Groundnut is one of the important oilseed crops worldwide, and India is ranked as the second-biggest producer of groundnuts. This crop can be affected by numerous diseases, which are key considerations contributing to quality degradation and productivity loss. Therefore, designing more reliable and automated solutions using machine learning (ML) and deep learning (DL) models to detect groundnut leaf disease becomes essential. With this motivation, this study presents a Robust Groundnut Leaf Disease Detection using Orca Predation Algorithm with Deep Learning (RGLDD-OPADL) method. The RGLDD-OPADL algorithm focuses on the classification and recognition of groundnut leaf diseases. In the presented RGLDD-OPADL method, different stages of operations are involved. To do this, the presented RGLDD-OPADL technique comprises median filtering (MF) approach to eliminate the noise. Besides, the SqueezeNet model is used to extract a set of features. The RGLDD-OPADL technique exploits the deep recurrent neural network (DRNN) model for disease detection and classification. Finally, the OPA-based hyperparameter optimization approach is utilized to select the DRNN parameters. The experimental outcomes of the RGLDD-OPADL methodology are tested on a benchmark dataset. The comprehensive comparison analysis portrayed the enhanced performance of the RGLDD-OPADL method in terms of different measures.