Diffusion-Based 3-D Radar Semantic Perception in Cluttered Agricultural Environments

Accurate and robust environmental perception is crucial for robot autonomous navigation. While current methods typically adopt optical sensors, such as cameras and LiDARs, as primary sensing modalities, their susceptibility to visual occlusion—such as dirt adhering to the lens or physical blockage of the sensor—often leads to degraded performance or complete system failure. In this paper, we focus on agricultural scenarios where robots are exposed to the risk of onboard sensor contamination. Leveraging radar’s strong penetration capability, we introduce a radar-based 3D environmental perception framework as a viable alternative. It comprises three core modules designed for dense and accurate semantic perception: first, parallel frame accumulation to enhance signal-to-noise ratio of raw radar data; second, a diffusion model-based hierarchical learning framework that first filters radar sidelobe artifacts then generates fine-grained 3D semantic point clouds; and third, a specifically designed sparse 3D network optimized for processing large-scale raw radar data. We conducted extensive benchmark comparisons and experimental evaluations on a self-built dataset collected in real-world agricultural field scenes. Results demonstrate that our method achieves superior structural and semantic prediction performance compared to existing methods, while simultaneously reducing computational and memory costs by 44.3% and 27.5%, respectively. Furthermore, our approach achieves complete reconstruction and accurate classification of thin structures such as poles and wires—which existing methods struggle to perceive—highlighting its potential for dense and accurate 3D radar perception.