[CFP] (Submissions due 31 December) - Embodied-AI for Sustainable and Intelligent Phytoprotection

 

Manuscript Submission Deadline 31 December 2026

 

 

 

 

The growing challenges of global food security, ecological sustainability, and climate change call for advanced technological solutions in plant protection. Conventional approaches often rely on limited data sources and lack the adaptability to address the complex and dynamic interactions among crops, pests, and environmental factors. Recent advances in artificial intelligence (AI), multimodal sensing, and distributed computing offer unprecedented opportunities to transform plant protection into an sustainable and intelligent system. By integrating multimodal data—such as visual imagery, spectral signals, text-based agronomic records, and time-series sensor measurements—with embodied AI systems capable of real-world interaction, we can enable more accurate diagnostics, timely interventions, and resource-efficient management. This Research Topic aims to explore how distributed signal processing, multimodal fusion, and embodied intelligence can be combined to build next-generation sustainable and intelligent phytoprotection systems that are scalable, adaptive, and ecologically aware.

 

While AI and sensing technologies have shown great potential in agriculture, key research gaps remain in integrating heterogeneous data streams and deploying intelligent systems in real-world field conditions. This Research Topic seeks to address these challenges by fostering research that bridges multimodal data fusion, distributed decision-making, and embodied AI for sustainable plant protection. Specific goals include: developing novel methods for fusing image, text, and time-series data from distributed sensor networks and drones; designing resource-aware learning algorithms that operate under communication and energy constraints in wireless sensor networks; and creating embodied AI agents—such as autonomous robots, drones, and solar insecticidal lamp (SIL)—that can perceive, reason, and act in complex agricultural environments. We also aim to explore how such systems can support non-destructive quality testing, precision spraying, and integrated pest management while minimizing ecological disruption.

 

We welcome contributions from researchers in signal processing, AI, computer vision, IoT, robotics, agronomy, plant science, and related fields. Topics of interest include, but are not limited to:

• Multimodal data fusion for detection and diagnosis of crop diseases, insect pests, weed damage, bird damage, and wind damage – integrating visual, spectral, textual, and temporal data using graph neural networks, distributed learning, or attention-based models.

• Resource-constrained intelligence in wireless sensor networks – distributed signal processing, in-network computing, and adaptive sampling for efficient data collection and model inference.

• Embodied AI for field operations – robotic, drones, and SIL with multimodal perception, navigation, and manipulation capabilities for targeted treatment and monitoring.

• Intelligent non-destructive testing and quality assessment – using hyperspectral imaging, deep learning, and sensor fusion to evaluate plant health and product quality.

• Adaptive and lifelong learning in dynamic environments – online learning, domain adaptation, and reinforcement learning for continuously improving plant protection models.

• Human-AI collaboration and decision support systems – interactive tools that combine expert knowledge and AI recommendations for crop diseases, insect pests, weed damage, bird damage, and wind damage.