Automated irrigation system adjusting to watering errors determined based on look-ahead and look-behind sensor data to reduce water usage and increase agricultural crop yield

A controller of an automated irrigation system, the automated irrigation system including a movable structure that is movable at least in a forward direction and has a plurality of nozzles for spraying a plurality of water to irrigate an aimed area of ground adjacent a current position of the movable structure, the controller comprising:

• one or more communication interfaces;
• one or more storage devices;
• and
• one or more processors;
• wherein the one or more communication interfaces are coupled to one or more nozzle actuators, each of the nozzle actuators being coupled to one or more of the nozzles and controlling an amount of water per unit time that is sprayed by the one or more of the nozzles to which the nozzle actuator is coupled according to one or more nozzle control signals;
• the one or more communication interfaces are further coupled to one or more sensors mounted to the movable structure, the one or more sensors firstly detecting one or more aspects in an unwatered area of ground toward which the movable structure is moving that is in front of the aimed area of ground, and secondly detecting the one or more aspects in a watered area of ground that is behind the aimed area of ground;
• and
• by executing a plurality of software instructions loaded from the one or more storage devices, the one or more processors are configured to:
• select a nozzle and determine a location in the unwatered area of ground that will be sprayed with water by the nozzle after the movable structure has moved in the forward direction;
• determine a target moisture characteristic of the location according to a target moisture data stored in the one or more storage devices;
• receive a look-ahead set of sensor data for the location from the one or more sensors;
• utilize a predictive model that at least takes into account the look-ahead set of sensor data and a prior look-behind set of sensor data from a last time the location was watered to generate a watering prescription for the nozzle;
• generate the one or more nozzle control signals such that the nozzle sprays water onto the location according to the watering prescription after the movable structure has moved in the forward direction and the unwatered area of ground has become the aimed area of ground;
• after the movable structure has further moved in the forward direction and the aimed area of ground has become the watered area of ground, receive an updated look-behind set of sensor data for the location from the one or more sensors;
• and
• update the predictive model for use a next time the movable structure returns to water the location, the predictive model being updated at least according to the updated look-behind set of sensor data and the target moisture characteristic for the location.

A method of controlling an automated irrigation system, the automated irrigation system including a movable structure that is movable at least in a forward direction and has a plurality of nozzles for spraying a plurality of water to irrigate an aimed area of ground adjacent a current position of the movable structure, the method comprising:

• communicating by one or more processors with one or more nozzle actuators, each of the nozzle actuators being coupled to one or more of the nozzles and controlling an amount of water per unit time that is sprayed by the one or more of the nozzles to which the nozzle actuator is coupled according to one or more nozzle control signals;
• communicating by the one or more processors with one or more sensors mounted to the movable structure, the one or more sensors firstly detecting one or more aspects in an unwatered area of ground toward which the movable structure is moving that is in front of the aimed area of ground, and secondly detecting the one or more aspects in a watered area of ground that is behind the aimed area of ground;
• selecting by the one or more processors a nozzle and determine a location in the unwatered area of ground that will be sprayed with water by the nozzle after the movable structure has moved in the forward direction;
• determining by the one or more processors a target moisture characteristic of the location according to a target moisture data stored in the one or more storage devices;
• receiving by the one or more processors a look-ahead set of sensor data for the location from the one or more sensors;
• utilizing by the one or more processors a predictive model that at least takes into account the look-ahead set of sensor data and a prior look-behind set of sensor data from a last time the location was watered to generate a watering prescription for the nozzle;
• generating by the one or more processors the one or more nozzle control signals such that the nozzle sprays water onto the location according to the watering prescription after the movable structure has moved in the forward direction and the unwatered area of ground has become the aimed area of ground;
• after the movable structure has further moved in the forward direction and the aimed area of ground has become the watered area of ground, receiving by the one or more processors an updated look-behind set of sensor data for the location from the one or more sensors;
• and
• updating by the one or more processors the predictive model for use a next time the movable structure returns to water the location, the predictive model being updated at least according to the updated look-behind set of sensor data and the target moisture characteristic for the location.