Multimodal Sensing for Individual Plant Phenotyping in Agriculture Robotics

The ANTONIO (Multimodal Sensing for Individual Plant Phenotyping in Agriculture Robotics) project is developed and implement a multi-sensor system to enable agri-robots to perform plant phenotyping and precision agriculture tasks. The system includes mobile sensors mounted on ground robots and drones, which provide high-throughput crop assessment and enable precision farming applications. The sensors include LIDAR, RGB/NIR cameras, stereo cameras/RGB-D sensors, multispectral cameras, thermographic vision, wheel encoders, and accelerometers/torque sensors. The sensory data is used for subsequent higher level processing steps, such as 3D mapping, situation awareness, crop assessment and recognition, and traversability assessment. The ANTONIO system helps to apply pesticides or fertilizers where needed, monitor crop health and yield estimation, inspect remote parts of the field, and enable controlled traffic farming. Sensor fusion techniques are also used to derive virtual sensors that compute information that would be too complex or expensive to obtain directly. Overall, the project provides a comprehensive solution for precision agriculture, enabling the optimization of crop yields and reducing the environmental impact of farming practices.