As an essential part of daily life, the drastic fluctuations in agricultural commodity prices significantly impact producers’ motivation and consumers’ quality of life, further exacerbating market uncertainty and unsustainability. The ability to scientifically and effectively predict agricultural commodity prices is of great significance for the rational deployment of market mechanisms, the timely adjustment of supply chains, and the promotion of food policy adjustments. This paper proposes a sustainable hybrid model SV-PSO-BiLSTM which integrates Seasonal-Trend decomposition procedure based on Loess (STL), Variational Mode Decomposition (VMD), Particle Swarm Optimization (PSO), and Bidirectional Long Short-Term Memory (BiLSTM) neural networks. This innovative approach first performs seasonal decomposition of the original data using the STL method, then applies the VMD method for double decomposition of the residual components, reconstructs the data based on sample entropy, and finally predicts agricultural commodity market prices using the BiLSTM network model optimized by the PSO algorithm. This paper investigates the market price dynamics of four agricultural commodities (chili, garlic, ginger, and pork) and one agricultural financial derivative (soybean futures). The experimental results indicate that the proposed SV-PSO-BiLSTM hybrid model achieves average values of 0.2241 for root mean square error (RMSE), 0.1665 for mean absolute error (MAE), 0.0207 for mean absolute percentage error (MAPE), and 0.9851 for the coefficient of determination (R²). These results surpass those of other comparative models, demonstrating stronger generalization, reliability, and stability. The research findings can provide effective guidance for the reasonable regulation of agricultural commodity market prices and further promote the healthy and sustainable development of the agricultural commodity industry.

Accurate yield prediction for major grain and oilseed crops, including soybean, corn, wheat, and rice, is essential for food-security assessment and precision field management. This study presents a structured integrative review of UAV-based crop yield prediction and follows PRISMA-guided procedures for literature search, screening, and evidence synthesis. Seventy peer-reviewed studies published between 2018 and 2025 were synthesized within a “Data–Ground Truth–Model–Decision” framework. Beyond summarizing UAV platforms, sensor configurations, feature-engineering strategies, and model architectures, the review explicitly distinguishes among microplot, field, and regional prediction scales, and evaluates the characteristics and limitations of yield-label acquisition methods, including manual harvest, plot-combine harvest, and combine yield-monitor data. Existing evidence indicates that the reliability of UAV-based yield prediction depends not only on optimal image acquisition windows, multi-source feature fusion, and model architecture, but also on scale-consistent yield labels, spatially aware validation strategies, and clearly defined model outputs, such as plot-level scalar yield, field-scale yield maps, and regional yield estimates. Major bottlenecks include scale mismatch between UAV imagery and yield labels, error propagation during yield-map generation, limited cross-year and cross-region transferability, weak causal interpretability, and difficulties in deploying models under complex operational field conditions. Future research should emphasize scale-explicit benchmark datasets, quality-controlled ground-truth yield acquisition, UAV–satellite–ground data fusion, spatiotemporal deep learning, and edge–cloud collaborative systems that can translate prediction outputs into agronomic decisions. This review provides a practical pathway for developing robust, interpretable, and deployable UAV-based yield prediction systems for major grain and oilseed crops.