying) or for driving assistance. Recently, an interest towards robotics applied in agriculture and gardening has arisen, with applications to automatic seeding and cropping or to plant disease control, etc. Autonomous lawn mowers are succesful market applications of gardening robotics. In this paper, we present a novel robot that is developed within the TrimBot2020 project, funded by the EU H2020 program. The project aims at prototyping the first outdoor robot for automatic bush trimming and rose pruning.

cs.RO↗