Method and system for enabling a vehicle occupant to report a hazard associated with the surroundings of the vehicle

A hazard reporting system adapted for enabling a vehicle occupant to report a hazard associated with the surroundings of a vehicle, said hazard reporting system comprising:

• a verbal report receiving unit adapted for:
• receiving, from the vehicle occupant via one or more microphones integrated within the vehicle, a verbal hazard reporting activation command;
• activating the hazard reporting system based on the received verbal hazard reporting activation command;
• receiving, via the one or more vehicle microphones and subsequent to activating the hazard reporting system, a verbal hazard report from said vehicle occupant, said verbal hazard report comprising information related to the hazard associated with the surroundings of said vehicle;
• a voice to text (VTT) transforming unit adapted for transforming, with support from a speech recognition service, said verbal hazard report into a resulting machine-readable hazard report;
• a data table providing unit adapted for providing a data set comprising one or more hazard categories,
• wherein the data set further comprises one or more sub-categories associated with each of the one or more hazard categories,
• wherein the hazard categories and sub-categories comprise differing machine-readable elements;
• an identifying unit adapted for identifying a hazard category of said data set considered corresponding to said machine-readable hazard report by comparing the machine-readable hazard report to the machine-readable elements of the data set,
• wherein the identifying unit is further adapted to apply, in real time, a machine learning classification model associated with said data set, to said machine-readable hazard report, said machine learning classification model being adapted to classify said machine-readable hazard report into the hazard category corresponding to said machine-readable hazard report;
• and
• a determining unit adapted for:
• determining, when a hazard category of said data set considered corresponding to said machine-readable hazard report has been identified, that said corresponding hazard category represents the hazard reported in the verbal hazard report,
• determining a confidence level for the identified hazard category;
• and
• assigning the identified hazard category to the machine-readable hazard report when the confidence level exceeds a pre-set confidence level threshold.

A non-transitory computer-readable medium having stored thereon sequences of instructions which, when executed by at least one processor, cause the at least one processor to perform a method for enabling a vehicle occupant to report a hazard associated with the surroundings of a vehicle, said method comprising:

• receiving, from the vehicle occupant via one or more microphones integrated within the vehicle, a verbal hazard reporting activation command;
• activating the hazard reporting system based on the received verbal hazard reporting activation command;
• receiving, via the one or more vehicle microphones and subsequent to activating the hazard reporting system, a verbal hazard report from said vehicle occupant, said verbal hazard report comprising information related to the hazard associated with the surroundings of said vehicle;
• transforming, with support from a speech recognition service, said verbal hazard report into a resulting machine-readable hazard report;
• providing a data set comprising one or more hazard categories,
• wherein the data set further comprises one or more sub-categories associated with each of the one or more hazard categories,
• wherein the hazard categories and sub-categories comprise differing machine-readable elements;
• and
• identifying a hazard category of said received verbal hazard report by comparing the machine-readable hazard report to the machine-readable elements of the data set,
• wherein identifying the hazard category of said received verbal hazard report further comprises:
• applying, in real time, a machine learning classification model associated with said data set, to said machine-readable hazard report, said machine learning classification model being adapted to classify said machine-readable hazard report into the hazard category corresponding to said machine-readable hazard report;
• determining a confidence level for the classification;
• and
• assigning the identified hazard category to the machine-readable hazard report when the confidence level exceeds a pre-set confidence level threshold.