Method and system for locating objects within a master space using machine learning on RF radiolocation

A object location system comprising:

• a processor;
• and
• a memory storing instructions that, when executed by the processor cause the processor to:
• receive a transmitted tag package from a Radio Frequency (RF) tag associated with an object, the transmitted tag package comprising a representation of a plurality of RF signals received by the RF tag from a respective plurality of RF signal sources;
• access a plurality of machine-learning data models, each of the plurality of machine-learning data models associated with a respective one of a plurality of subspaces, and each one of the plurality of machine-learning data models trained by:
• generating a session identifier for a measurement session;
• retrieving a customer site number associated with a master space;
• identifying a zone label and coordinates for one of the plurality of subspaces within the master space based on customer site information;
• associating the zone label with a location number;
• storing the session identifier, zone label, coordinates, and location number in a location list;
• generating a scan identifier for a radio frequency (RF) signal sampling operation;
• collecting RF signal strength indicator (RSSI) values from a plurality of RF beacons within the one of the plurality of subspaces;
• associating the collected RSSI values with the location number;
• generating a measurement log comprising the session identifier, zone label, scan identifier, and collected RSSI values;
• and
• processing the measurement log to create the one of the plurality of machine-learning data models, the one of the plurality of machine-learning data models associated with the one of the plurality of subspaces;
• provide the representation of RF signals to each of the plurality of machine-learning data models;
• receive, from each of the plurality of machine-learning data models, a correlation between the representation of RF signals and that machine-learning data model;
• determine a highest correlation of the received correlations;
• select, from the plurality of machine-learning models, a candidate machine-learning data model from which the highest correlation was received;
• and
• predict that the object is located in a subspace of the plurality of subspaces associated with the candidate machine-learning data model.