×

Agricultural information sensing and retrieval

  • US 10,453,018 B2
  • Filed: 10/28/2015
  • Issued: 10/22/2019
  • Est. Priority Date: 01/14/2014
  • Status: Active Grant
First Claim
Patent Images

1. An agricultural data search engine, comprising:

  • a processor generating an index of performance data, stored in a data store, the performance data being indicative of performance of a plurality of different agricultural machines having different operators, the index indexing the performance data based on machine and worksite context criteria and on performance metric values;

    a set of sensors detecting current machine and worksite context data corresponding to a selected agricultural machine;

    a performance metric generator generating a current performance metric value indicative of machine performance of the selected agricultural machine;

    a search system detecting a data retrieval trigger;

    a criteria identifier component selecting context filter criteria from a context filter criteria hierarchy based on the detected current machine and worksite context data;

    a context filter system filtering the performance data using the index and the selected context filter criteria to identify a first set of performance data;

    a score range identifier component identifying a score range, relative to the current performance metric value;

    a performance metric filter system filtering the first set of performance data using the index and the identified score range to modify the first set of performance data to identify a second set of performance data that is different from the first set of performance data;

    a data volume analysis component identifying a quantity of underlying data, in the data store, underlying the second set of performance data and from which the second set of performance data was generated and, if the quantity of underlying data is above a first quantity value, then reducing the score range and repeating filtering the first set of performance data to decrease the quantity of underlying data and, if the quantity of underlying data is below a second quantity value, then increasing the score range and repeating the step of filtering the first set of performance data to increase the quantity of underlying data;

    a data retrieval component generating a database control signal to retrieve the underlying data underlying the second set of performance data; and

    an output component returning the retrieved underlying training data based on the detected data retrieval trigger.

View all claims
  • 1 Assignment
Timeline View
Assignment View
    ×
    ×