Agri-Sim: Agricultural Simulation Platform for Embodied Intelligence Evaluation in Greenhouse Robotics

Shuhan Shi; Zhenfeng Xue; Minghao Mei; Chao Zheng; Nan Li; Zhonghua Miao

cs.ROAug 29, 2026

Agri-Sim: Agricultural Simulation Platform for Embodied Intelligence Evaluation in Greenhouse Robotics

Authors: Shuhan Shi, Zhenfeng Xue, Minghao Mei, Chao Zheng, Nan Li, Zhonghua Miao

Organizations: School of Mechatronic Engineering and Automation, Shanghai University, Shanghai 200444, China · Sino-European School of Technology of Shanghai University, Shanghai University, Shanghai 200444, China · School of Future Technology (Institute of Artificial Intelligence), Shanghai University, Shanghai 200444, China

Abstract

Agricultural-robot development requires simulation environments that can jointly support realistic scene construction, virtual sensing, autonomous navigation, motion planning, and manipulation-task execution. This paper presents Agri-Sim, a Unity and ROS2-based simulation platform for the closed-loop development and functional evaluation of agricultural robots. The platform contains a configurable tomato-greenhouse environment, a mobile dual-arm harvesting robot, virtual RGB-D, LiDAR, IMU, and joint sensors, and a bidirectional communication interface between Unity and ROS2. Unity is responsible for scene rendering, rigid-body dynamics, collision detection, virtual sensing, and task-state execution, whereas ROS2 and MoveIt 2 provide localization, navigation, collision-aware motion planning, inverse kinematics, and trajectory generation. Autonomous greenhouse navigation and dual-arm tomato harvesting were used to evaluate the complete simulation workflow. The experiments covered virtual sensor publication, ROS2-based navigation, collision-aware motion planning, mobile-base control, tomato acquisition, inter-arm handover, and box placement. The results demonstrate that Agri-Sim supports closed-loop integration and repeatable functional evaluation of navigation and manipulation workflows in a controlled virtual greenhouse, providing a practical foundation for subsequent algorithm development and Sim-to-Real studies.

Figures & tables

Figure 1: Motivation for Agri-Sim. The platform bridges the gap between costly and difficult-to-repeat real-world development and conventional simulation with limited agricultural-scene, sensor, and system fidelity.

System or study	Primary platform	Agricultural scene	Multimodal sensing	Ground navigation	Manipulation	ROS integration	Unified closed loop
Iqbal et al. [ 12 ]	Robot simulator	Yes	Partial	Yes	–	Partial	–
Ivanovic et al. [ 13 ]	Gazebo–Blender	Yes	Yes	–	–	Yes	Partial
Agri-fly [ 30 ]	Custom UAV simulator	Yes	Yes	–	–	Partial	Partial
Gazebo Plants [ 6 ]	Gazebo	Yes	–	–	Yes	Yes	Partial
Li and Xiang [ 16 ]	Unreal Engine–ROS	Yes	Partial	Partial	–	Yes	Partial
DigiHortiRobot [ 8 ]	Digital-twin framework	Yes	Yes	Partial	Yes	Yes	Partial

Table 1: Comparison of representative agricultural robot simulation systems and the proposed Agri-Sim platform. “Partial” indicates that the capability is limited to a subsystem or is not demonstrated as part of a complete closed-loop workflow.

Figure 2: Core components of Agri-Sim and their integration in the Unity HDRP greenhouse simulation.

Figure 3: Construction of the virtual greenhouse: (a) greenhouse structural model and (b) crop-populated tomato-growing scene.

Figure 4: Environmental-control panel for configuring simulation date, time, weather, seasonal conditions, temperature, humidity, and rendering quality.

Figure 5: Mobile dual-arm harvesting robot and its principal modules: rear enclosure, omnidirectional chassis, dual-arm assembly, and lifting mechanism.

Figure 6: Representative virtual sensor outputs: (a) IMU measurements, (b) RGB observation, (c) LiDAR visualization, and (d) depth observation.

Figure 7: Bidirectional ROS2–Unity communication framework for virtual observations, target poses, robot-state feedback, and robot-motion commands.

Data	Direction	ROS2 message type
RGB/depth images	Unity $\rightarrow$ ROS2	sensor_msgs/msg/ Image
Camera parameters	Unity $\rightarrow$ ROS2	sensor_msgs/msg/ CameraInfo
Point cloud	Unity $\rightarrow$ ROS2	sensor_msgs/msg/ PointCloud2
IMU data	Unity $\rightarrow$ ROS2	sensor_msgs/msg/ Imu
Joint states	Unity $\rightarrow$ ROS2	sensor_msgs/msg/ JointState
Coordinate transforms	Unity $\rightarrow$ ROS2	tf2_msgs/msg/ TFMessage

Table 2: Principal ROS2–Unity Communication Interfaces

Item	Configuration
Unity operating system	Windows 11 Pro
Unity CPU	Intel Core i7-14700KF
Unity GPU	NVIDIA RTX 5090 GPU
ROS operating system	Ubuntu 22.04
ROS distribution	ROS2 Humble
ROS GPU	NVIDIA RTX 3080 Ti

Table 3: Principal experimental configuration.

Figure 8: Representative frames of the autonomous greenhouse-navigation sequence: (a1)–(a3) the mobile robot enters the crop-row track; (b1)–(b3) the robot traverses along the track between two crop rows; and (c1)–(c3) the robot performs a row-switching maneuver to enter the adjacent track.

Navigation outcome	Trials	Rate	95% CI
Success	45	90%	78.6–95.7%
Failure	5	10%	—
Total	50	100%	—

Table 4: Autonomous navigation results over 50 repeated trials.

Figure 9: Representative frames of the dual-arm tomato-harvesting workflow: (a) target selection; (b) pre-grasp positioning; (c) tomato grasping; (d) preparation for inter-arm handover; (e) inter-arm handover; (f) transfer toward the harvesting box; (g) placement into the box; and (h) task completion.

Outcome	Trials	Rate
Complete harvesting sequence successful	43	86%
Initial grasp failure	2	4%
Inter-arm handover failure	2	4%
Box-placement failure	1	2%
Motion-planning failure	2	4%
Total	50	100%

Table 5: Tomato-harvesting results over 50 repeated trials.

Robotic Harvesting

See all 22 papers

AgriGen: Large-Scale Scene Generation Framework for Photorealistic Agricultural Robotics Simulation
Sep 22, 2026Utkarsh Bajpai, Serge Tleiji, Cédric Pradalier +1Robotics Simulation Synthetic Data Generation for Robotics
Georgia Tech-CNRS IRL2958, France · Notre Dame University, Lebanon; work done during internship at 1
AirSimAG: A High-Fidelity Simulation Platform for Air-Ground Collaborative Robotics
Mar 24, 2026Yangjie Cui, Xin Dong, Boyang Gao +3Robotics Simulation Robotics
School of Aeronautic Science and Engineering, Beihang University, Beijing 100191, China · Hangzhou International Innovation Institute, Beihang University, Hangzhou 311115, China · Institution of Unmanned System, Beihang University, Beijing 100191, China
Integrated cloud-based architecture for robot-robot and human-robot collaboration using ROS 2--MQTT in Mediterranean Greenhouses
Jun 21, 2026F. Cañadas-Aránega, M. Muñoz, J. C. Moreno +1Robotics Multi-Robot Coordination
Department of Informatics, CIESOL, ceiA3, Universidad de Almería, Ctra.2026 Sacramento s/n, Almería, 04120, Spain · Department of Engineering, CIESOL, ceiA3, Universidad de Almería, Ctra. Sacramento s/n, Almería, 04120, Spain

Sep 22, 2026J/K move · Enter open · S save

AgriGen: Large-Scale Scene Generation Framework for Photorealistic Agricultural Robotics Simulation

Utkarsh Bajpai, Serge Tleiji, Cédric Pradalier, Stéphanie Aravecchia

Georgia Tech-CNRS IRL2958, France · Notre Dame University, Lebanon; work done during internship at 1

Robotics Simulation Synthetic Data Generation for Robotics 3D Scene Generation Agricultural Robotics

Open paper AskPDF arXiv

Agricultural robotics is advancing rapidly, yet progress remains constrained by limited field access, lack of control over field conditions, geographic variability, and seasonal crop cycles. These factors make it difficult and costly to acquire diverse agricultural datasets, resulting in limited evaluation and reduced system robustness. While other robotics domains have scaled learning and evaluation through high-fidelity simulation, agricultural robotics still lacks comparably capable tools. In this paper, we present a ROS-integrated framework, built on Isaac Sim, for large-scale procedural generation of agricultural environments. The framework supports photorealistic rendering, physics simulation, and domain randomization at scales relevant to robotics research, with built-in support for row crops, orchards, and vineyards and straightforward extensibility to additional crop categories. Project Page: https://baj31415.github.io/agrigen/

Abstract

Robotic Harvesting

Explore similar work

AgriGen: Large-Scale Scene Generation Framework for Photorealistic Agricultural Robotics Simulation

AirSimAG: A High-Fidelity Simulation Platform for Air-Ground Collaborative Robotics

Integrated cloud-based architecture for robot-robot and human-robot collaboration using ROS 2--MQTT in Mediterranean Greenhouses