Electronic vulnerability detection and measuring system and method for susceptibility or vulnerability of truck fleet to occurring accident events

1 . A method for detecting a susceptibility or vulnerability to damage impacts caused by occurring accident events with a measurable impact strength for a carriage vehicle fleet comprising a plurality of carriage vehicles, the method comprising:

capturing, via a telematics circuit that communicates with mobile telematics devices associated with the carriage vehicles, behavioral driving data of the carriage vehicles of the carriage vehicle fleet, the mobile telematics devices comprising sensory and measuring devices at least including a GPS sensor for measuring speed and/or location parameter values at a predefined measuring frequency per time unit;

capturing, via a first data interface, the behavioral driving data of the plurality of carriage vehicle of the carriage vehicle fleet as a driving data input signal;

capturing, via a first data interface, from a carriage vehicle fleet database carriage vehicle data as a fleet input signal from the carriage vehicle fleet database, the carriage vehicle data comprising: vehicle data including physical parameter measurements of carriage vehicle characteristics for the carriage vehicles of the carriage vehicle fleet, carriage vehicle driver data including physical parameter measurements of driver characteristics, and/or carriage vehicle usage data including physical parameter measurements of carriage vehicle usage characteristics;

transmitting the fleet input signal and the driving data input signal to processing circuitry;

analyzing, by a machine learning module of the processing circuitry, the data provided by the fleet input signal and the driving data input signal of at least one carriage vehicle of the carriage vehicles by using one or more machine learning structures;

generating, by the machine learning module, a vulnerability index measure for each of the carriage vehicles of the carriage vehicle fleet, the one of the one or more machine learning structures of the machine learning module being configured to provide dimensionality reduction by selecting and extracting data variables from the carriage vehicle data, the carriage vehicle driver data, and the carriage vehicle usage data;

automatically generating, by an aggregating module of the processing circuitry, an aggregated vulnerability score measure for the carriage vehicle fleet based on the vulnerability index measures of the carriage vehicles;

automatically generating, by a forward-looking modelling module of the processing circuitry, a predicted vulnerability index measure of a predicted damage of at least one carriage vehicle of the carriage vehicle fleet by simulating a physically impacting event on the at least one carriage vehicle based on the vulnerability index measures generated by the machine learning module, the one or more machine learning structures of the machine learning module being realized as an unsupervised machine learning structure analyzing the fleet input signal and the driving data input signal, and providing validation for risk index values for the carriage vehicles of the carriage vehicle fleet, wherein the machine learning module provides validated data for the forward-looking modelling module; and

generating, by a signal generator, the aggregated vulnerability score measure as an output signal indicating a vulnerability score measure of the carriage vehicle fleet.

2 . The method according to claim 1 , further comprising automatically requesting fleet input signal data and behavioral driving data for carriage vehicles added to the carriage vehicle fleet via an application programming interface providing access to third party information data about the carriage vehicles of the carriage vehicle fleet hosted in a fleet data processing device.

3 . The method according toe claim 2 , wherein

the automatic requesting is conducted daily, and

the aggregated vulnerability score measure for the carriage vehicle fleet is updated accordingly.

4 . An electronic vulnerability detection and measuring system for detecting a susceptibility or vulnerability to damage impacts caused by occurring accident events with a measurable impact strength for a carriage vehicle fleet comprising a plurality of carriage vehicles, the system comprising:

an electronic driving monitoring system with a telematics circuit communicating with mobile telematics devices associated with the carriage vehicles of the carriage vehicle fleet, the mobile telematics devices at least capturing behavioral driving data of the carriage vehicles of the carriage vehicle fleet, and the mobile telematics devices comprising sensory and measuring devices at least including a GPS sensor for measuring speed and/or location parameter values at a predefined measuring frequency per time unit;

a first data interface associated with a first data access device configured to capture the behavioral driving data of the plurality of carriage vehicle of the carriage vehicle fleet as a driving data input signal, the first data access device being connected to the telematics circuit or at least one of the carriage vehicles;

a second data interface associated with a second data access device to a carriage vehicle fleet database for capturing carriage vehicle data as a fleet input signal from the carriage vehicle fleet database, the carriage vehicle data comprising: vehicle data including physical parameter measurements of carriage vehicle characteristics for the carriage vehicles of the carriage vehicle fleet, carriage vehicle driver data including physical parameter measurements of driver characteristics, and/or carriage vehicle usage data including physical parameter measurements of carriage vehicle usage characteristics;

processing circuitry configured to receive the fleet input signal and the driving data input signal, the processing circuitry implementing:

a machine learning module configured to analyze the data provided by the fleet input signal and the driving data input signal of at least one carriage vehicle of the carriage vehicles by using one or more machine learning structures and generating a vulnerability index measure for each of the carriage vehicles of the carriage vehicle fleet, the one of the one or more machine learning structures of the machine learning module being configured to provide dimensionality reduction by selecting and extracting data variables from the carriage vehicle data, the carriage vehicle driver data, and the carriage vehicle usage data,

an aggregating module configured to automatically generate an aggregated vulnerability score measure for the carriage vehicle fleet based on the vulnerability index measures of the carriage vehicles, and

a forward-looking modelling module configured to automatically generate a predicted vulnerability index measure of a predicted damage of at least one carriage vehicle of the carriage vehicle fleet by simulating a physically impacting event on the at least one carriage vehicle based on the vulnerability index measures generated by the machine learning module, the one or more machine learning structures of the machine learning module being realized as an unsupervised machine learning structure analyzing the fleet input signal and the driving data input signal, and providing validation for risk index values for the carriage vehicles of the carriage vehicle fleet, wherein the machine learning module provides validated data for the forward-looking modelling module; and

a signal generator configured to provide the aggregated vulnerability score measure as an output signal indicating a vulnerability score measure of the carriage vehicle fleet.

5 . The system according to claim 4 , wherein the predefined measuring frequency per time unit is 30 per seconds.

6 . The system according to claim 4 , wherein the vulnerability index measure is generated as a forecasted probability value for an occurrence of an accident event at a risk measure range between 0 and 1.

7 . The system according to claim 4 , wherein the electronic driving monitoring system is designed as an Advanced Driver Assistance System (ADAS), an intelligent driving monitoring system (IDMS), an autonomous driving system (ADS), and/or on-board diagnostics (OBD) system.

8 . The system according to claim 4 , wherein the electronic driving monitoring system is designed to capture physical parameter measurements of the carriage vehicle characteristics, driver characteristics, and/or carriage vehicle usage characteristics based on advanced driver assistance functionalities including mean speed, maximum speed, percentiles of speed, fatigue driving hours per specified distance, night driving, rush hour driving, urban driving, rural driving, highway driving, over speed driving, and/or warning signal rate.

9 . The system according to claim 8 , wherein the sensory and measuring devices are configured to monitor advanced driver assistance functionalities of the at least one carriage vehicle of the carriage vehicles including autonomous emergency braking, lane departure monitoring, forward collision monitoring, unsafe following monitoring, steering assistance, automatic emergency steering, cross traffic alert, adaptive cruise control, blind spot detection, crosswind stabilization, driver monitoring, and/or pedestrian detection/avoidance.

10 . The system according to claim 4 , wherein the electronic driving monitoring system includes interfaces capturing carriage vehicle parameter values of speed reduction, impact/final speed, impact position, braking distance, warning inception, ADAS feature inception, maximum braking deceleration, maximum braking time, and speed range for brake activation.

11 . The system according to claim 4 , wherein the forward-looking modelling module is configured to generate a predicted risk index value and/or a predicted warning rate level based on the physical parameter measurements of the carriage vehicle characteristics, the driver characteristics, and/or the carriage vehicle usage characteristics using a Markov chain model.

12 . The system according to claim 11 , wherein the machine learning module comprises a deep neural network structure configured to generate the risk index values for the carriage vehicles of the carriage vehicle fleet.

13 . The system according to claim 4 , wherein the carriage vehicle data, and/or the carriage vehicle driver data, and/or the carriage vehicle usage data at least comprise (i) speed and location measured by the GPS sensor at a frequency of 30 per seconds, (ii) warning alerts on driving behaviors comprising warnings from an Advanced Driver Assistance System (ADAS) at least on forward collision and/or unsafe following and/or lane departure, and/or warning alerts from a Driver Monitoring System (DMS) at least on fatigue measures, phone use measures and/or smoking, (iii) warning detail data at least comprising speed and acceleration at a frequency of 5 times per second, from 5 seconds prior to 5 seconds after each warning, and (iv) risk-transfer data.

14 . A digital platform providing a digital channel for automated measurement-driven risk-transfer analysis and hosting the system according to claim 4 , wherein the digital platform comprises a cloud-based infrastructure accessible via a digital network and at least one application programming interface for accessing third party information data about the carriage vehicles of the carriage vehicle fleet from a fleet data processing unit device.

15 . The digital platform according to claim 14 , wherein the platform is configured to host:

a memory configured to store risk index values, predicted risk index values, and/or risk score measures provided by the system,

a risk-transfer processing module, implemented by the processing circuitry, configured to apply the aggregated vulnerability score measure to a risk-transfer indicator of the carriage vehicle fleet, and

a communication module, implemented by the processing circuitry, configured to communicate with users of the system.