Systems and methods detecting use of mounted phones in motor vehicles

A system for detecting physical interaction with a mounted electronic device, the system comprising:

• one or more processors;
• and
• a non-transitory computer-readable medium storing instructions which when executed by the one or more processors, cause the one or more processors to perform operations including:
• receiving gyroscopic data from one or more gyroscopic sensors incorporated within the electronic device, the gyroscopic data including a set of values corresponding to measurements output from each of the one or more gyroscopic sensors, each measurement being relative to at least one axis of a plurality of axes;
• determining a frequency composition from the set of values for at least one axis of the plurality of axes;
• defining a set of frequency bands for the at least one axis of the plurality of axes based on the frequency composition, wherein each frequency band of the set of frequency bands represents a portion of the frequency composition;
• calculating, for each frequency band of the set of frequency bands, a relative magnitude of an energy value associated with the portion of the frequency composition represented by the frequency band;
• generating an input for a classifier model, the input including the relative magnitude corresponding to each frequency band of the set of frequency bands and one or more ancillary data values;
• receiving an output from the classifier model in response to the input, the output including a probability value indicating a probability that a user is physically interacting with the electronic device;
• determining that the probability value exceeds a threshold;
• and
• transmitting, in response to determining that the probability value exceeds the threshold, an indication that the user is physically interacting with the electronic device to a remote device.

A non-transitory computer-readable medium storing instructions which when executed by one or more processors, cause the one or more processors to perform operations including:

• receiving gyroscopic data from one or more gyroscopic sensors incorporated within an electronic device, the gyroscopic data including a set of values corresponding to measurements output from each of the one or more gyroscopic sensors, each measurement being relative to at least one axis of a plurality of axes;
• determining a frequency composition from the set of values for at least one axis of the plurality of axes;
• defining a set of frequency bands for the at least one axis of the plurality of axes based on the frequency composition, wherein each frequency band of the set of frequency bands represents a portion of the frequency composition;
• calculating, for each frequency band of the set of frequency bands, a relative magnitude of an energy value associated with the portion of the frequency composition represented by the frequency band;
• generating an input for a classifier model, the input including the relative magnitude corresponding to each frequency band of the set of frequency bands and one or more ancillary data values;
• receiving an output from the classifier model in response to the input, the output including a probability value indicating a probability that a user is physically interacting with the electronic device;
• determining that the probability value exceeds a threshold;
• and
• transmitting, in response to determining that the probability value exceeds the threshold, an indication that the user is physically interacting with the electronic device to a remote device.

A method of detecting physical interaction with an electronic device, the method comprising:

• receiving gyroscopic data from one or more gyroscopic sensors incorporated within the electronic device, the gyroscopic data including a set of values corresponding to measurements output from each of the one or more gyroscopic sensors, each measurement being relative to at least one axis of a plurality of axes;
• determining a frequency composition from the set of values for at least one of the plurality of axes;
• defining a set of frequency bands for the at least one axis of the plurality of axes based on the frequency composition, wherein each frequency band of the set of frequency bands represents a portion of the frequency composition;
• calculating, for each frequency band of the set of frequency bands, a relative magnitude of an energy value associated with the portion of the frequency composition represented by the frequency band;
• generating an input for a classifier model, the input including the relative magnitude corresponding to each frequency band of the set of frequency bands and one or more ancillary data values;
• receiving an output from the classifier model in response to the input, the output including a probability value indicating a probability that a user is physically interacting with the electronic device;
• determining that the probability value exceeds a threshold;
• and
• transmitting, in response to determining that the probability value exceeds the threshold, an indication that the user is physically interacting with the electronic device to a remote device;
• receiving, from the remote device, a command to modify the electronic device in response to transmitting the indication;
• and
• executing the command to modify the electronic device, the modification including increasing a sampling rate of at least one gyroscope of the one or more gyroscope sensors.