Feedback Based Correction System for Adaptive Management in Smart Irrigation

Smart sprinkler systems that offer precise and accurate irrigation can be impacted by various external factors that were not previously considered. The proposed system commences with a four-point calibration process wherein markers are placed at equidistant cardinal points around the sprinkler. During calibration, water jets are directed towards the markers, and any deviation is estimated using image processing techniques that are unique to each marker. These values are used to adjust the sprinkler's water jets and re-target the markers with the correction factor until the water jets accurately hit the calibration target. The proposed system stores correction factors for dry patch data points and supplies them to the machine learning model.