Cognitive sensor fusion management

A system, comprising:

• one or more processors;
• and
• a non-transitory computer-readable storage medium containing instructions which, when executed on the one or more processors, cause the one or more processors to perform operations including:
• obtaining one or more data streams from one or more sensors in a cognitive sensor fusion management system;
• automatically identifying a type of data and related data sources for a sensor fusion management model of the cognitive sensor fusion management system;
• selecting a subset of data streams from the one or more data streams based on the identified type of data and related data sources for the sensor fusion management model, wherein the subset of data streams are obtained from sensors having different types and capability, and wherein the selecting is further based on evaluating a respective capability of a sensor associated with each respective data stream of the subset of data streams;
• determining whether a combination of the subset of data streams comprises sufficient information for achieving a desired level of accuracy associated with implementing the sensor fusion management model using the combination of the subset of data streams, the determining based on using a machine learning network to analyze the combination of the subset of data streams and the desired level of accuracy against a power consumption at each respective sensor of a subset of sensors included in the one or more sensors;
• and
• determining one or more modifications to at least the subset of data streams or one or more sensors based on whether the combination of the subset of data streams comprises sufficient information for achieving the desired level of accuracy.

A non-transitory machine-readable storage medium, including instructions configured to cause a data processing apparatus to perform operations including:

• obtaining one or more data streams from one or more sensors in a cognitive sensor fusion management system;
• automatically identifying a type of data and related data sources for a sensor fusion management model of the cognitive sensor fusion management system;
• selecting a subset of data streams from the one or more data streams based on the identified type of data and related data sources for the sensor fusion management model, wherein the subset of data streams are obtained from sensors having different types and capability, and wherein the selecting is further based on evaluating a respective capability of a sensor associated with each respective data stream of the subset of data streams;
• determining whether a combination of the subset of data streams comprises sufficient information for achieving a desired level of accuracy associated with implementing the sensor fusion management model using the combination of the subset of data streams, the determining based on using a machine learning network to analyze the combination of the subset of data streams and the desired level of accuracy against a power consumption at each respective sensor of a subset of sensors included in the one or more sensors;
• and
• determining one or more modifications to at least the subset of data streams or one or more sensors based on whether the combination of the subset of data streams comprises sufficient information for achieving the desired level of accuracy.