agriFrame: Agricultural framework to remotely control a rover inside a greenhouse environment
Saail Narvekar, Soofiyan Atar, Vishal Gupta, Lohit Penubaku, Kavi Arya
TL;DR
agriFrame tackles the bottleneck of seasonal crop data collection by enabling a digital-twin style simulation and remote control of a greenhouse rover. The approach combines a Gazebo-based greenhouse simulator with a VPN-secured, ROS-driven control pipeline to facilitate simulator-to-real deployment. Key contributions include a scalable greenhouse simulator, a three-block network architecture, and a sim-to-real workflow that supports remote testing with live video and ROS topic exchange. The results demonstrate platform-dependent performance (e.g., $RTF$ in [0,1] and perception FPS up to 42.5 on high-end GPUs) and highlight practical limitations, pointing to ROS2 adoption and physics-enabled, larger-scale plant models for broader impact.
Abstract
The growing demand for innovation in agriculture is essential for food security worldwide and more implicit in developing countries. With growing demand comes a reduction in rapid development time. Data collection and analysis are essential in agriculture. However, considering a given crop, its cycle comes once a year, and researchers must wait a few months before collecting more data for the given crop. To overcome this hurdle, researchers are venturing into digital twins for agriculture. Toward this effort, we present an agricultural framework(agriFrame). Here, we introduce a simulated greenhouse environment for testing and controlling a robot and remotely controlling/implementing the algorithms in the real-world greenhouse setup. This work showcases the importance/interdependence of network setup, remotely controllable rover, and messaging protocol. The sophisticated yet simple-to-use agriFrame has been optimized for the simulator on minimal laptop/desktop specifications.