US 11,636,424 B1 · System and method to detect failures at a subset of load cells

Abstract

Load cells are used in a fixture of a facility to acquire weight data. However, some of the load cells used in the fixture may fail. One way to test a subset of load cells is to compare data from a first subset of load cells to a second subset of load cells using a linear regression. For example, the linear regression may help determine predicted values for the first subset of load cells. Actual values from the first subset of load cells are then compared to the predicted values. If the differences between the actual values and predicted values for the first subset of load cells are greater than a threshold, a determination is made that the first subset of load cells has failed. Weight data from the second subset of load cells may then be used to determine hypotheses for describing interactions at the fixture.

Claims20

Dependent claims

A method comprising:

accessing first weight data obtained by a first subset of load cells of a fixture;
accessing second weight data obtained by a second subset of load cells of the fixture, wherein the second subset of load cells is mutually exclusive of the first subset of load cells;
comparing the first weight data and the second weight data using at least a linear regression;
determining a failure of the first subset of load cells of the fixture based on the comparing;
generating one or more hypotheses for describing an interaction at the fixture based on the second weight data;
selecting one hypothesis from the one or more hypotheses to describe the interaction at the fixture, wherein the interaction comprises one or more of picking an item from the fixture or placing the item at the fixture;
and
updating an inventory quantity of the item based on the one hypothesis that is selected.

5 more elements

A system comprising:

a shelf;
a plurality of load cells configured to acquire weight data associated with the shelf;
and
a computing device comprising:
a memory storing first computer-executable instructions;
and
a hardware processor to execute the first computer-executable instructions to:
acquire first weight data from a first subset of the plurality of load cells;
acquire second weight data from a second subset of the plurality of load cells;
compare the first weight data and the second weight data using at least a linear regression;
determine a failure of the first subset of the plurality of load cells based on the comparison of the first weight data and the second weight data;
generate one or more hypotheses for describing an interaction at the shelf using the second weight data;
and
initiate one or more mitigation actions responsive to the failure of the first subset of the plurality of load cells, the one or more mitigation actions comprising one or more of:
disregarding the first weight data from the first subset of the plurality of load cells,
deactivating the first subset of the plurality of load cells, or
generating a notification associated with the failure of the first subset of the plurality of load cells.

14 more elements

Dependencies are inferred from the claim text.

System and method to detect failures at a subset of load cells

Life of this patent

System and method to detect failures at a subset of load cells

Life of this patent

Abstract

Claims20