Systems, and methods for controlling a modular system for improved real-time yield monitoring and sensor fusion of crops in an orchard are disclosed. According to some embodiments of the invention, a modular system for improved real-time yield monitoring and sensor fusion may include a collection vehicle, a modular processing unit, a volume measurement module, a three-dimensional point-cloud scanning module, an inertial navigation system, and a post-processing server. As the collection vehicle travels through an orchard, the volume measurement module calculates volume measurements of the windrow, the three-dimensional point-cloud scanning module assembles point-clouds of each plant in the orchard, and the inertial navigation system calculates geodetic positions of the collection vehicle. The modular processing unit may fuse the collected data together and transmit the fused data set to a post-processing server. The post-processing server may process the geodetic position data for errors which may be used for geo-referencing the fused data.