Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces AgriJetsonBench, a comprehensive benchmarking tool designed to evaluate agricultural vision models on NVIDIA Jetson edge platforms, focusing on critical deployment factors such as power consumption and sustained performance. By utilizing existing agricultural datasets and exporting models through ONNX and TensorRT, the study provides a detailed analysis of model performance under various operational conditions. Key findings reveal that the Jetson Orin Nano Super significantly outperforms the AGX Orin in terms of frames per second and energy efficiency for specific workloads, highlighting the importance of tailored deployment strategies over mere hardware comparisons.
The Jetson Orin Nano Super outperforms the AGX Orin in agricultural vision tasks, achieving 145.82 FPS with YOLO11s while consuming less energy, challenging conventional hardware selection criteria.
Agricultural vision models are often selected from validation accuracy and reported frames per second, but deployment on embedded agricultural edge-GPU systems also depends on timing boundary, numeric precision, power mode, board-input energy, and sustained-run validity. We present AgriJetsonBench, a reproducible deployment benchmark for crop/weed detection and segmentation on NVIDIA Jetson AGX Orin 64GB and Jetson Orin Nano Super. Existing locked agricultural datasets were used as fixed deployment workloads, and seven model families were exported through ONNX and TensorRT. The benchmark combines pure TensorRT engine timing, external board-input power logging, internal Jetson telemetry, battery-continuation testing, run-validity screening, and archived reproducibility artifacts. Main energy tests used YOLO11s and SegFormer-B0 at batch size 1 and $640 \times 640$. Under matched 15 W INT8 operation, Orin Nano Super outperformed AGX Orin on both workloads: YOLO11s reached 145.82 FPS and 0.0767 J/inference versus 80.90 FPS and 0.1814 J/inference, and SegFormer-B0 reached 47.18 FPS and 0.2808 J/inference versus 31.95 FPS and 0.5013 J/inference. In native SegFormer-B0 FP16 operation, AGX Orin achieved higher throughput and lower p95 latency than Nano Super, 103.61 FPS and 9.441 ms versus 67.47 FPS and 15.020 ms, with similar energy per inference. Time-aligned analysis of 76,691 internal--external power samples showed high temporal association but a pooled internal-minus-external bias of $-1.988$ W. The results show that agricultural Jetson deployment decisions should be based on workload, precision, power mode, and measurement boundary rather than board class alone.