System and method for determining recurrent and non-recurrent road congestion to mitigate congestion

Abstract

A system for mitigating congestion in a road network is provided. The system includes a plurality of motor vehicles located at associated locations in the road network. Each vehicle has one or more sensors producing inputs and a telematics control unit (TCU) for producing at least one location signal of a location of the associated motor vehicle and at least one event signal of an event related to the associated motor vehicle, wherein the location signal and the event signal correspond to high speed vehicle telemetry data (HSVT data) based on the inputs from the sensors. The system further includes a computer in communication with a display device and the TCU. The computer includes a processor and a computer readable medium including instructions such that the processor is programmed to identify a congestion and determine whether the congestion is a recurring or non-recurring congestion.

Description

Technical Field

[0001] This disclosure relates to road network congestion, and more specifically, to systems and methods for using real-time high-speed vehicle telemetry data (HSVT data) collected from motor vehicles at a city scale to identify the root causes of road congestion in order to alleviate congestion. Background Technology

[0002] Transportation agencies often lack a comprehensive understanding of the root causes of road congestion and its associated impacts on transportation systems. Furthermore, they may lack the knowledge to categorize active road congestion types at different locations (such as recurring and infrequent congestion). Municipalities currently typically engage civil engineering consultants for annual reviews and analyses to assess and determine basic road infrastructure failures and capacity limitations. These consultants may use partially deployed static sensors along roads for these annual reviews. However, these static sensors suffer from limited data collection coverage (spatial and temporal), time-consuming and / or expensive deployment, and often lead to erroneous recommendations due to a lack of up-to-date data.

[0003] Therefore, while existing systems for alleviating road congestion have achieved their intended purpose, new and improved systems are needed to address these issues. Summary of the Invention

[0004] According to several aspects of this disclosure, a system is provided for identifying, classifying, mitigating, and finding the root causes of road network congestion. The system includes multiple motor vehicles located at multiple associated locations within a road network. Each motor vehicle has one or more sensors for generating inputs. Each motor vehicle also has a telematics control unit (TCU) for generating at least one location signal of the associated motor vehicle's location and at least one event signal of an event related to the associated motor vehicle, the at least one location signal and the at least one event signal corresponding to high-speed vehicle telemetry data (HSVT data) based on inputs from at least one sensor. The system also includes a display device and a computer communicating with the display device and the TCU of the associated motor vehicle. The computer includes one or more processors coupled to and receiving HSVT data from the TCU of the associated motor vehicle. The computer also includes a non-transitory computer-readable storage medium (CRM) including instructions that cause the processor to be programmed to identify road network congestion at a current time slot based on the HSVT data. The processor is also programmed to determine whether the road network congestion is recurring or non-recurring based on HSVT data over a period of time. The processor is also programmed to determine the source of recurring congestion and the cause of infrequent congestion. The processor is also programmed to generate notification signals associated with recurring congestion, the source and location of recurring congestion, infrequent congestion, and / or the cause and location of infrequent congestion, such that, in response to a display device receiving a notification signal from the processor, the display device displays one of the associated recurring congestion, the source and location of recurring congestion, infrequent congestion, and the cause and location of infrequent congestion.

[0005] In one aspect, the sensor is at least one of a GPS unit communicating with the TCU, a thermocouple, a humidity sensor, a brake sensor, an airbag sensor, an ADAS module, a sensing sensor suite, and a motion sensor.

[0006] In another aspect, the processor is programmed to use a causal reasoning model to determine the causal relationship between causal factors in the first position in the first time slot and infrequent congestion in the second position in the second time slot.

[0007] In another aspect, the processor is also programmed to use a causal reasoning model based on the following:

[0008]

[0009] in This indicates that at time slot t, at position p i Non-recurring congestion in the middle; among them Indicates in Located at position p iHistorical congestion in the middle; of which m k Represents element The weighted vector function at time slot tk; where Y represents the infrequent congestion vector in historical data; where k represents the k-th historical time slot before the current time slot t; where Indicates in Located at position p j Historical non-recurring congestion in the region; where n k It is associated with at least one causal factor; where X represents a causal factor that favors the formation of infrequent congestion event Y; where The autocorrelation function representing the relationship between the vector of infrequent congestion events Y and the vector of causal factors X is typically assumed to follow a normal (Gaussian) distribution. ; where O k Represents the elements of the autocorrelation function The weighting function; where k represents the k-th historical time slot before the current time slot t; and where This represents random Gaussian noise.

[0010] On the other hand, the processor is also programmed to determine, based on statistical equations, the causal factors that cause infrequent congestion in the first position in the first time slot to occur in the second position in the second time slot:

[0011]

[0012] in Let X represent the causal relationship determination function, which is used to determine whether a non-recurring congestion event Y is caused by the attributable event X; where X is the causal relationship determination function. This represents the functional operation between the weighted vector n and the weighted vector o; where n represents the weighted vector of the attribution event vector; and o represents the weighted vector of the autocorrelation function between the attribution event vector and infrequent congestion events. The function operation represents the weighted vector m; where m represents the weighted vector of infrequent congestion events.

[0013] In another aspect, the processor is also programmed to determine the sources of recurring congestion by: using an origin-destination matrix derived from historical HSVT data to determine a first route capacity estimate for the road link; using direct measurements from the HSVT dataset collected, for example, in real time or over a predetermined time period (e.g., a whole day), to determine a second route capacity estimate for the road link; comparing the first and second route capacity estimates; and calibrating the processor based on the comparison between the first and second route capacity estimates.

[0014] In another aspect, the processor is programmed to create and use the OD matrix by: determining the spectral TS data extrapolation of historical HSVT data; determining the OD matrix based on the spectral TS data extrapolation; determining route assignments based on the OD matrix; and determining a first route capacity estimate based on the route assignments and the OD matrix.

[0015] In another aspect, the processor is also programmed to generate a partial origin-destination matrix (OD matrix) and partial trajectory data based on HSVT data. The processor is further programmed to determine the OD matrix by: determining OD matrix estimation based on partial trajectory data and HSVT data; determining dynamic route allocation and road capacity estimation based on the OD matrix estimation; and determining static route allocation and road capacity estimation based on partial trajectory data and HSVT data.

[0016] In another aspect, the processor is also programmed to extrapolate the spectral TS data based on the spectral decomposition:

[0017]

[0018] In another aspect, the processor is also programmed to extrapolate the spectral TS data based on the objective function formula:

[0019]

[0020] In another aspect, the processor is also programmed to use an iterative method to reduce the mean square error (MSE) between the extrapolation of the spectral TS data and the underlying facts.

[0021] According to several aspects of this disclosure, a computer is provided for identifying, classifying, mitigating, and finding the root causes of road network congestion. The system includes multiple motor vehicles located at multiple locations within a road network. Each motor vehicle includes one or more sensors for generating inputs. Each motor vehicle also includes a telematics control unit (TCU) that generates one or more location signals of the associated motor vehicle's location and one or more event signals of events related to the associated motor vehicle, wherein the location signals and event signals correspond to high-speed vehicle telemetry data (HSVT data) based on inputs from the sensors. The computer includes one or more processors coupled to the TCU of the associated motor vehicle, and the processor receives the HSVT data from the TCU. The computer also includes a non-transitory computer-readable storage medium (CRM) including instructions that cause the processor to be programmed to identify road network congestion at a current time slot based on the HSVT data. The processor is also programmed to determine whether the road network congestion is recurring or non-recurring based on the HSVT data over a period of time. The processor is further programmed to determine the sources of recurring congestion and the causes of non-recurring congestion. The processor is also programmed to generate notification signals associated with recurring congestion, sources of recurring congestion, non-recurring congestion, and / or causes of non-recurring congestion, such that in response to the display device receiving the notification signals from the processor, the display device displays one of the associated recurring congestion, sources of recurring congestion, non-recurring congestion, and causes of non-recurring congestion.

[0022] In one aspect, the processor is also programmed to use a causal inference model to determine the causal relationship between causal factors in the first position in the first time slot and infrequent congestion in the second position in the second time slot.

[0023] In another aspect, the processor is also programmed to use a causal reasoning model based on the following:

[0024]

[0025] in This indicates that at time slot t, at position p i Non-recurring congestion in the middle; among them Indicates in Located at position p i Historical congestion in the middle; of which m k Represents element The weighted vector function at time slot tk; where Y represents the infrequent congestion vector in historical data; where k represents the k-th historical time slot before the current time slot t; where Indicates in Located at position p j Historical non-recurring congestion in the region; where n kIt is associated with at least one causal factor; where X represents a causal factor that favors the formation of infrequent congestion event Y; where The autocorrelation function representing the relationship between the vector of infrequent congestion events Y and the vector of causal factors X is typically assumed to follow a normal (Gaussian) distribution. ; where O k Represents the elements of the autocorrelation function The weighting function; where k represents the k-th historical time slot before the current time slot t; and where This represents random Gaussian noise.

[0026] On the other hand, the processor is also programmed to determine, based on statistical equations, the causal factors that cause infrequent congestion in the first position in the first time slot to occur in the second position in the second time slot:

[0027]

[0028] in Let X represent the causal relationship determination function, which is used to determine whether a non-recurring congestion event Y is caused by the attributable event X; where X is the causal relationship determination function. This represents the functional operation between the weighted vector n and the weighted vector o; where n represents the weighted vector of the attribution event vector; and o represents the weighted vector of the autocorrelation function between the attribution event vector and infrequent congestion events. The function operation represents the weighted vector m; where m represents the weighted vector of infrequent congestion events.

[0029] In another aspect, the processor is also programmed to determine the sources of recurring congestion by: using an origin-destination matrix (OD matrix) derived from historical HSVT data to determine a first route capacity estimate for the road link; using direct measurements derived from the HSVT dataset to determine a second route capacity estimate for the road link; comparing the first and second route capacity estimates; and calibrating the processor based on the comparison between the first and second route capacity estimates.

[0030] According to several aspects of this disclosure, an operating system provides a method for identifying, classifying, mitigating road network congestion, and identifying the root causes of road network congestion. The system includes multiple motor vehicles located at multiple locations within a road network. Each motor vehicle has one or more sensors for generating inputs. Each motor vehicle also includes a telematics control unit (TCU) that generates one or more location signals of the associated motor vehicle's location and one or more event signals of events related to the associated motor vehicle, wherein the location signals and event signals correspond to high-speed vehicle telemetry data (HSVT data) based on inputs from the sensors. The system also includes a computer having one or more processors coupled to the TCU of the associated motor vehicle. The computer also has a non-transitory computer-readable storage medium (CRM) including instructions. The method includes using the processor to identify road network congestion at a current time slot based on the HSVT data. The method also includes using the processor to determine, based on HSVT data over a period of time, whether the road network congestion is recurring and / or non-recurring. The method further includes using the processor to determine the source of recurring congestion or the cause of non-recurring congestion. The method also includes using a processor to generate a notification signal associated with recurring congestion, a source of recurring congestion, infrequent congestion, or a cause of infrequent congestion, such that in response to a display device receiving the notification signal from the processor, the display device displays one of the associated recurring congestion, a source of recurring congestion, infrequent congestion, or a cause of infrequent congestion.

[0031] In one aspect, the method also includes using a processor to employ a causal inference model to determine the causal relationship between causal factors in a first location in a first time slot and infrequent congestion in a second location in a second time slot.

[0032] In another aspect, the method also includes using a processor to apply a causal reasoning model based on the following:

[0033]

[0034] in This indicates that at time slot t, at position p i Non-recurring congestion in the middle; among them Indicates in Located at position p i Historical congestion in the middle; of which m k Represents element The weighted vector function at time slot tk; where Y represents the infrequent congestion vector in historical data; where k represents the k-th historical time slot before the current time slot t; where Indicates in Located at position p j Historical non-recurring congestion in the region; where n kIt is associated with at least one causal factor; where X represents a causal factor that favors the formation of infrequent congestion event Y; where The autocorrelation function representing the relationship between the vector of infrequent congestion events Y and the vector of causal factors X is typically assumed to follow a normal (Gaussian) distribution. ; where O k Represents the elements of the autocorrelation function The weighting function; where k represents the k-th historical time slot before the current time slot t; and where This represents random Gaussian noise.

[0035]

[0036] in Let X represent the causal relationship determination function, which is used to determine whether a non-recurring congestion event Y is caused by the attributable event X; where X is the causal relationship determination function. This represents the functional operation between the weighted vector n and the weighted vector o; where n represents the weighted vector of the attribution event vector; and o represents the weighted vector of the autocorrelation function between the attribution event vector and infrequent congestion events. The function operation represents the weighted vector m; where m represents the weighted vector of infrequent congestion events.

[0037] In another aspect, the method further includes a processor identifying sources of recurring congestion. The method also includes a processor using an origin-destination matrix (OD matrix) derived from historical HSVT data to determine a first route capacity estimate for road links. The method further includes a processor using direct measurements from the HSVT dataset to determine a second route capacity estimate for road links. The method also includes a processor comparing the first and second route capacity estimates. The method further includes calibrating the processor based on the comparison between the first and second route capacity estimates.

[0038] Other areas of application will become apparent from the description provided herein. It should be understood that the descriptions and specific examples are intended for illustrative purposes only and are not intended to limit the scope of this disclosure.

[0039] The present invention also includes the following technical solutions.

[0040] Technical Solution 1. A system for identifying, classifying, mitigating, and finding the root causes of road network congestion, the system comprising:

[0041] Multiple motor vehicles located at multiple associated locations in a road network, each of the motor vehicles having at least one sensor for generating input, and each of the motor vehicles also having a telematics control unit (TCU) for generating at least one position signal of the associated motor vehicle's location and at least one event signal of an event related to the associated motor vehicle, the at least one position signal and the at least one event signal corresponding to high-speed vehicle telemetry data (HSVT data) based on input from at least one sensor.

[0042] Display device; and

[0043] A computer that communicates with a display device and a TCU of an associated motor vehicle, the computer comprising:

[0044] At least one processor, said processor being coupled to and receiving HSVT data from the TCU of the associated motor vehicle; and

[0045] A non-transitory computer-readable storage medium, including instructions, such that at least one processor is programmed to:

[0046] Identify road network congestion at the current time slot based on HSVT data;

[0047] Based on HSVT data over a period of time, road network congestion is determined to be at least one of recurring congestion and non-recurring congestion;

[0048] Identify at least one of the sources of recurring congestion and the causes of infrequent congestion; and

[0049] Generate a notification signal associated with at least one of recurring congestion, the source and location of recurring congestion, infrequent congestion, and the cause and location of infrequent congestion, such that in response to a display device receiving the notification signal from at least one processor, the display device displays, based on HSVT data, one of the associated recurring congestion, the source and location of recurring congestion, infrequent congestion, and the cause and location of infrequent congestion.

[0050] Technical Solution 2. The system according to Technical Solution 1, wherein the at least one sensor includes at least one of a GPS unit communicating with the TCU, a thermocouple, a humidity sensor, a brake sensor, an airbag sensor, an ADAS module, a sensing sensor kit, and a motion sensor.

[0051] Technical Solution 3. The system according to Technical Solution 2, wherein the at least one processor is programmed to use a causal inference model to determine the causal relationship between causal factors in a first location in a first time slot and infrequent congestion in a second location in a second time slot.

[0052] Technical Solution 4. The system according to Technical Solution 3, wherein the at least one processor is programmed to use the causal reasoning model according to the following:

[0053]

[0054] in This indicates that at time slot t, at position p i Non-recurring congestion in the middle; among them Indicates in Located at position p i Historical congestion in the middle; of which m k Represents element The weighted vector function at time slot tk; where Y represents the infrequent congestion vector in historical data; where k represents the k-th historical time slot before the current time slot t; where Indicates in Located at position p j Historical non-recurring congestion in the region; where n k It is associated with at least one causal factor; where X represents a causal factor that favors the formation of infrequent congestion event Y; where The autocorrelation function representing the relationship between the vector of infrequent congestion events Y and the vector of causal factors X is typically assumed to follow a normal (Gaussian) distribution. ; where O k Represents the elements of the autocorrelation function The weighting function; where k represents the k-th historical time slot before the current time slot t; and where This represents random Gaussian noise.

[0055] Technical Solution 5. The system according to Technical Solution 3, wherein the at least one processor is further programmed to determine, based on statistical equations, causal factors at the first position in the first time slot leading to infrequent congestion at the second position in the second time slot:

[0056]

[0057] in Let X represent the causal relationship determination function, which is used to determine whether a non-recurring congestion event Y is caused by the attributable event X; where X is the causal relationship determination function. This represents the functional operation between the weighted vector n and the weighted vector o; where n represents the weighted vector of the attribution event vector; and o represents the weighted vector of the autocorrelation function between the attribution event vector and infrequent congestion events. The function operation represents the weighted vector m; where m represents the weighted vector of infrequent congestion events.

[0058] Technical Solution 6. The system according to Technical Solution 3, wherein the at least one processor is further programmed to determine the source of recurring congestion in the following manner:

[0059] The origin-destination matrix (OD matrix) derived from historical HSVT data is used to determine the first route capacity estimate for road links;

[0060] The second route capacity estimate for road links is determined using direct measurements from the HSVT dataset;

[0061] Compare the capacity estimates of the first and second routes; and

[0062] The at least one processor is calibrated based on a comparison between the first and second route capacity estimates.

[0063] Technical Solution 7. The system according to Technical Solution 6, wherein the at least one processor is programmed to create and use the OD matrix in the following manner:

[0064] Determine the spectral spatiotemporal data extrapolation of historical HSVT data.

[0065] The OD matrix is ​​determined based on extrapolation of spectral TS data and available partial OD matrix data;

[0066] Route assignment is determined based on the OD matrix; and

[0067] The capacity estimate of the first route is determined based on route assignment and the OD matrix.

[0068] Technical Solution 8. The system according to Technical Solution 7, wherein the at least one processor is programmed to generate a partial origin-end point matrix and partial trajectory data based on HSVT data, and the at least one processor is further programmed to determine the OD matrix in the following manner:

[0069] The origin-end point matrix estimation (OD Matrix Estimation) is determined based on partial trajectory data and HSVT data.

[0070] Dynamic route allocation and road capacity estimation are determined based on OD matrix estimation; and

[0071] Static route assignments and road capacity estimates are determined based on partial trajectory data and HSVT data.

[0072] Technical Solution 9. The system according to Technical Solution 8, wherein the at least one processor is further programmed to determine the extrapolation of spectral TS data based on spectral decomposition:

[0073] .

[0074] Technical Solution 10. The system according to Technical Solution 9, wherein the at least one processor is further programmed to determine the extrapolation of the spectrum TS data based on the objective function formula:

[0075] .

[0076] Technical Solution 11. The system according to Technical Solution 10, wherein the at least one processor is further programmed to use an iterative method to reduce the mean square error (MSE) between the extrapolation of the spectral TS data and the underlying facts.

[0077] Technical Solution 12. A computer for identifying, classifying, mitigating road network congestion, and finding the root causes of road network congestion, the system having multiple motor vehicles located at multiple locations in a road network, each of the motor vehicles having at least one sensor for generating input, and each of the motor vehicles further having a telematics control unit (TCU) for generating at least one location signal of the associated motor vehicle's location and at least one event signal of an event related to the associated motor vehicle, the at least one location signal and the at least one event signal corresponding to high-speed vehicle telemetry data (HSVT data) based on input from the at least one sensor, the computer comprising:

[0078] At least one processor, said processor being coupled to and receiving HSVT data from the TCU of the associated motor vehicle; and

[0079] A non-transitory computer-readable storage medium, including instructions, such that at least one processor is programmed to:

[0080] Identify road network congestion at the current time slot based on HSVT data;

[0081] Based on HSVT data over a period of time, road network congestion is determined to be at least one of recurring congestion and non-recurring congestion;

[0082] Identify at least one of the sources of recurring congestion and the causes of infrequent congestion; and

[0083] Generate a notification signal associated with at least one of recurring congestion, the source and location of recurring congestion, infrequent congestion, and the cause and location of infrequent congestion, such that in response to a display device receiving the notification signal from at least one processor, the display device displays, based on HSVT data, one of the associated recurring congestion, the source and location of recurring congestion, infrequent congestion, and the cause and location of infrequent congestion.

[0084] Technical Solution 13. The computer according to Technical Solution 12, wherein the at least one processor is programmed to use a causal reasoning model to determine the causal relationship between causal factors in a first location in a first time slot and infrequent congestion in a second location in a second time slot.

[0085] Technical Solution 14. The computer according to Technical Solution 13, wherein the at least one processor is programmed to use the causal reasoning model according to the following:

[0086]

[0087] in This indicates that at time slot t, at position p i Non-recurring congestion in the middle; among them Indicates in Located at position p i Historical congestion in the middle; of which m k Represents element The weighted vector function at time slot tk; where Y represents the infrequent congestion vector in historical data; where k represents the k-th historical time slot before the current time slot t; where Indicates in Located at position p j Historical non-recurring congestion in the region; where n k It is associated with at least one causal factor; where X represents a causal factor that favors the formation of infrequent congestion event Y; where The autocorrelation function representing the relationship between the vector of infrequent congestion events Y and the vector of causal factors X is typically assumed to follow a normal (Gaussian) distribution. ; where O k Represents the elements of the autocorrelation function The weighting function; where k represents the k-th historical time slot before the current time slot t; and where This represents random Gaussian noise.

[0088] Technical Solution 15. The computer according to Technical Solution 13, wherein the at least one processor is further programmed to determine, based on a statistical equation, causal factors in the first position at the first time slot leading to infrequent congestion in the second position at the second time slot:

[0089]

[0090] in Let X represent the causal relationship determination function, which is used to determine whether a non-recurring congestion event Y is caused by the attributable event X; where X is the causal relationship determination function. This represents the functional operation between the weighted vector n and the weighted vector o; where n represents the weighted vector of the attribution event vector; and o represents the weighted vector of the autocorrelation function between the attribution event vector and infrequent congestion events. The function operation represents the weighted vector m; where m represents the weighted vector of infrequent congestion events.

[0091] Technical Solution 16. The computer according to Technical Solution 13, wherein the at least one processor is further programmed to determine the source of recurring congestion in the following manner:

[0092] The origin-destination matrix (OD matrix) derived from historical HSVT data is used to determine the first route capacity estimate for road links;

[0093] The second route capacity estimate for road links is determined using direct measurements from the HSVT dataset;

[0094] Compare the capacity estimates of the first and second routes; and

[0095] The at least one processor is calibrated based on a comparison between the first and second route capacity estimates.

[0096] Technical Solution 17. A method for operating a system for identifying, classifying, mitigating, and finding the root causes of road network congestion, the system having multiple motor vehicles located at multiple locations in a road network, each of the motor vehicles having at least one sensor for generating input and a telematics control unit (TCU) for generating at least one location signal of the associated motor vehicle's location and at least one event signal of an event related to the associated motor vehicle, the at least one location signal and the at least one event signal corresponding to high-speed vehicle telemetry data (HSVT data) based on the input from the at least one sensor, the system further including a computer having at least one processor coupled to the TCU of the associated motor vehicle and a non-transitory computer-readable storage medium including instructions, the method comprising:

[0097] The at least one processor is used to identify road network congestion in the current time slot based on HSVT data;

[0098] The at least one processor is used to determine, based on HSVT data over a period of time, whether road network congestion is at least one of recurring congestion and infrequent congestion;

[0099] The at least one processor is used to determine at least one of the sources of recurring congestion and the causes of infrequent congestion; and

[0100] The at least one processor generates a notification signal associated with at least one of recurring congestion, the source and location of recurring congestion, non-recurring congestion, and the cause and location of non-recurring congestion, such that in response to a display device receiving the notification signal from the at least one processor, the display device displays, based on HSVT data, one of the associated recurring congestion, the source and location of recurring congestion, non-recurring congestion, and the cause and location of non-recurring congestion.

[0101] Technical Solution 18. The method according to Technical Solution 17 further includes:

[0102] The at least one processor uses a causal inference model to determine the causal relationship between causal factors in the first position in the first time slot and infrequent congestion in the second position in the second time slot.

[0103] Technical Solution 19. The method according to Technical Solution 18 further includes:

[0104] The causal inference model is used by at least one processor according to the following:

[0105]

[0106] in This indicates that at time slot t, at position p i Non-recurring congestion in the middle; among them Indicates in Located at position p i Historical congestion in the middle; of which m k Represents element The weighted vector function at time slot tk; where Y represents the infrequent congestion vector in historical data; where k represents the k-th historical time slot before the current time slot t; where Indicates in Located at position p j Historical non-recurring congestion in the region; where n k It is associated with at least one causal factor; where X represents a causal factor that favors the formation of infrequent congestion event Y; where The autocorrelation function representing the relationship between the vector of infrequent congestion events Y and the vector of causal factors X is typically assumed to follow a normal (Gaussian) distribution. ; where O k Represents the elements of the autocorrelation function The weighting function; where k represents the k-th historical time slot before the current time slot t; and where Represents random Gaussian noise; and

[0107] The at least one processor determines, according to the following statistical equation, the causal factors that cause infrequent congestion in the second position in the first time slot, at the first location, lead to the infrequent congestion in the second location in the second time slot:

[0108]

[0109] in Let X represent the causal relationship determination function, which is used to determine whether a non-recurring congestion event Y is caused by the attributable event X; where X is the causal relationship determination function. This represents the functional operation between the weighted vector n and the weighted vector o; where n represents the weighted vector of the attribution event vector; and o represents the weighted vector of the autocorrelation function between the attribution event vector and infrequent congestion events. The function operation represents the weighted vector m; where m represents the weighted vector of infrequent congestion events.

[0110] Technical Solution 20. The method according to Technical Solution 19 further includes using the at least one processor to determine the source of recurring congestion in the following manner:

[0111] The at least one processor uses the origin-destination matrix (OD matrix) derived from historical HSVT data to determine the first route capacity estimate for the road link;

[0112] The at least one processor uses direct measurements from the HSVT dataset to determine a second route capacity estimate for the road link;

[0113] The first and second route capacity estimates are compared using the at least one processor; and

[0114] The at least one processor is calibrated using the at least one processor based on a comparison between the first and second route capacity estimates. Attached Figure Description

[0115] The accompanying drawings described herein are for illustrative purposes only and are not intended to limit the scope of this disclosure in any way.

[0116] Figure 1It is a schematic diagram of a road network with multiple road edges and multiple road intersections, wherein an example of the system includes multiple motor vehicles traveling along associated road edges and through associated road intersections.

[0117] Figure 2 yes Figure 1 A schematic diagram of the system shows that it includes a motor vehicle with an associated telematics control unit (TCU) and a computer communicating with the TCU for identifying, classifying, mitigating road network congestion, and finding the root causes of road network congestion.

[0118] Figure 3 It is used for operation Figure 1 A flowchart of a non-restrictive example of the system's approach. Detailed Implementation

[0119] The following description is merely exemplary and is not intended to limit this disclosure, its application, or its uses. Although the accompanying drawings illustrate examples, they are not necessarily drawn to scale, and certain features may be enlarged to better illustrate and explain specific aspects of the illustrative examples. Any one or more of these aspects may be used alone or in combination with each other. Furthermore, the exemplary descriptions herein are not intended to be exhaustive or otherwise limit or restrict the precise forms and configurations shown in the drawings and disclosed in the following detailed description. The exemplary illustrations are described in detail below with reference to the accompanying drawings.

[0120] refer to Figure 1A non-limiting example of System 100 uses real-time high-speed vehicle telemetry data (HSVT data) collected at a city scale from multiple motor vehicles 102 to proactively identify congestion locations, track congestion spread, and predict congestion evolution, enabling transportation agencies to manage traffic and / or allowing motor vehicles to adopt alternative routes depending on congestion levels. To this end, as detailed below, System 100 uses causal relationship analysis and random forest analysis to identify the root causes of road congestion and determine whether two spatiotemporally random events are causally related to other events that contribute to congestion formation. Specifically, System 100 identifies non-recurring causes of congestion associated with aperiodic or random factors. Non-limiting examples of these random factors may include traffic accidents, construction zones, hazardous weather, special events (concerts, sporting events, etc.), and shockwaves. However, it is conceivable that non-recurring causes can be associated with any other suitable random factor. System 100 also identifies recurring causes of congestion associated with periodic intervals or predictable times (such as peak hours or evening hours) of basic road infrastructure failures. Non-limiting examples of these basic road infrastructure failures can include highway merging, insufficient road capacity, inadequate traffic control, or traffic control malfunctions. However, it is conceivable that recurring causes could be associated with any other suitable basic road infrastructure failure. System 100 utilizes a data-driven approach to determine causal relationships to predict and mitigate congestion in a more accurate, timely, and cost-effective manner than known static sensors currently used by civil engineering consultants and / or transportation agencies.

[0121] In this non-limiting example, system 100 includes motor vehicles 102 located at multiple locations within road network 103 for traveling along associated road edges 104 and through associated road intersections 106. As described in detail below, system 100 can determine that congested intersections and congestion edges connected to each other form a first sub-figure 108 of road congestion, and system 100 can also determine that two congestion edges connected to each other form a second sub-figure 110 of road congestion.

[0122] refer to Figure 2Each motor vehicle 102 includes a telematics control unit 112 (TCU) for generating one or more position signals of the associated motor vehicle 102's location and one or more event signals of events related to the associated motor vehicle 102 using telematics services. The position signals and event signals correspond to high-speed vehicle telemetry data (HSVT data) based on inputs from one or more sensors 113 described below. In this non-limiting example, the TCU 112 is a microcontroller (a complete computer on a single electronic chip), a microprocessor, or a field-programmable gate array (FPGA). The TCU 112 wirelessly connects the associated motor vehicle 102 to cloud services or other vehicles via a mobile network using V2X or P2P standards. The TCU 112 connects and communicates with various subsystems and collects telemetry data via the data and control bus (CAN) in the motor vehicle 102. This data includes elements such as location, speed, engine data, and connection quality. It can also provide in-vehicle connectivity via Wi-Fi and BlueTooth and enable e-Call functionality in relevant markets. The TCU 112 communicates with suitable sensors 113 of the vehicle 102 to collect telemetry data. Non-limiting examples of these components may include a GPS unit 114 that tracks the latitude and longitude values ​​of the vehicle 102, enabling the TCU 112 to generate a position signal based on these values. Other non-limiting examples of these sensors 113 may include a humidity sensor 116, a brake sensor 118, an airbag sensor 120, an ADAS module 122, a perception sensor suite 123, a motion sensor 124, and one or more mobile communication units 126, along with an external interface for mobile communication (GSM, GPRS, Wi-Fi, WiMax, LTE, or 5G) that provides the tracked values ​​to a central computer 128 or database server as described below. The vehicle 102 also includes a amount of memory 130 for storing GPS values ​​or intelligently storing information about vehicle sensor data in the absence of movement.

[0123] As described in detail below, system 100 may include a local road model with peer-to-peer (P2P) or edge computing, utilizing the TCU 112 of a motor vehicle communicating with the TCU 112 of other motor vehicles 102. Edge computing is a distributed computing paradigm that brings computation and data storage closer to the data source to improve response time and save bandwidth. System 100 may also include cloud computing with a global road model utilizing a remote computer 128 or server. Computer 128 includes one or more processors 134 and a non-transient computer-readable storage medium 132 (“CRM”) including instructions, such that processor 134 is programmed to receive HSVT data from sensors 113 of the associated motor vehicle 102.

[0124] Processor 134 is also programmed to identify road network congestion at the current time slot based on HSVT data. More specifically, processor 134 is further programmed to determine, based on analysis of HSVT data over time (e.g., days, weeks, months, etc.), whether the road network congestion is at least one of recurring and infrequent congestion, to distinguish whether the observed congestion is part of a recurring or recurring pattern (by location, time of day, day of week, etc.). Processor 134 is programmed to use a causal inference model to determine the causal relationship between causal factors at a first location in a first time slot and infrequent congestion at a second location in a second time slot.

[0125] Equation 1

[0126] Non-limiting examples of causal factors may include, but are not limited to, weather and related road conditions, traffic accidents, construction zones, shock waves, police activity, special sporting / public / concert events, or other temporary road / lane closures; This indicates that at time slot t, at position p i Non-recurring congestion in the middle; among them Indicates in Located at position p i Historical congestion in the middle; of which m k Represents element The weighted vector function at time slot tk; where Y represents the infrequent congestion vector in historical data; where k represents the k-th historical time slot before the current time slot t; where Indicates in Located at position p j Historical non-recurring congestion in the region; where n k It is associated with at least one causal factor; where X represents a causal factor that favors the formation of infrequent congestion event Y; where The autocorrelation function representing the relationship between the vector of infrequent congestion events Y and the vector of causal factors X is typically assumed to follow a normal (Gaussian) distribution. ; where O k Represents the elements of the autocorrelation function The weighting function; where k represents the k-th historical time slot before the current time slot t; and where The noise represents random Gaussian noise. Those skilled in the art will understand that Equation 1 is a non-limiting example illustrating how infrequent congestion events can be modeled as a weighted sum of several influencing factors, including: historical infrequent congestion events, attributable events that may have caused the infrequent congestion events, the autocorrelation between the infrequent congestion events and the attributable events that may have caused the infrequent congestion events, and random Gaussian noise. Other non-limiting examples or equivalent schemes of Equation 1 or other modeling methods can be conceived for modeling infrequent congestion events.

[0127] Processor 134 is also programmed to determine the source and location of recurring congestion and the cause and location of infrequent congestion. Processor 134 is further programmed to determine, based on statistical equations, the causal factors at the first location in the first time slot that lead to infrequent congestion at the second location in the second time slot.

[0128] Equation 2

[0129] in This represents a causal relationship determination function, used to determine whether a non-recurring congestion event Y is caused by an attributable event X (such as weather conditions, construction zones, police incidents, etc.); where Let represent the functional operation between the weighted vector n and the weighted vector o; where n represents the weighted vector of the attribution event vector (as described in Equation 1); and o represents the weighted vector of the autocorrelation function between the attribution event vector and the infrequent congestion event (as described in Equation 1). The function operation represents the weighted vector m; where m represents the weighted vector of infrequent congestion events.

[0130] Those skilled in the art will understand that Equation 2 is a non-limiting example used to illustrate whether an infrequent congestion event is caused by a given attributable event, which can be determined by a weighted factor associated with several influencing factors, including: historical infrequent congestion events, attributable events that may have caused the infrequent congestion event, autocorrelation between the infrequent congestion event and the attributable events that may have caused the infrequent congestion event, and random Gaussian noise. Other equivalent schemes or variations of Equation 2 or other modeling methods are conceivable for determining whether an infrequent congestion event Y is caused by an attributable event X.

[0131] The processor 134 is also programmed to determine the source of recurring congestion by: using an origin-destination matrix derived from historical HSVT data to determine a first route capacity estimate for the road link; using direct measurements from the HSVT dataset (e.g., historical traffic volume and speed) to determine a second route capacity estimate for the road link; comparing the first and second route capacity estimates; and calibrating the processor based on the comparison between the first and second route capacity estimates.

[0132] The processor 134 is also programmed to create and use the OD matrix in the following ways: determining the spectral TS data extrapolation of historical HSVT data; determining the OD matrix based on the spectral TS data extrapolation, such as a city-wide OD matrix or a regional OD matrix; determining route assignments based on the OD matrix; and determining an initial route capacity estimate based on the route assignments and the OD matrix.

[0133] Processor 134 is also programmed to generate a partial origin-destination matrix and partial trajectory data based on HSVT data, and is further programmed to determine the OD matrix by: determining an OD matrix estimation based on the partial trajectory data and HSVT data; determining dynamic route assignment and road capacity estimation based on the OD matrix estimation; and determining static route assignment and road capacity estimation based on the partial trajectory data and HSVT data. Processor 134 is also programmed to determine spectral TS data extrapolation based on spectral decompositions such as Principal Component Analysis (PCA) or Singular Vector Decomposition (SVD).

[0134] Equation 3

[0135] Equation 4

[0136] Processor 134 is also programmed to extrapolate the spectrum TS data based on the objective function formula:

[0137] Equation 5

[0138] Equation 6

[0139] Those skilled in the art will understand that the above equations are non-limiting examples illustrating a spectral decomposition method that can use partial OD matrix data to aid in data extrapolation. However, equivalent schemes or variations of these equations, or other modeling methods using spectral decomposition and reconstruction, are conceivable.

[0140] Processor 134 is also programmed to use an iterative approach to reduce the mean square error (MSE) between the extrapolation of the spectral TS data and the underlying facts.

[0141] The processor 134 is also programmed to generate notification signals associated with at least one location of recurring congestion, the source of recurring congestion, the location of infrequent congestion, and the cause of infrequent congestion, such that in response to a display device receiving a notification signal from the processor, the display device displays the location of recurring congestion, the source of recurring congestion, the location of infrequent congestion, and the cause of infrequent congestion based on HSVT data. As disclosed in U.S. Patent Application No. 17 / 575,017 (the entire contents of which are incorporated herein by reference), congestion is a system state identified by comparing observed speeds and travel times on a road link with a reference non-congestion state that occurs normally when observed traffic volume or derived flow reaches road link capacity, derived using historical HSVT data.

[0142] In one non-limiting example, the display device could be a screen in a motor vehicle used to notify vehicle occupants of road congestion, allowing them to drive the vehicle along an alternative route free from congestion. In another non-limiting example, the display device could be a screen in an autonomous vehicle used to notify vehicle occupants of road congestion and instruct the autonomous vehicle to travel along an alternative route free from congestion. In yet another non-limiting example, the display device could be a monitor on a desktop computer used by a municipality to analyze road congestion and modify traffic control infrastructure to better manage urban traffic, send first responders to incidents, indicate the location of snowplows clearing snow-covered road networks, or take any other appropriate action.

[0143] Now for reference Figure 3 Provides operation Figure 1 This is a non-limiting example of method 200 of system 100. Method 200 begins at block 202 with the use of processor 134 to identify road network congestion at the current time slot based on HSVT data.

[0144] At box 204, method 200 further includes using processor 134 to determine, based on HSVT data, whether road network congestion is at least one of recurring and infrequent congestion conditions.

[0145] At box 206, method 200 further includes using processor 134 to determine at least one of the sources of recurring congestion and the causes of anomalous congestion. More specifically, method 200 may include using processor 134 to employ a causal inference model to determine a causal relationship between causal factors in a first location in a first time slot and anomalous congestion in a second location in a second time slot. Method 200 may include using processor 134 to employ a causal inference model according to Equation 1 above.

[0146] Method 200 also includes using processor 134 to determine, according to Equation 2 above, the causal factors in the first position in the first time slot that cause non-recurring congestion in the second position in the second time slot.

[0147] Method 200 also includes using processor 134 to determine the source of recurring congestion by using processor to determine a first route capacity estimate of the road link using the origin-destination matrix (OD matrix); using processor to determine a second route capacity estimate of the road link using direct measurement results; using processor to compare the first and second route capacity estimates; and using processor to calibrate the processor based on the comparison between the first and second route capacity estimates.

[0148] At block 208, method 200 further includes using processor 134 to generate a notification signal associated with recurring congestion, the source of recurring congestion, non-recurring congestion, and / or the cause of non-recurring congestion. In response to receiving the notification signal from the processor, display device 136 displays one of the associated recurring congestion, the source of recurring congestion, non-recurring congestion, and the cause of non-recurring congestion based on HSVT data.

[0149] Computers and computing devices typically include computer-executable instructions, which can be executed by one or more computing devices, such as those listed above. Computer-executable instructions can be compiled or interpreted from computer programs created using various programming languages ​​and / or techniques, including but not limited to JAVA, C, C++, MATLAB, SIMUEDGE, STATEFLOW, VISUALBASIC, JAVASCRIPT, PERL, HTML, TENSORFLOW, PYTHON, PYTORCH, KERAS, etc., individually or in combination. Some of these applications can be compiled and executed on virtual machines, such as the Java Virtual Machine, the Dalvik Virtual Machine, etc. Typically, a processor (e.g., a microprocessor) receives instructions from memory, computer-readable media, etc., and executes those instructions to perform one or more processes, including one or more processes described herein. Various computer-readable media can be used to store and transfer such instructions and other data. Files in a computing device are typically collections of data stored on computer-readable media, such as storage media, random access memory, etc.

[0150] CRM (also known as processor-readable media) is involved in providing data (e.g., instructions) that can be read by a computer (e.g., by the computer's processor). Such media can take many forms, including but not limited to non-volatile and volatile media. Non-volatile media can include, for example, optical discs or magnetic disks, and other persistent storage. Volatile media can include, for example, dynamic random access memory (DRAM), which typically constitutes main memory. Such instructions can be transmitted via one or more transmission media, including coaxial cables, copper wires, and optical fibers, including wires that include a system bus connected to the processor of the ECU. Common forms of computer-readable media include, for example, floppy disks, retractable disks, hard disks, magnetic tape, any other magnetic media, CD-ROMs, DVDs, any other optical media, punched cards, paper tape, any other physical media with a perforated pattern, RAM, PROM, EPROM, FLASH EEPROM, any other memory chips or cassette tapes, or any other media that a computer can read.

[0151] In some examples, system elements may be implemented as computer-readable instructions (e.g., software) on one or more computing devices, stored on an associated computer-readable medium (e.g., a disk, memory, etc.). A computer program product may include instructions stored on a computer-readable medium for performing the functions described herein.

[0152] Regarding the media, processes, systems, methods, heuristic teaching methods, etc., described herein, it should be understood that although the steps of such processes, etc., have been described as occurring in a certain ordered sequence, such processes can also be practiced by means of steps performed in an order different from that described herein. It should be further understood that some steps can be performed simultaneously, other steps can be added, or some steps described herein can be omitted. In other words, the description of processes herein is provided to illustrate certain embodiments and should in no way be construed as limiting the claims.

[0153] Therefore, it should be understood that the above description is intended to be exemplary and not restrictive. Many embodiments and applications beyond the examples provided will be apparent to those skilled in the art upon reading the above description. The scope of the invention should not be determined by reference to the above description, but rather by reference to the appended claims and the full scope of equivalents claiming rights to those claims. Future developments are anticipated and intended to occur in the art discussed herein, and the disclosed systems and methods will be incorporated into such future embodiments. In summary, it should be understood that the invention is capable of modifications and variations and is limited only by the following claims.

[0154] All terms used in the claims are intended to be given their simple and common meaning as understood by those skilled in the art, unless explicitly indicated otherwise herein. In particular, the use of singular articles such as “a,” “that,” “the,” etc., should be interpreted as referring to one or more of the indicated elements, unless the statement of the claims explicitly limits it to the contrary.

Claims

1. A system for identifying, classifying, mitigating, and finding the root causes of road network congestion, the system comprising: Multiple motor vehicles located at multiple associated locations in a road network, each of the motor vehicles having at least one sensor for generating input, and each of the motor vehicles also having a telematics control unit for generating at least one position signal of the associated motor vehicle's location and at least one event signal of an event related to the associated motor vehicle, the at least one position signal and the at least one event signal corresponding to high-speed vehicle telemetry data based on input from at least one sensor; Display device; and A computer that communicates with a display device and a telematics control unit of an associated motor vehicle, the computer comprising: At least one processor, said processor being coupled to and receiving high-speed vehicle telemetry data from the telemetry control unit of the associated motor vehicle; and A non-transitory computer-readable storage medium, including instructions, such that at least one processor is programmed to: Identify road network congestion in the current time slot based on high-speed vehicle telemetry data; Based on high-speed vehicle telemetry data over a period of time, road network congestion is determined to be at least one of frequent congestion and infrequent congestion; Identify at least one of the sources of recurring congestion and the causes of infrequent congestion; and Generate a notification signal associated with at least one of recurring congestion, the source and location of recurring congestion, non-recurring congestion, and the cause and location of non-recurring congestion, such that in response to a display device receiving the notification signal from at least one processor, the display device displaying a signal associated with one of recurring congestion, the source and location of recurring congestion, non-recurring congestion, and the cause and location of non-recurring congestion based on high-speed vehicle telemetry data; The at least one processor is further programmed to determine the source of recurring congestion in the following manner: The origin-endpoint matrix derived from historical high-speed vehicle telemetry data is used to determine the first route capacity estimate for the road link; The second route capacity estimate for road links is determined using direct measurements from high-speed vehicle telemetry datasets. Compare the capacity estimates of the first and second routes; and The at least one processor is calibrated based on a comparison between the first and second route capacity estimates. The at least one sensor includes at least one of the following: a GPS unit communicating with a remote information processing control unit, a thermocouple, a humidity sensor, a brake sensor, an airbag sensor, an ADAS module, a sensing sensor kit, and a motion sensor. The at least one processor is programmed to use a causal inference model to determine the causal relationship between causal factors in a first location in a first time slot and infrequent congestion in a second location in a second time slot. The at least one processor is programmed to use the causal reasoning model according to the following: in This indicates that at time slot t, at position p i Non-recurring congestion in the middle; among them Indicates in Located at position p i Historical congestion in the middle; of which m k Represents element The weighted vector function at time slot tk; where Y represents the infrequent congestion vector in historical data; where k represents the k-th historical time slot before the current time slot t; where Indicates in Located at position p j Historical non-recurring congestion in the region; where n k It is associated with at least one causal factor; where X represents a causal factor that favors the formation of infrequent congestion event Y; where The autocorrelation function representing the relationship between the vector of infrequent congestion events Y and the vector of causal factors X is Gaussian distributed. ; where O k Represents the elements of the autocorrelation function The weighting function; where k represents the k-th historical time slot before the current time slot t; and where This represents random Gaussian noise.

2. The system according to claim 1, wherein, The at least one processor is also programmed to determine, based on statistical equations, causal factors in the first position within the first time slot that cause infrequent congestion in the second position within the second time slot: in Let X represent the causal relationship determination function, which is used to determine whether a non-recurring congestion event Y is caused by the attributable event X; where X is the causal relationship determination function. This represents the functional operation between the weighted vector n and the weighted vector o; where n represents the weighted vector of the attribution event vector; and o represents the weighted vector of the autocorrelation function between the attribution event vector and infrequent congestion events. The function operation represents the weighted vector m; where m represents the weighted vector of infrequent congestion events.

3. The system according to claim 1, wherein, The at least one processor is programmed to create and use the origin-end point matrix in the following manner: Determine the extrapolation of the spatiotemporal data of historical high-speed vehicle telemetry data; The origin and destination matrix is ​​determined based on extrapolation of spectral spatiotemporal data and available partial origin and destination matrix data; Route allocation is determined based on the origin-destination matrix; and The capacity estimate of the first route is determined based on route allocation and the origin-destination matrix.

4. The system according to claim 3, wherein, The at least one processor is programmed to generate a partial origin-destination matrix and partial trajectory data based on high-speed vehicle telemetry data, and the at least one processor is further programmed to determine the origin-destination matrix in the following manner: The origin-end point matrix is ​​estimated based on partial trajectory data and high-speed vehicle telemetry data; Dynamic route allocation and road capacity estimation are determined based on origin-destination matrix estimation; and Static route allocation and road capacity estimation are determined based on partial trajectory data and high-speed vehicle telemetry data.

5. The system according to claim 4, wherein, The at least one processor is also programmed to determine spectral spatiotemporal data extrapolation based on spectral decomposition.

6. The system according to claim 5, wherein, The at least one processor is also programmed to determine the extrapolation of spectral spatiotemporal data based on the objective function formulation.

7. The system according to claim 6, wherein, The at least one processor is also programmed to use an iterative method to reduce the mean square error between the extrapolation of spectral spatiotemporal data and the underlying facts.

8. A computer for identifying, classifying, mitigating road network congestion, and finding the root causes of road network congestion, the system having multiple motor vehicles located at multiple locations in a road network, each of the motor vehicles having at least one sensor for generating input, and each of the motor vehicles further having a telematics control unit for generating at least one location signal of the location of an associated motor vehicle and at least one event signal of an event related to the associated motor vehicle, the at least one location signal and the at least one event signal corresponding to high-speed vehicle telemetry data based on input from the at least one sensor, the computer comprising: At least one processor, said processor being connected to and receiving high-speed vehicle telemetry data from the telemetry control unit of the associated motor vehicle; and A non-transitory computer-readable storage medium, including instructions, such that at least one processor is programmed to: Identify road network congestion in the current time slot based on high-speed vehicle telemetry data; Based on high-speed vehicle telemetry data over a period of time, road network congestion is determined to be at least one of frequent congestion and infrequent congestion; Identify at least one of the sources of recurring congestion and the causes of infrequent congestion; and Generate a notification signal associated with at least one of recurring congestion, the source and location of recurring congestion, non-recurring congestion, and the cause and location of non-recurring congestion, such that in response to a display device receiving the notification signal from at least one processor, the display device displaying a signal associated with one of recurring congestion, the source and location of recurring congestion, non-recurring congestion, and the cause and location of non-recurring congestion based on high-speed vehicle telemetry data; The at least one processor is further programmed to determine the source of recurring congestion in the following manner: The origin-endpoint matrix derived from historical high-speed vehicle telemetry data is used to determine the first route capacity estimate for the road link; The second route capacity estimate for road links is determined using direct measurements from high-speed vehicle telemetry datasets. Compare the capacity estimates of the first and second routes; and The at least one processor is calibrated based on a comparison between the first and second route capacity estimates. The at least one processor is programmed to use a causal inference model to determine the causal relationship between causal factors in a first location in a first time slot and infrequent congestion in a second location in a second time slot. The at least one processor is programmed to use the causal reasoning model according to the following: in This indicates that at time slot t, at position p i Non-recurring congestion in the middle; among them Indicates in Located at position p i Historical congestion in the middle; of which m k Represents element The weighted vector function at time slot tk; where Y represents the infrequent congestion vector in historical data; where k represents the k-th historical time slot before the current time slot t; where Indicates in Located at position p j Historical non-recurring congestion in the region; where n k It is associated with at least one causal factor; where X represents a causal factor that favors the formation of infrequent congestion event Y; where The autocorrelation function representing the relationship between the vector of infrequent congestion events Y and the vector of causal factors X is Gaussian distributed. ; where O k Represents the elements of the autocorrelation function The weighting function; where k represents the k-th historical time slot before the current time slot t; and where This represents random Gaussian noise.

9. The computer according to claim 8, wherein, The at least one processor is also programmed to determine, based on statistical equations, causal factors in the first position within the first time slot that cause infrequent congestion in the second position within the second time slot: in Let X represent the causal relationship determination function, which is used to determine whether a non-recurring congestion event Y is caused by the attributable event X; where X is the causal relationship determination function. This represents the functional operation between the weighted vector n and the weighted vector o; where n represents the weighted vector of the attribution event vector; and o represents the weighted vector of the autocorrelation function between the attribution event vector and infrequent congestion events. The function operation represents the weighted vector m; where m represents the weighted vector of infrequent congestion events.

10. A method of operating a system for identifying, classifying, mitigating, and finding the root causes of road network congestion, the system having multiple motor vehicles located at multiple locations in a road network, each of the motor vehicles having at least one sensor for generating input and a telematics control unit for generating at least one location signal of the location of an associated motor vehicle and at least one event signal of an event related to the associated motor vehicle, the at least one location signal and the at least one event signal corresponding to high-speed vehicle telemetry data based on the input from the at least one sensor, the system further comprising a computer having at least one processor coupled to the telematics control unit of the associated motor vehicle and a non-transitory computer-readable storage medium including instructions, the method comprising: The at least one processor is used to identify road network congestion in the current time slot based on high-speed vehicle telemetry data; The at least one processor is used to determine, based on high-speed vehicle telemetry data over a period of time, whether road network congestion is at least one of recurring congestion and infrequent congestion. The at least one processor is used to determine at least one of the sources of recurring congestion and the causes of infrequent congestion; and The at least one processor generates a notification signal associated with at least one of recurring congestion, the source and location of recurring congestion, non-recurring congestion, and the cause and location of non-recurring congestion, such that in response to the display device receiving the notification signal from the at least one processor, the display device displays one of the associated recurring congestion, the source and location of recurring congestion, non-recurring congestion, and the cause and location of non-recurring congestion based on high-speed vehicle telemetry data. The method further includes using the at least one processor to determine the source of recurring congestion in the following manner: The at least one processor uses a origin-endpoint matrix derived from historical high-speed vehicle telemetry data to determine a first route capacity estimate for the road link; The at least one processor uses direct measurements from a high-speed vehicle telemetry dataset to determine a second route capacity estimate for the road link; The first and second route capacity estimates are compared using the at least one processor; and The at least one processor is calibrated using a comparison between the first and second route capacity estimates. The at least one processor uses a causal inference model to determine the causal relationship between causal factors in the first time slot at the first location and infrequent congestion in the second time slot at the second location. The causal inference model is used by at least one processor according to the following: in This indicates that at time slot t, at position p i Non-recurring congestion in the middle; among them Indicates in Located at position p i Historical congestion in the middle; of which m k Represents element The weighted vector function at time slot tk; where Y represents the infrequent congestion vector in historical data; where k represents the k-th historical time slot before the current time slot t; where Indicates in Located at position p j Historical non-recurring congestion in the region; where n k It is associated with at least one causal factor; where X represents a causal factor that favors the formation of infrequent congestion event Y; where The autocorrelation function representing the relationship between the vector of infrequent congestion events Y and the vector of causal factors X is Gaussian distributed. ; where O k Represents the elements of the autocorrelation function The weighting function; where k represents the k-th historical time slot before the current time slot t; and where This represents random Gaussian noise; and The at least one processor determines, according to the following statistical equation, the causal factors that cause infrequent congestion in the second position in the first time slot, at the first location, lead to the infrequent congestion in the second location in the second time slot: in Let X represent the causal relationship determination function, which is used to determine whether a non-recurring congestion event Y is caused by the attributable event X; where X is the causal relationship determination function. This represents the functional operation between the weighted vector n and the weighted vector o; where n represents the weighted vector of the attribution event vector; and o represents the weighted vector of the autocorrelation function between the attribution event vector and infrequent congestion events. The function operation represents the weighted vector m; where m represents the weighted vector of infrequent congestion events.

Images