Science AI Journal
Agricultural and Biological Sciences · Research topic

Open research questions in Smart Agriculture and AI

664 unresolved questions extracted from the limitations and future-work sections of 1,057 Smart Agriculture and AI papers in our library. Each links back to the study that raised it.

What the literature leaves open

  • Exploring the transferability of the models to other crops, - Investigating the use of other optimization algorithms for the model, - Developing a more robust model that can handle variations in lighting, background noise, and image quality

    Deep learning technique for grape leaf disease classification via deep high attention stage-by-stage forward Taylor network · 2026 · DOI

  • The paper identifies a gap in existing models for classifying grape leaf diseases, including limitations in handling variations in lighting, background noise, and image quality. The paper also mentions the need for a method that can consider key environmental factors to enhance predictive precision.

    Deep learning technique for grape leaf disease classification via deep high attention stage-by-stage forward Taylor network · 2026 · DOI

  • The paper identifies the challenge of premature convergence, which is addressed through Adaptive Genetic Optimization. The paper also highlights the need for a model that can generalize across varying lighting, noise, and environmental conditions. The paper discusses the challenge of balancing the learning of spatial and contextual features related to maize leaf blight classification.

    Hybrid CNN–Transformer Model for Maize Leaf Blight Classification Using Adaptive Genetic Optimization · 2026 · DOI

  • The paper identifies the limitations of conventional diagnosis methods, which are time-consuming, subjective, and prone to human error. The paper also highlights the need for a model that can combine spatial and contextual features for improved classification.

    Hybrid CNN–Transformer Model for Maize Leaf Blight Classification Using Adaptive Genetic Optimization · 2026 · DOI

  • Exploring the application of machine learning in agriculture for multiple environmental factors. Developing models that can handle multiple outputs and provide accurate predictions. Investigating the use of other machine learning algorithms for predicting agricultural environmental indices.

    Multi-Output Linear Regression Model for Real-Time Prediction of Agricultural Environmental Indices in Smart Farming Applications · 2026 · DOI

  • Many works focus on single-output prediction models that target a single agricultural variable at a time. These models do not account for interdependencies among multiple environmental factors. There is a need for a model that can predict various critical variables simultaneously.

    Multi-Output Linear Regression Model for Real-Time Prediction of Agricultural Environmental Indices in Smart Farming Applications · 2026 · DOI

  • The lack of consistent measurement of data infrastructure. The need for rigorous identification of causal effects at the county level. The complexity of evaluating the underlying impact pathways of digital infrastructure.

    Impact of digital infrastructure on agricultural green transformation: theory and China’s experience · 2026 · DOI

  • Few studies have focused on digital-related influencing factors. Existing research has not taken a perspective based on digital infrastructure to achieve a rigorous identification of causal effects. The underlying impact pathways of digital infrastructure on agricultural green transformation are not systematically evaluated.

    Impact of digital infrastructure on agricultural green transformation: theory and China’s experience · 2026 · DOI

  • Conventional deep learning models often fail to balance local feature extraction with global context. Traditional manual inspection of diseased leaves is slow, inconsistent, and error-prone. The need for a model that can capture both local and global features for accurate disease classification.

    LXViT: a hybrid hierarchical CNN-ViT framework for lemon leaf disease classification with multi-method explainable AI integration · 2026 · DOI

  • 76% backbones demonstrated competent performance, they were limited by their respective lack of local inductive Article in PressScientific Reportshttps://doi.

    LXViT: a hybrid hierarchical CNN-ViT framework for lemon leaf disease classification with multi-method explainable AI integration · 2026 · DOI

  • the lack of a non-destructive identification method for cultivated ginseng growth years - the need for a method to combine leaf hyperspectral and RGB image features - the limitation of prior methods in achieving high accuracy for growth year identification

    A growth-year classification model for cultivated ginseng based on feature–decision-level fusion of leaf hyperspectral and RGB image features · 2026 · DOI

  • Investigating the application of DA-OHB to other types of IoT equipment; Developing methods to address node heterogeneity and rare-fault imbalance; Exploring the use of additional data sources, such as maintenance logs and physical inspection records

    Cross-Node Fault Diagnosis of Solar Insecticidal Lamp IoT Equipment for Reliable Precision Pest Monitoring Using a Diagnosability-Aware Health Baseline · 2026 · DOI

  • The scarcity and imbalance of labelled faults hinder the development of effective diagnosis methods. Existing methods do not account for the complexity of variable interactions and the directionality of faults. There is a need for a diagnosability-aware health baseline that combines data-observable features with operating-state gates and direction-sensitive evidence scores.

    Cross-Node Fault Diagnosis of Solar Insecticidal Lamp IoT Equipment for Reliable Precision Pest Monitoring Using a Diagnosability-Aware Health Baseline · 2026 · DOI

  • sensing–structure co-design - flexible functional materials - task-oriented multimodal sensing - transferable state estimation - standardized evaluation frameworks

    Sensing Technologies in Robotic Manipulators for Low-Damage Fruit and Vegetable Grasping: Principles, Integration, and Applications · 2026 · DOI

  • a substantial gap remains between laboratory-scale demonstrations and robust, long-term deployment across diverse objects and complex environments - the coordination between manipulation and internal quality assessment requires further investigation - the pronounced biological variability of fruits and vegetables poses a major challenge to the generalization of sensing models

    Sensing Technologies in Robotic Manipulators for Low-Damage Fruit and Vegetable Grasping: Principles, Integration, and Applications · 2026 · DOI

  • High initial costs and technical complexities of operating drones, - Limited operational time due to battery constraints, - Limited payload capacity of agricultural drones

    A Review of Indian-Based Drones in the Agriculture Sector: Issues, Challenges, and Solutions · 2025 · DOI

  • The gap between potential and practice of drone adoption in Indian agriculture. Regulatory barriers and high costs hinder widespread implementation of drones. Limited awareness and technical complexities of operating drones present major barriers to wider adoption among small-scale farmers.

    A Review of Indian-Based Drones in the Agriculture Sector: Issues, Challenges, and Solutions · 2025 · DOI

  • The need for a reliable weed detection system that is accessible to a wide range of end users. The lack of cost-efficient sensor systems for weed management.

    Weed Detection from Unmanned Aerial Vehicle Imagery Using Deep Learning—A Comparison between High-End and Low-Cost Multispectral Sensors · 2024 · DOI

  • The general- izability of the model is limited by the present dataset, which included one genotype, four clonal greenhouse-grown trees, one sampling period, and one preparation/imaging workflow.

    Explainable growth stage classification of cacao (Theobroma cacao L.) leaves and key feature visualization using vision transformer and transfer learning · 2026 · DOI

  • Such paired data are limited because organ annotation is expensive, and following the same plants over time requires repeated, registered imaging.

    OneGrow: Unified Temporal Plant Image and Mask Generation · 2026 · DOI

  • This enables adaptive and differentiated management, in which pesticides are used only as a last resort when biological regulation processes are insufficient.

    Machines agricoles et numérique au service de l'agroécologie pour repenser la protection des cultures au-delà des seuls pesticides · 2026 · DOI

  • Yet, their sustainability is threatened by two underexplored challenges: vulnerability to cyberattacks and instability under fluctuating energy supplies.

    Sustainable Smart Farm Networks: A Decision Theory-Guided Deep Reinforcement Learning Approach · 2026 · DOI

  • Traditional centralized deep learning approaches, although effective, are constrained by data privacy concerns, limited data sharing, and poor generalization under non-identically distributed (non-IID) data scenarios.

    Federated EfficientNet for soybean leaf disease classification using a modified MOON framework · 2026 · DOI

  • Research limitations/implications Future research should explore dynamic models that incorporate real-time behavioral data to expand the analysis to more diverse agricultural ecosystems.

    Customer influence on value delivery in agricultural product development: the Brazilian farmers' perspective · 2026 · DOI

  • Artificial intelligence (AI) is increasingly shaping how agrifood systems function in the United States, yet the role of federal policy in guiding its use, oversight, and broader consequences is still not well defined.

    How U.S. Federal Artificial Intelligence (AI) policy is shaping agrifood systems: an integrative review · 2026 · DOI

Most-cited papers in Smart Agriculture and AI

Most recent work

Find a gap in your own Smart Agriculture and AI sub-topic

This page shows what the Smart Agriculture and AI literature already flags as unresolved. To narrow it to your specific question, run the guided finder — it searches the gap library on demand and checks candidates against 250M+ OpenAlex works.

Open the Research Gap Finder →

Related topics in Agricultural and Biological Sciences

Agricultural Development and Management
535 open questions
Agricultural Innovations and Practices
472 open questions
Rural development and sustainability
453 open questions
Soil Carbon and Nitrogen Dynamics
396 open questions
Land Rights and Reforms
374 open questions
Agriculture, Land Use, Rural Development
350 open questions
Plant Molecular Biology Research
300 open questions
Plant-Microbe Interactions and Immunity
295 open questions
664 open questions have been extracted from the limitations and future-work passages of 1,057 Smart Agriculture and AI papers in our library. Each one below links back to the study that raised it, so you can read the original claim in context.

Tools for your next paper

Gaps & Literature SearchFind open questions in any field — 120,000+ open gaps mined from our own library.Pre-CheckIs your paper ready? A Tier 1–5 acceptance probability in seconds.Journal FinderA ranked shortlist of target journals from a 17,500-venue index.Duplicate Publication CheckerCheck prior-publication & salami-slicing risk across 8 sources.Citation GeneratorFreeA DOI or title → APA, MLA, Vancouver, Chicago, BibTeX, RIS. No signup.Graphical Abstract MakerFreeTurn your findings into a colorblind-safe graphical abstract.

Compare the category — Honest roundups of the AI research tools, ours listed alongside the alternatives.

Command palette

Jump anywhere, run any action.