Method and system for generating traffic flow data

Abstract

To provide a method for generating traffic data indicative of a traffic volume and / or traffic density within a navigable network in an area covered by an electronic map.SOLUTION: A method generally comprises: obtaining data indicative of a count of devices associated with vehicles traversing the navigable element represented by a segment of an electronic map in respect of a given time, where the count of devices is based on positional data and associated timing data relating to the movement of a plurality of devices associated with vehicles along the navigable element represented by the segment; and using the determined count data and a scaling coefficient to obtain data indicative of an estimated traffic volume for the segment in respect of the given time. The scaling coefficient is time dependent.SELECTED DRAWING: Figure 14

Description

[Technical Field]

[0001] The present invention relates to a method and system for generating data indicative of traffic volume within a navigable network, the navigable network being within an area covered by an electronic map, the electronic map including a plurality of segments representing navigable elements of the navigable network. [Background technology]

[0002] Traffic volume (also called traffic "flow") is a measure of the number of vehicles passing through a given cross section of a navigable element, such as a road, during a specified period of time.

[0003] Traffic volume is an important parameter for determining the average travel speed (or conversely, travel time) associated with road elements of a road network. The average travel speeds (or travel times) associated with different road elements of a road network may be taken into account when planning a route through the road network. For example, each road element of a navigable network may be represented by a road segment on an electronic map. The fastest route through the navigable network may be planned using the average speed (or travel time) data associated with the road segments, for example, using an appropriate cost function. Similarly, the average speed of the travel data may be used in determining the exact arrival time of the route. The more accurate the traffic volume data that can be obtained for the road elements of the road network, the more accurately the fastest route and / or estimated arrival time can be determined.

[0004] In addition to being important in the context of navigation, traffic volume is generally a key quantity for characterizing traffic conditions within a road network. Therefore, having knowledge of such traffic data can be very beneficial for traffic management and control purposes. For example, along with traffic speed, traffic volume is a key parameter for many traffic management and control applications. In general, traffic volume data may be used for a variety of applications to provide a more complete operational performance measurement. For example, traffic volume data can provide insight into real-time flows through a network, which may be useful for monitoring major events or occurrences occurring within the network, including monitoring the impact of travel information on detour routes (which typically do not include traffic monitoring systems). As another example, traffic volume data may be used to determine traffic demand patterns, e.g., for the calibration and validation of traffic signal patterns. Traffic volume data can also be used to estimate road capacity, e.g., for use in transportation planning models. As yet another example, traffic volume data can be combined with data reporting delays (or costs) caused by traffic congestion to estimate transportation costs.

[0005] Traditionally, traffic volume has been measured by directly counting the number of vehicles at a location within a road network using either manual or automatic counting methods. Automatic counting can be performed by employing various sensors at desired locations within the road network. For example, the most widely used technique for automatic counting relies on inductive sensing (e.g., inductive loop sensors embedded in the road network), although it is known to use video or radar sensors to automatically count vehicles passing through a given cross section of the road. These types of sensors can be expensive to install and maintain, and their availability varies greatly from location to location. Therefore, while such direct counting methods can provide accurate data, they cannot be easily scaled to provide wider coverage of the road network.

[0006] FIG. 9 shows a map area containing road segments. Expressways are shown in darker gray. The figure also shows the locations of inductive loops on the highways (dark spots 500 on the highways). Non-highway road segments generally do not have inductive loops. These road segments are shown in light gray or using a single gray line. This indicates that only a small percentage of road segments are equipped with traffic flow detectors.

[0007] Map data used by navigation applications is typically designed specifically for use by route guidance algorithms, using location data from a positioning system, e.g., a GPS or GNSS system. For example, roads can be described as lines, i.e., vectors (e.g., start point, end point, and road direction; an entire road is composed of hundreds of such segments, each uniquely defined by start / end direction parameters). A map is a set of such road vectors, data associated with each vector (e.g., speed limit, direction of travel), as well as points of interest (POIs), road names, and other geographic features such as parking lot boundaries, river boundaries, etc., all of which are defined in terms of vectors. All map features (e.g., road vectors, POIs, etc.) are typically defined in a coordinate system that corresponds to or is related to the coordinate system of a positioning system, such as a GPS system, allowing locations as determined through the positioning system to be located on the associated roads shown on the map and allowing optimal routes to destinations to be planned.

[0008] Within a given road network, several vehicles are associated with devices that include location detection means (e.g., GPS devices). Such devices can transmit location data that indicate their location, and therefore the location of the vehicle, with respect to time. Such data can be called "probe data," or more specifically, "vehicle probe data." Another commonly used term for such data is "floating car data." Each device (or vehicle) can be called a "probe."

[0009] Thus, probe data transmitted by a device associated with a vehicle provides an indication of the vehicle's movement through a network. In some embodiments, a device associated with a vehicle that transmits probe data may be a device running a navigation application. Such a device may be referred to as a "navigation device." Such a navigation device may include, for example, a dedicated navigation device, or any mobile device (e.g., a cell phone, tablet, or wearable device such as a watch) running a suitable navigation application, or may be implemented using an integrated in-vehicle navigation system. However, probe data may include location data obtained from any device associated with the vehicle and having positioning capability. For example, the device may include means for accessing and receiving information from a cellular communication network, such as a WiFi access point or a GSM device, and using this information to determine its location. Typically, however, the device includes a Global Navigation Satellite System (GNSS) receiver, such as a GPS receiver, for receiving satellite signals indicative of the receiver's position at a particular time, which preferably receives updated location information at regular intervals. Such devices may include navigation devices, mobile telecommunications devices with positioning capabilities, wearable devices with positioning capabilities, location sensors, etc. For example, a navigation application may cause the device running the application to periodically sample at least the device's current location. Such samples of location data are sometimes referred to as "probe" data samples. A probe data sample includes at least the location of the device and may include data indicative of the time to which that location pertains.

[0010] Optionally, other data can be included in the probe data sample. For example, the probe data sample can include a latitude coordinate value, a longitude coordinate value, and a time value, and optionally additional information such as one or more of a heading, a travel speed, an altitude, etc. The device is configured to transmit the probe sample data to a server. Such samples of location data, i.e., probe data samples, can be collected by the server from multiple devices traveling through navigable elements of a navigable network within a geographic area. The navigable elements can be represented by segments of an electronic map. The devices may be associated with vehicles traveling through elements of the network. The probe data sample obtained from a given device indicates the path traveled by the particular device.

[0011] Techniques for determining traffic volume using probe data have been proposed. Such techniques are advantageous in that traffic volume data can be obtained for any segment of interest without the cost and lack of flexibility associated with traditional techniques that use a fixed infrastructure of sensors. However, probe data is typically collected from only a portion of the total vehicle fleet. The penetration level can be defined as the percentage of vehicles from which probe data is collected. Penetration level may also be referred to as penetration rate, sample rate, (instrument) degree, or (relative) percentage (degree). For consistency, the term "penetration level" is used herein. Currently, the percentage of vehicles from which data is collected (i.e., "penetration level") is only on the order of about 10%, and even less in some areas. The low penetration rate and potentially uneven sampling rate mean that traffic volume is typically not (and cannot be) determined directly from probe data.

[0012] The electronic map includes a number of segments representing navigable elements (e.g., road elements) of a navigable network (e.g., a road network). Traffic volume for a segment of the electronic map may be estimated using counts of movements of the element represented by segment s by a device associated with a vehicle at time t indicated by probe data (i.e., counts of probes moving through the segment at the relevant time), and a scaling factor.

[0013] therefore,

number

[0014] The scaling factor represents the prevalence level, which in this example is the inverse of the prevalence level.

[0015] For ease of notation, time t may be expressed in time units of the aggregation time interval Δt. For example, one such system uses time as Δt, where time t is an hourly time indicator (t has a precision of 1 hour).

[0016] Thus, the scaling factor k is used to project the probe counts associated with a segment for a given time (e.g., an aggregation time interval) onto the traffic volume of the segment for the given time. It will be appreciated that the factor k (and therefore the penetration level Θ) is critical to the ability to perform this projection accurately.

[0017] For ease of reference, characteristics such as traffic volume, measured traffic volume or movement counts, whether measured or according to probe data, may be referred to herein with respect to navigable elements, e.g., navigable segments, e.g., road segments, of an electronic map representing road elements of an associated navigable network, e.g., a road network. Even if not explicitly stated, it will be understood that such characteristics, e.g., traffic volume or counts, etc., refer to the corresponding characteristics of the real-world navigable element represented by the segment.

[0018] Traffic flow detectors, such as inductive loops, can be used to directly measure the count of vehicles along a navigable element. The measured total traffic volume for a given navigable segment s at time t is expressed as Y(s,t). The total traffic volume Y(s,t) for a segment may be obtained corresponding to the measured count of vehicles passing along the road element represented by the segment at a relevant time, e.g., during a relevant aggregation time interval. This sensed or measured traffic volume can be used to estimate a scaling factor k that enables projecting traffic volume from the probe count for the segment at a relevant time. In other words, the measured count of vehicles serves as ground truth (ground truth data), from which the factor k can be directly estimated by comparing the measured count with a sample probe count X(s,t) for the same time t. As mentioned above, for simplicity of notation, the time t may be expressed in time units of Δt, e.g., using 1 hour as Δt, where t is a time indicator (t has a precision of 1 hour).

[0019] A common technique is to use a traffic flow detector to determine a constant scaling factor indicative of the occupancy level, e.g., a factor k that is the inverse of the occupancy level. While using a constant factor k is a commonly used approach, the applicant has recognized that this may introduce errors as occupancy levels vary at different times and / or between different road segments.

[0020] To illustrate this issue, experiments were performed using two sets of inductive loops. The inductive loops in the first set had k = 5.68 (Θ = 17.6%). Using this result to analyze the second set of inductive loops, a mean relative prediction error (MRE) of 12.9% was obtained. The observed error between the estimated actual traffic flow kX(s,t) and the measured traffic flow Y(s,t), both for the second set of inductive loops, is shown in Figure 10.

[0021] The figure shows the error distribution determined using the observed probe data and inductive loop data for the second set. The error distribution indicates that a better method is needed to estimate the coefficients used to obtain total traffic volume from the probe data.

[0022] WO 2019 / 158438, entitled "Methods and Systems for Generating Traffic Volume or Traffic Density Data" in the name of TomTom Traffic BV and published on August 22, 2019, describes a technique for estimating traffic volume on a road segment using average speeds of probes detected on the road segment and road segment parameters.

[0023] Another method for estimating traffic volume using probe data is described in U.S. Patent Application Publication No. 2015 / 0120174, entitled "Traffic Volume Estimation," in the name of HERE Global BV, and published on April 30, 2015. However, the method described in U.S. Patent Application Publication No. 2015 / 0120174 still suffers from various drawbacks.

[0024] Accordingly, applicants have recognized that a need remains for improved methods and systems for providing traffic data for navigable networks based on probe data. Summary of the Invention

[0025] According to a first aspect of the present invention, there is provided a method of generating traffic data indicative of traffic volume in a navigable network in an area covered by an electronic map, said electronic map including a plurality of segments representing navigable elements of the navigable network, said method comprising, for one or more segments of said electronic map: obtaining data indicative of a count of devices associated with a vehicle moving along the navigable element represented by the segment with respect to a given time, the count of devices being based on position data and associated timing data relating to movement of a plurality of devices associated with a vehicle along the navigable element represented by the segment; using the determined count data and a scaling factor to obtain data indicative of an estimated traffic volume for the segment for the given time, the scaling factor being a time-dependent scaling factor, the method including using the scaling factor for the given time when obtaining the estimated traffic volume for the segment.

[0026] Thus, according to the present invention, a count of devices associated with vehicles moving through a navigable element represented by a segment of the electronic map at a given time is obtained, the count being based on, i.e., obtained using, position data and associated timing data (i.e., vehicle probe data) relating to the movement of a plurality of devices associated with vehicles along the navigable element represented by the segment, the count data being used together with a scaling factor to obtain an estimated traffic volume for the segment for the given time, the estimated traffic volume indicating an estimated traffic volume for the navigable element represented by the segment for the given time.

[0027] The method may be performed with respect to one or more segments, which may be referred to as segments of interest. Any of the steps described herein in connection with determining traffic volumes and / or scaling factors for a segment (of interest) may be performed with respect to one or more additional segments of interest.

[0028] According to the invention, the scaling factors are time-dependent scaling factors: the scaling factors relating to (i.e. applicable to) the given time are used in obtaining the estimated traffic volume. Thus, rather than using a scaling factor that is constant over time, the (value of) the scaling factor will vary depending on the time considered, so that the same (value of) scaling factor is used for the required estimated traffic volume regardless of the time to which it relates.

[0029] The scaling factors are preferably estimated, i.e., calculated, scaling factors, rather than such factors being measured, although, as explained below, the process of estimating the estimated scaling factors may involve using measured data.

[0030] The scaling factor is used in estimating traffic volume for the segment. Traffic volume is a measure of the number of vehicles passing through a given cross section of a road during a specified time period. Thus, in an embodiment, to estimate traffic volume for a segment, the number of probe counts (i.e., "sample volume") for a given time (e.g., within a given time interval) is determined, and the determined sample volume is scaled using an appropriate scaling factor to estimate total traffic volume for the segment. The scaling factor may, for example, indicate a penetration rate that is inversely proportional to it.

[0031] The given time (to which the count data and scaling factor, and therefore the estimated traffic volume, pertain) is preferably a time interval. The time interval may be a recurring time interval, for example, a given time interval for a given day of the week. In a preferred embodiment, the given time is a time interval for a given day of the week.

[0032] The count of devices associated with vehicles moving through the navigable element represented by the segment with respect to the given time may be the count of such devices associated with vehicles moving through the navigable element at the given time, e.g., in a given time interval.

[0033] To determine a device count or perform other operations using vehicle position data, i.e., probe data, the position data must be aggregated over a time interval. In embodiments where the given time is a time interval, the time interval may be the time interval used in the aggregation of the position data, for example, to obtain the count of devices moving through the navigable element. The time interval may be an aggregation time interval over which the position data is aggregated when obtaining the count of devices. It will be understood that obtaining a count of devices moving through a navigable element represented by a segment will typically involve aggregating position data related to device movements associated with a vehicle, i.e., probe data along the element over a time window or “aggregation interval.” Any devices moving through the element within the aggregation interval will be counted when obtaining the count for that aggregation interval. If the aggregation time interval is a repeating interval, such as a day-of-week interval, probe data related to movements of the element over different weeks but the same time interval on the same day of the week may be counted for a time interval between 3:00 PM and 4:00 PM on Tuesday.

[0034] The time interval may be of any desired size. Typically, traffic volume is reported in vehicles per hour (or even vehicles per hour per lane in the case of multi-lane roads). In some embodiments, the interval is an hourly interval. However, the size of the aggregation time interval may generally be selected as desired, e.g., depending on the application. For example, for dynamic traffic phenomena such as traffic congestion, it may be desirable to report traffic volume at relatively short intervals, and the sample volume may be aggregated over a period of about 1 minute to about 1 hour. In other cases, such as traffic light calibration or traffic planning, it may be desirable to report traffic volume over longer intervals.

[0035] The time interval may be one of a set of predetermined time intervals. The time interval is preferably a regular time interval. For example, the time interval may be one of a set of predetermined time intervals obtained by dividing each day of the week into predetermined time units. The time unit may be an hour unit, or if more or less granularity is desired, a smaller unit such as 10 minutes, or a larger unit such as 20 minutes. Each time interval may be identified by a time index.

[0036] The given time (to which the count data and scaling factor, and therefore estimated traffic volume, pertains) may be obtained based on an indicated time of interest. The method may include receiving data indicating a segment of interest for which estimated traffic volume data is desired, using the data indicating the time of interest to identify the given time, and using data indicating a time of interest. The method may include using the data indicating the segment of interest for which estimated traffic volume data is desired and the data indicating the time of interest to obtain the data indicating the count of devices associated with vehicles moving through the navigable element represented by the segment for the given time. In a preferred embodiment in which the given time is a time interval, the data indicating the time of interest can be used to identify a time interval for obtaining the count data. The time interval may be a time interval that encompasses the time of interest.

[0037] The scaling factor is time-dependent in that the applicable value of the scaling factor varies depending on the time considered, e.g. depending on the time at which the estimated traffic volume of a segment is required.

[0038] The method may further comprise obtaining (a value of) the scaling factor for the given time period to be used in obtaining the estimated traffic volume for the segment. The method may comprise selecting the scaling factor from a set of scaling factors, each scaling factor relating to a different time period, e.g., a different time interval. In another embodiment, the method comprises determining, i.e., deriving, the scaling factor for the given time period. Whether or not the method comprises determining the scaling factor, the scaling factor is preferably an estimated scaling factor.

[0039] While it is envisaged that the scaling factor may be derived and stored for subsequent use, such that obtaining the scaling factor simply involves looking up an appropriate value applicable to the time of interest, the present invention allows for the scaling factor to be easily derived, for example using a database of probe data and measured vehicle count data for a segment of an electronic map. Advantageously, therefore, the scaling factor may be estimated on demand, for example in response to a request for estimated traffic volume. This may allow the most recent probe and measured count data to be taken into account. In some embodiments, the scaling factor is determined, i.e., derived, on the fly.

[0040] Whether or not the method includes deriving the scaling factor, the scaling factor used is applicable to the given time for which the estimated traffic volume is required. Preferably, the given time is a given time interval, one of a set of predetermined time intervals (e.g., obtained by dividing each day of the week into predetermined time units). If the scaling factor forms part of a set of scaling factors, the set of scaling factors may be a predetermined set of scaling factors. The set of scaling factors may include a scaling factor for each predetermined time interval. It will be appreciated that, for ease of processing, the given time to which the scaling factor relates is preferably the same as the given time interval for which the estimated traffic volume data is required (and to which the count data relates). However, this does not necessarily have to be the case, as long as the scaling factor used relates to a time interval related to the given time, e.g., a time interval that includes or is included in the given time interval to which the count (and therefore traffic volume) relates.

[0041] The method may include using received data indicative of the time of interest in obtaining the scaling factor to be used (which may include deriving the factor or selecting the factor from a set of predetermined factor).

[0042] In a preferred embodiment of the invention in any of its aspects, a count of devices associated with vehicles traveling on the segment is determined for a plurality of different given times, such as time intervals, and the determined count data and respective scaling factors are used to obtain the estimated traffic volume for the segment for each one of the plurality of different given times, a different respective scaling factor being used for each different given time.

[0043] The step of obtaining the data indicative of the count of devices based on the location (i.e., probe) data obtained from devices associated with a vehicle may simply involve looking up the applicable count of devices for the relevant segment and time. Thus, in some embodiments, the method includes obtaining the count data from a database of counts of devices moving through a navigable element of the navigable network represented by a segment of the electronic map for different times, e.g., time intervals. The time interval may correspond to a time interval used in the method of the present invention, or the appropriate count data may be derived from count data for other time intervals in the database (e.g., by summing count data for smaller time intervals). In other embodiments, the method extends to the step of determining the count data.

[0044] In some embodiments, the method includes obtaining position data and associated timing data regarding movement of a plurality of devices associated with a vehicle along the navigable element represented by the segment, and using the position data and associated timing data to determine the count of devices associated with vehicles moving along the segment with respect to the given time. This can be performed using appropriate filtering of the position data with respect to time. The count of devices may be obtained using only position data and associated timing data regarding movement of devices associated with a vehicle along the navigable element, i.e., using only probe data, and without using other forms of data obtained, for example, from sensors associated with the navigable element represented by the segment.

[0045] The scaling factor indicates the popularity level of the segment, for example, the scaling factor may be inversely related to popularity level.

[0046] The present invention therefore recognizes that the most appropriate value of the scaling factor for a segment depends on the time period under consideration. In other words, the time period under consideration must be taken into account in order to be able to more accurately estimate traffic volume based on counting vehicle segment movements according to probe data for the segment. For a given segment, the scaling factor that most accurately projects probe counts to estimated traffic volume may vary, for example, between parts of a day and / or between days of the week.

[0047] According to the invention, the time-dependent scaling factors may be obtained in any suitable manner, including obtaining the scaling factor for the given time from a database of scaling factors for different times, and in other embodiments, the method extends to determining the scaling factor (or, in embodiments, a set of scaling factors for different times).

[0048] The scaling factor may be one of a set of scaling factors obtained or obtainable for different times, the values ​​of which form a continuous value with respect to time, or one of a set of discrete scaling factors obtained or obtainable for different times (or time intervals). For example, a set of discrete scaling factors may be provided, each for a different time interval. The scaling factor may be obtained using a time-dependent function. The function may be configured to provide scaling factors for different times that form a continuous value with respect to time, or a set of discrete values ​​with respect to time.

[0049] Various techniques can be used to estimate the time-dependent scaling factors, which can be used to provide sets of such scaling factors for different times.

[0050] The scaling factor may be determined, i.e., derived, using data indicative of a count of vehicles traveling through one or more navigable elements represented by a segment of the electronic map for the given time period, as detected by at least one traffic detector associated with the elements, and a count of devices associated with vehicles traveling through each such navigable element for the given time period, as determined using position data and associated timing data relating to the movement of multiple devices associated with vehicles along the navigable elements. Thus, a scaling factor for a segment in which both such data types are present may be derived based on a relationship between the count of vehicles traveling through the element represented by the segment for a time period (e.g., an aggregation time interval) as determined using at least one traffic detector and as determined based on probe data. The scaling factor for a segment of interest may be based on such data related to the segment itself and / or such data related to one or more reference segments. For simplicity, the time period may correspond to the same time period or time interval used in determining the counts based on probe data to obtain the estimated traffic volume; however, other time intervals may be used, provided the resulting scaling factor is relevant to the given time period for which the estimated traffic volume is required, e.g., within or including the time period. Thus, in a simple embodiment, the scaling factor may be a measured scaling factor obtained using probe data and measured count data for segments for which both forms of data are available for the given time. This is in contrast to prior art arrangements in which a constant scaling factor is used regardless of the time considered. However, in a preferred embodiment, the scaling factor is estimated using probe data and measured count data for other segments.

[0051] Various possibilities can be envisioned for deriving scaling factors for segments of interest for different times based on a comparison of traffic counts based on probe data and traffic detector data for at least some segments for which both types of data are available for different times. Such segments can include the segment of interest and / or a set of one or more reference segments. The present invention allows for obtaining scaling factors for segments of interest for which traffic detector data is not available, and therefore the segments used in determining the scaling factors can include a set of one or more reference segments rather than the segment of interest. The scaling factors are estimated scaling factors. Several forms of aggregation between different (reference) segments can be envisioned to obtain an overall scaling factor value for each considered time.

[0052] In some preferred embodiments, the scaling factor for the (segment of interest) is an estimated scaling factor, estimated using data indicative of the similarity of a probe profile associated with the (segment of interest) to a respective one of one or more reference probe profiles, each reference probe profile being associated with a respective reference segment. As used herein, a probe profile or reference probe profile for a segment or reference segment refers to a profile indicative of changes in the count of devices associated with a vehicle moving through the navigable element represented by the segment or reference segment over time, determined based on position data and associated timing data relating to the movement of a plurality of devices associated with a vehicle along the navigable element represented by the segment or reference segment.

[0053] The scaling factor may be estimated using data indicating the similarity of the probe profile for the segment to each of a plurality of reference probe profiles, each associated with a different one of a plurality of reference segments. The scaling factor may be estimated using a set of similarity parameters indicating the similarity of the probe profile for the segment to the reference probe profiles. The method may include determining such a set of similarity parameters. The set of similarity parameters may include a similarity parameter indicating the similarity of the probe profile for the segment to the reference probe profile for each of the reference probe profiles. The similarity parameters may be continuous or may be expressed using a predetermined similarity scale, which may include discrete levels.

[0054] The method may include using a kernel function to determine the data, e.g., a similarity parameter, indicative of the similarity of the probe profile to a reference probe profile. The kernel function may be a non-negative kernel function. The function may accept two vector arguments and output a single real number in a predetermined range. The predetermined range may correspond to a desired predetermined similarity range, e.g., 0 to 1. The kernel function maps the similarity of two probe profiles to a function result of a real value. The kernel function may be a radial basis function. The number may be the similarity parameter. Thus, in an embodiment, the similarity profile is obtained using a kernel function.

[0055] Preferably, the or each reference segment is a segment representing a navigable element associated with at least one traffic detector. Thus, in these embodiments, each reference segment is a segment for which traffic detector data is available (e.g., to a server performing the methods described herein). Such data allows for determining a measured count of vehicles traveling on the segment at a given time (i.e., a given time interval). A traffic detector may be any device or system capable of detecting the presence of vehicles on the navigable element represented by the segment.

[0056] It will be appreciated that if measured traffic data for a segment is available, the exact value of the scaling factor can be determined for a given time based on a comparison of the measured count of vehicles traveling on that segment for that time with a count determined using probe data for that segment for that time.

[0057] In some embodiments, the scaling factors are estimated based on a set of one or more reference scaling factors, each reference scaling factor being a scaling factor determined for a respective one of one or more reference segments for the given time period based on data obtained from the at least one traffic detector associated therewith.

[0058] The reference scaling factor may be determined using: a count of devices associated with vehicles traveling the navigable element represented by the reference segment for the given time based on position data and associated timing data regarding the movement of a plurality of devices associated with vehicles along the navigable element represented by the reference segment; and a measured count of vehicles traveling the reference segment for the given time determined based on data measured by the at least one traffic detector associated with the reference segment. The reference scaling factor may be determined by dividing the measured count of vehicles by the count of vehicles determined using the position data regarding the movement of the plurality of devices (i.e., using vehicle probe data).

[0059] The given time to which the data used in estimating the scaling factor relates is preferably a time interval, e.g., an aggregation time interval. The given time preferably corresponds to a given time during which the count of devices moving through the navigable element represented by the segment for which estimated traffic volume is required is obtained and used to obtain data indicative of the estimated traffic volume for the segment for the given time. However, this does not necessarily have to be the case, as long as the given time used in estimating the scaling factor results in a scaling factor that is applicable to the given time for which estimated traffic volume is required, e.g., at least approximately corresponds to, e.g., overlaps with, or is included within, the given time, e.g., time interval.

[0060] Preferably, similarity data obtained as described according to any of the above techniques is used together with said reference scaling factor data in deriving scaling factors for said segments.

[0061] It can be assumed that the scaling factor of a segment of interest is more likely to be similar to the reference scaling factor associated with a reference segment having a reference probe profile that is more similar to the probe profile of the segment of interest. In embodiments, the similarity data is used to determine the contribution of each reference scaling factor to the estimated scaling factor determined for the segment of interest. In other words, the similarity data may be used to weight the contribution of each reference scaling factor. The method may be performed such that a reference scaling factor associated with a reference segment having a reference probe profile that is more similar to the probe profile associated with the segment of interest provides a greater contribution to the estimated scaling factor than a reference scaling factor associated with a reference segment having a reference probe profile that is less similar to the probe profile associated with the segment of interest. The greater the similarity of the reference probe profile associated with a reference segment to the probe profile associated with the segment of interest, the greater the contribution of the reference scaling factor to the estimated scaling factor of the segment of interest.

[0062] However, regardless of whether the method for determining the scaling factor includes consideration of the similarity between a probe profile for a segment and one or more reference probe profiles, in general, the scaling factor for a segment may be based on a set of one or more (preferably multiple) reference scaling factor(s), each reference scaling factor being a scaling factor determined for a set of one or more reference segments for the given time period using data obtained from at least one traffic detector associated with the reference segment. Thus, in such embodiments, the or each reference segment is a segment representing a navigable element associated with at least one traffic detector. The set of one or more reference segments preferably corresponds to the set of reference segments obtained in embodiments in which probe profile similarity data is obtained. Each reference scaling factor may provide a contribution to the estimated scaling factor.

[0063] In any embodiment using a reference scaling factor, the reference scaling factor may be determined using: a count of devices associated with vehicles traveling the navigable element represented by the reference segment for the given time based on position data and associated timing data regarding the movement of a plurality of devices associated with vehicles along the navigable element represented by the reference segment; and a measured count of vehicles traveling the reference segment for the given time determined based on data measured by the at least one traffic detector associated with the reference segment. The reference scaling factor may be determined by dividing the measured count of vehicles by the count of vehicles determined using the position data regarding the movement of the plurality of devices (i.e., using vehicle probe data).

[0064] Preferably, the scaling factor (for the segment of interest) is based on a set of multiple such reference scaling factors, and may be based on the sum thereof.

[0065] It has been found that some reference segments may be more important than others in deriving the correct scaling factor for a segment. In embodiments that use similarity data, this may be the case regardless of the similarity between the reference segment and the segment for which a scaling factor is required. Preferably, in any embodiment that uses reference scaling factors, the scaling factor for the (interest) segment is based on a weighted sum of the multiple reference scaling factors.

[0066] One or more sets of weighting values ​​may be derived, including a weighting value for each one of the set of reference segments, each weighting value indicating a weight to be assigned to the reference scaling factor associated with the reference segment when determining a scaling factor for the segment. The set of weighting values ​​may be defined by a vector. The scaling factor for the segment may be obtained using a set or multiple sets of such weighting values. Multiple sets of weighting values ​​may be obtained for different factors that affect the contribution of the reference scaling factor to the overall estimated scaling factor for the segment. The overall contribution of the reference scaling factor may be based on different factors, for example, weighting values ​​related to similarity to the segment in question and / or the results of implementing a linear regression training model.

[0067] The contribution of a given reference scaling factor to the estimated scaling factor of the (interest) segment may be based at least in part on the similarity of the reference probe profile associated with the reference segment with which the reference scaling factor is associated to the probe profile associated with the interest segment.

[0068] Alternatively or additionally, the relative importance of different reference segments, i.e., the reference scaling factors associated therewith in determining the scaling factor for a segment, can be found, for example, based on a comparison of the results obtained using an algorithm implementing the method for estimating scaling factors (e.g., based on similarity data and reference segment data) with the scaling factor associated with that segment based on measured data, i.e., traffic sensor data, for which both measured data and probe data are available, thereby capturing factors that influence the relative importance of reference segments in determining the estimated scaling factor for a segment of interest that are not due to probe profile similarity.

[0069] In some embodiments, the contribution of each reference scaling factor to the estimated scaling factor for the segment of interest is determined at least in part using a linear regression training model. For example, a set of weighting values ​​for use in obtaining a weighted sum of the plurality of reference scaling factors may be obtained using a linear regression training model. The model may use, for each reference segment, data indicative of a measured count of vehicles traveling through the navigable element represented by the reference segment for the given time, determined based on data measured by the at least one traffic detector associated with the navigable element represented by the reference segment. Such data may be used as ground truth data.

[0070] The step of determining the contribution of each reference scaling factor to the estimated scaling factor of the segment of interest using the linear regression training model may include determining estimated reference scaling factors for one or more of the reference segments for the given time period, comparing, for each such reference segment, the estimated reference scaling factor of the measured reference segment with the reference scaling factor of the segment obtained using data measured by the at least one traffic detector associated with the reference segment, and determining whether any adjustment of the contribution (e.g., of a set of weighting values) is necessary. The method may include adjusting the contribution of the reference scaling factor to the estimated reference scaling factor (e.g., adjusting a set of weighting values ​​with respect to the reference scaling factor) so that the resulting estimated reference scaling factor more closely matches the measured reference scaling factor. The steps of comparing the estimated reference scaling factor with the measured reference scaling factor, determining whether any adjustment is necessary, and, if necessary, performing such adjustment may be performed iteratively. The method may include obtaining a set of weighting values ​​using the linear regression training model in such a manner.

[0071] The method according to any of the embodiments described herein may be repeated to obtain estimated traffic volumes for the segment for one or more further given times, and for each such further given time, an applicable scaling factor may be obtained, e.g., estimated, relative to the further given time, e.g., based on the same further given time.

[0072] The invention extends to the step of estimating the scaling factor for the segment for the given time, and to a method of deriving a function that can be used to estimate the scaling factor for the segment for the given time.

[0073] According to a further aspect of the present invention, there is provided a method for generating traffic data indicative of traffic volume in a navigable network in an area covered by an electronic map, the electronic map including a plurality of segments representing navigable elements of the navigable network, the method comprising: receiving data indicative of a segment of interest, the segment for which estimated traffic volume data is desired, and data indicative of a time of interest; obtaining, for each of a set of a plurality of the segments of the electronic map, data indicative of a count of devices associated with vehicles moving through the navigable element represented by the segment for a given time selected based on the time of interest, the count of devices being based on position data and associated timing data relating to movement of a plurality of devices associated with vehicles along the navigable element represented by the segment; identifying a reference subset of the set of segments of the electronic map, each segment of the reference subset of segments being associated with at least one traffic detector; obtaining, for each of the reference subset of the set of segments of the electronic map, data indicative of a measured count of vehicles moving through the navigable element represented by the segment for the given time based on data obtained from the at least one traffic detector associated therewith; obtaining a reference scaling factor for each of the reference subset of segments, each reference scaling factor being a scaling factor determined for the given time based on the measured count of vehicles traveling on the segment for the given time, determined based on data measured by the at least one traffic detector associated with the segment, and a count of devices associated with vehicles traveling on the segment for the given time based on the position data and associated timing data regarding movement of the devices along the segment; determining an estimated scaling factor for the segment of interest for the given time based on the or each reference scaling factor obtained; In obtaining the estimated traffic volume for the segment, using the estimated scaling factor for the given time based on the position data and associated timing data regarding movement of a plurality of devices along the navigable element represented by the segment of interest with respect to time, and the data indicative of the count of devices associated with vehicles moving along the navigable element represented by the segment of interest with respect to the given time; A method is provided which includes:

[0074] The invention in this further aspect may include any or all of the features described in relation to the other aspects and embodiments of the invention.

[0075] The method may include any of the features described above in relation to the various steps of estimating the traffic data and / or scaling factors.

[0076] The time of interest is the time to which the estimated traffic volume applies.

[0077] The given time to which the count data based on probe data is applied may be a given time interval as described above.

[0078] The method may include, for each of the set of segments of the electronic map, obtaining position data and associated timing data relating to movement of a plurality of devices associated with vehicles along the navigable element represented by the segment, and using the position data and associated timing data to determine, for each of the set of segments, the count of devices associated with vehicles traveling on the segment for the given time selected based on the time of interest. The count of devices may be obtained using only the position data and associated timing data relating to movement of devices associated with vehicles along the navigable element, i.e., using only probe data and no other forms of data.

[0079] In other embodiments, the required count data based on probe data may be obtained from a database of such count data for different times, for example time intervals.

[0080] The obtained counts of movements of elements represented by segments of the reference subset of segments based on traffic detector data may be based entirely on data obtained from the at least one traffic detector associated with the elements. Obtaining the counts of movements based on measured data for a segment may include obtaining the counts from a database of such counts (e.g., a database containing such measured count data for different times), or may extend to determining the counts based on data from the at least one traffic detector.

[0081] The reference subset of the set of segments preferably comprises a plurality of segments and may include all segments associated with at least one traffic detector in the set of segments under consideration.

[0082] The reference scaling factor may be obtained by dividing the measured count of a vehicle by the count determined using probe data.

[0083] Each reference scaling factor may provide a contribution to the estimated scaling factor.

[0084] The method includes: for each segment in the set of reference segments, obtaining a reference probe profile indicating a change in a count of devices associated with a vehicle traversing the navigable element represented by the segment with respect to time, the change being determined based on position data and associated timing data relating to the movement of devices associated with a vehicle along the navigable element represented by the segment; and obtaining, for the segment of interest, a probe profile indicative of changes in the count of devices associated with vehicles moving through the navigable element represented by the segment with respect to time based on position data and associated timing data relating to movement of devices associated with vehicles along the navigable element represented by the segment; The estimated scaling factor for the segment of interest for the given time further depends on the similarity of the probe profile associated with the segment of interest to each of one or more of the reference probe profiles.

[0085] Thus, the method may include determining an estimated scaling factor for the segment of interest for the given time using data indicative of the similarity of the probe profile associated with the segment of interest to each (and preferably each) of one or more of the acquired reference probe profiles (i.e., reference probe profiles for one or more or each of the set of reference segments).

[0086] The probe profile and reference profile are each based on probe data. The method may further include determining the reference probe profile and probe profile for the segment of interest based on acquired position data and associated timing data relating to movement of a plurality of devices associated with a vehicle along the navigable element represented by the segment. Such profiles may be acquired using existing count data for different times, for example from a database, or such count data may be acquired using the probe data.

[0087] The method may include comparing the probe profile for the segment of interest to each of a plurality of reference probe profiles (e.g., each reference profile considered) and determining a similarity parameter indicative of the similarity of the probe profile for the segment of interest to the reference probe profiles. Thus, the method may include obtaining a set of similarity parameters indicative of the similarity of the probe profile for the segment of interest to each of the reference probe profiles. The scaling factor for the segment of interest may be estimated using the set of similarity parameters.

[0088] The similarity data may be used together with the reference scaling coefficient data in deriving the estimated scaling coefficients for the segment of interest, for example, to determine the contribution of each reference scaling coefficient to the estimated scaling coefficient determined for the segment of interest.

[0089] The method may include determining the estimated scaling factor for the segment of interest based on each obtained reference scaling factor and a similarity of the probe profile associated with the segment of interest to each of the reference profiles.

[0090] The estimated scaling factor for the segment of interest may be based on a weighted sum of the reference scaling factors. In some embodiments, the contribution of each reference scaling factor to the estimated scaling factor for the segment of interest is determined at least in part using a linear regression training model. A set of weighting values ​​for use in obtaining the weighted sum of the plurality of reference scaling factors may be obtained using a linear regression training model. The linear regression training model may use, for each reference segment, data indicative of a measured count of vehicles traveling through the navigable element represented by the reference segment for the given time, determined based on data measured by the at least one traffic detector associated with the navigable element represented by the reference segment. Such data may be used as ground truth data.

[0091] Determining the contribution of each reference scaling factor to the estimated scaling factor for the segment of interest may be performed in any of the ways described above.

[0092] The contribution of a reference scaling factor to an overall estimated scaling factor for a segment of interest may be based at least in part on a degree of similarity of the reference probe profile associated with the reference segment with which the reference scaling factor is associated to the probe profile associated with the segment of interest. The safety segment for which the estimated scaling factor is determined may be a segment representing a navigable element not associated with any traffic detector. Thus, the segment of interest may not form part of the subset of reference segments.

[0093] According to a further aspect of the present invention, there is provided a method of estimating traffic volume for a given time for a given segment of an electronic map representing a navigable network in an area, the electronic map including a plurality of segments representing navigable elements of the navigable network in the area, the navigable network in the area including navigable sections associated with at least one traffic detector and navigable sections not associated with any traffic detector, the given segment being a segment representing at least a portion of a navigable section of the navigable network in the area not associated with any traffic detector, the electronic map further including a plurality of reference segments, each reference segment representing a portion of a navigable section of the navigable network in the area associated with a traffic detector. a reference segment representing at least a portion of a route along which a route is traveled, the reference segment being associated with data indicative of a respective reference scaling factor for the given time, the reference scaling factor being based on a measured count of vehicles traveling along the at least a portion of the route represented by the reference segment for the given time, and a count of devices associated with vehicles traveling along the at least a portion of the route represented by the reference segment for the given time, the measured count of vehicles being based on data measured by the at least one traffic detector associated with the route, and the count of devices associated with vehicles being based on position data and associated timing data relating to movement of a plurality of devices along the at least a portion of the route represented by the reference segment; estimating the traffic volume for the given segment for the given time using data indicating a count of devices associated with vehicles traveling along at least a portion of the section of the navigable network represented by the given segment for the given time and an estimated scaling factor for the given segment for the given time, wherein the estimated scaling factor for the segment is based on the reference scaling factor associated with each of a subset of one or more reference segments of the electronic map associated with the given segment, and the count of devices is based on position data and associated timing data regarding movement of a plurality of devices along at least a portion of the navigable network represented by the given segment for the given time; generating data indicative of the estimated traffic volume for the given segment for the given time; A method is provided which includes:

[0094] The invention in this further aspect may include any or all of the features described in relation to the other aspects and embodiments of the invention.

[0095] The method may include any of the features described above in relation to the various steps of estimating the traffic data and / or scaling factors.

[0096] The given time is the time to which the estimated traffic volume applies.

[0097] As in earlier aspects and embodiments, the given time may be any suitable time, such as a current time or a future time. The given time may also be a past time. The given time may be obtained based on an indicated time of interest. The time of interest may be determined in any of the ways described with respect to earlier aspects and embodiments of the present invention. Similarly, the given segment may be determined in any of the ways described above and may be referred to as a segment of interest. Data indicative of the segment of interest and / or time of interest may be received from any suitable source, as previously described.

[0098] The given time to which the count data based on probe data applies may be a given time interval as described above. The given time may also be a recurring time interval. For example, the given time may be a time interval on a given day of the week.

[0099] The method may be performed for one or more given segments of the electronic map, in which case the described steps would be performed for each given segment considered.

[0100] As described in connection with the preceding aspects and embodiments of the present invention, the navigable network within the considered area includes navigable sections associated with at least one traffic detector and navigable sections not associated with any traffic detector. A navigable section may include at least a portion of one or more navigable elements of the navigable network. In some embodiments, the navigable section corresponds to each navigable element of the navigable network.

[0101] A navigable segment associated with at least one traffic detector is such that a measured count of vehicles traveling on said segment at a given time is available for said segment. In contrast, such measured count data is not available for a navigable segment that is not associated with any traffic detector. A traffic detector may be defined as described above.

[0102] The given segment is a segment that represents at least a portion of a navigable section of the navigable network in the area that is not associated with any traffic detector. Thus, the given segment is a segment for which measured count data is not available. The given segment may represent a navigable element for which measured count data is not available (i.e., not associated with any traffic detector). The or each given segment may be considered a "non-reference" segment.

[0103] A navigable section includes at least a portion of one or more navigable elements of the navigable network.

[0104] The reference segments may be as defined above. Each reference segment represents at least a portion of a navigable section for which measured count data is available. Each reference segment may represent a navigable element for which measured count data is available (i.e., associated with at least one traffic detector). The reference segment may represent at least a portion of one or more navigable elements of the navigable network that form the section for which measured count data is available. A navigable section may be considered to be associated with a traffic detector(s) when measured count data is available for the navigable section, regardless of where the traffic detector(s) are located. Thus, a particular portion of the section represented by the reference segment need not itself comprise a traffic detector, provided that measured count data applicable to the portion of the section represented by the reference segment is available based, for example, on data obtained from one or more traffic detectors positioned at any suitable location to determine a count of vehicles traveling along the section.

[0105] In these further aspects or embodiments of the invention, it will be appreciated that, indeed, according to any of the aspects or embodiments described herein, the given segment may be a segment representing at least a portion of a navigable section of the navigable network in an area where data indicating an absolute count (absolute number) of vehicles traveling on at least a portion of the navigable section is not available. Conversely, the or each reference segment is a segment representing at least a portion of a navigable section of the navigable network in an area where data indicating an absolute count (absolute number) of vehicles traveling on at least a portion of the navigable section is available. Thus, more broadly, "non-reference" and "reference" segments may be segments for which absolute count data is available and for which it is not. Such absolute count data may be data obtained from a traffic detector associated with the section, as described herein.

[0106] Each reference segment is associated with data indicating a respective reference scaling factor for the given time. The reference scaling factor is based on a measured count of vehicles traveling at least a portion of the section represented by the reference segment for the given time and a count of devices associated with vehicles traveling at least a portion of the section represented by the reference segment for the given time. The measured count of vehicles is based on data measured by the at least one traffic detector associated with the section, and the count of devices associated with vehicles is based on position data and associated timing data regarding movement of multiple devices along at least a portion of the section. As described above, the at least one traffic detector may be associated in any manner with the section represented by the reference segment to provide measured count data for the portion of the section represented by the reference segment that is applicable to (and may not necessarily be located on) the portion of the section represented by the reference segment.

[0107] Thus, the reference scaling factor is based on the measured count data applicable to the segment at the given time (i.e., as measured by the at least one traffic detector) and a vehicle probe data-based count for the segment for the given time. The reference scaling factor may be based on a ratio (proportion) of the measured count to probe data count for the segment at the given time. The reference scaling factor may be obtained by dividing the measured count of vehicles by the count determined using probe data.

[0108] The reference segment may be associated in any suitable manner with data indicative of a reference scaling factor for the given time. Each reference segment may be associated with data indicative of a time-dependent reference scaling factor profile from which the reference scaling factor for the given time may be obtained. The reference scaling factor profile indicates a change in the reference scaling factor for the reference segment with respect to time. The reference scaling factor for the given time may be obtained from the scaling factor profile. The reference scaling factor profile may be based at least in part on a probe profile for the reference segment. The probe profile indicates a change in count of devices associated with a vehicle traversing at least a portion of the navigable section represented by the given reference segment with respect to time, determined based on position data and associated timing data relating to movement of a plurality of devices associated with a vehicle along at least a portion of the navigable section represented by the reference segment. The reference scaling factor profile may be based on such a probe profile and a profile indicative of a change in the measured count of vehicles traversing at least a portion of the section represented by the reference segment with respect to time.

[0109] The reference scaling factors can be based on live data and / or historical data. For example, regardless of whether the given time is past, present, or future, historical probe data in the form of a probe profile may be used in obtaining the reference scaling factors. The probe profile indicates the count of devices used in obtaining the reference scaling factors. The measured count data used may also be based on live data and / or historical data. Thus, the probe data and measured count data used to provide the reference scaling factors may be based on live data, historical data, or any combination thereof, provided that the resulting factors can be considered applicable to the current time. Advantageously, if the given time is the current time, the reference scaling factors are based at least in part on live data, e.g., live measured counts and / or live probe data. This may result in more accurate factors. Live data is data that can be considered to reflect current conditions in the navigable network. Historical data may or may not be used additionally.

[0110] The traffic volume of the given segment for the given time is estimated using a count of devices associated with vehicles traveling on at least a portion of the section of the navigable network represented by the given segment for the given time, i.e., vehicle probe data for the segment for the given time, and an estimated scaling factor for the given segment for the given time, the estimated scaling factor being based on one or more of the reference scaling factors.

[0111] The count based on probe data and measured data for the reference segment may be based on live data and / or historical data. For example, historical probe data in the form of a probe profile may be used regardless of whether the given time is past, present, or future. Live data may be used in addition to or as a substitute for historical data when the given time is the current time. A probe profile indicates changes in the count of devices associated with a vehicle traversing at least a portion of the navigable section represented by the given reference segment over time, determined based on position data and associated timing data regarding the movement of multiple devices associated with a vehicle along at least a portion of the navigable section represented by the reference segment.

[0112] The reference scaling factors associated with one, more than one, or all of the reference segments may be used in determining the traffic volume for the given segment. Thus, the estimated scaling factors may be based on a subset of a set of reference scaling factors associated with some of the reference segments, which subset may be a single one of the reference scaling factors or may include a plurality thereof.

[0113] The method may further include determining a subset of one or more of the plurality of reference segments of the electronic map associated with the given segment, the estimated scaling factor being based on a reference scaling factor associated with each of the set of one or more of the reference segments. The navigable network may include various sections that may give rise to reference segments of the electronic map. Some of these reference segments may be more relevant than others when determining the estimated scaling factor for the given segment. Thus, in these embodiments, only the reference scaling factors associated with reference segments that may be considered associated with the given segment are considered when determining the estimated scaling factor for the given segment. The subset of reference segments associated with the given segment may be a single reference segment or multiple reference segments that form a subset of the total number of reference segments of the electronic map representing sections of the navigable network in the area associated with a traffic detector. Thus, the given segment may be associated with a single reference segment or multiple reference segments.

[0114] The subset of one or more reference segments may be associated with the data representing the given segment in the electronic map. Thus, the association between the reference segments and the segment of interest may be known. In other embodiments, the method may extend to identifying the subset of one or more reference segments.

[0115] Whether or not the method extends to identifying the subset of one or more reference segments associated with the given segment, the subset of one or more reference segments may be selected in any suitable manner.

[0116] The subset of one or more reference segments may be determined based on a comparison of a probe profile associated with the given segment to reference probe profiles associated with some of the reference segments, the probe profile indicating a change in count of devices associated with vehicles traversing at least a portion of the navigable section represented by the given segment over time, determined based on position data and associated timing data regarding the movement of multiple devices associated with vehicles along at least a portion of the navigable section represented by the segment, and the reference probe profile indicating a change in count of devices associated with vehicles traversing at least a portion of the navigable section represented by the reference segment over time, determined based on position data and associated timing data regarding the movement of multiple devices associated with vehicles along at least a portion of the navigable section represented by the reference segment. Thus, the probe profile and reference probe profile are time-dependent probe profiles, such as weekly probe profiles.

[0117] The subset of one or more reference segments may include (or correspond to) one or more reference segments having a reference probe profile determined to be most similar to the probe profile of the given segment. The similarity may be assessed in any suitable manner. For example, a single most similar reference segment may be identified. Alternatively, each reference probe profile may be assigned a respective similarity value, and the reference probe profiles may be ranked in order of similarity. A predetermined number of the most similar probe profiles may be selected, or all reference probe profiles having a similarity above a predetermined threshold may be included in the set of one or more reference segments.

[0118] Alternatively or additionally, the selection of the subset of one or more reference segments may be based at least in part on the proximity of the reference segments to the given segment. The proximity may be temporal and / or spatial proximity. For example, the subset of reference segments may include or correspond to a predetermined number of closest reference segments in terms of travel time or distance. Alternatively, the subset of reference segments may include or correspond to segments within a predetermined travel time or distance of the given segment. The distance and / or travel time may be measured based on a straight-line path between the segments or along the road network.

[0119] Alternatively or additionally, the selection of the subset of one or more reference segments may be based at least in part on the similarity of one or more characteristics of the reference segments to the given segment. For example, the characteristics may include functional road class. Any relevant characteristics may be taken into account. Each reference segment may be assigned a similarity value indicating its similarity to the given segment. For example, the subset of reference segments may include or correspond to a subset of reference segments considered most similar to the given segment (e.g., a predetermined number of the segments, or segments having a similarity above a given threshold, etc.).

[0120] Any one or more of the above measurements may be used to attempt to obtain the subset of reference segments for use in determining the estimated scaling factor that may be expected to be associated with the given segment. When multiple factors are used, any suitable technique may be used to obtain a subset of reference segments that simultaneously meets all the criteria considered. Some weighting of the different criteria may also be used.

[0121] The estimated scaling factor for the given segment may be determined in any of the ways described above using the one or more reference scaling factors on which it is based.

[0122] The estimated scaling factor for the given segment may be estimated using data indicative of a similarity of a probe profile associated with the given segment to each of a set of one or more reference probe profiles, each reference probe profile being associated with a respective one of the one or more reference segments whose reference scaling factor is used in determining the estimated scaling factor. The probe profile may indicate a change in a count of devices associated with a vehicle traversing at least a portion of the navigable section represented by the given segment with respect to time, determined based on position data and associated timing data regarding movement of a plurality of devices associated with a vehicle along at least a portion of the navigable section represented by the segment, and the reference probe profile may indicate a change in a count of devices associated with a vehicle traversing at least a portion of the navigable section represented by the reference segment with respect to time, determined based on position data and associated timing data regarding movement of devices associated with a vehicle along at least a portion of the navigable section represented by the reference segment.

[0123] When multiple reference scaling factors are considered, the contribution of a given reference scaling factor to the estimated scaling factor for the given segment may be based at least in part on the similarity of the reference probe profile associated with the reference segment to which the reference scaling factor is associated to the probe profile associated with the given segment.

[0124] Alternatively or additionally, the contribution of each reference scaling factor to the estimated scaling factor may be based at least in part on the proximity of the reference segment associated with the reference scaling factor to the given segment. The proximity may be temporal or spatial proximity and may be of any of the types described above in connection with determining the subset of reference segments. A greater weight may be assigned to reference scaling factors associated with reference segments that are closer to the given segment. The contribution of a reference scaling factor associated with a given reference segment to the estimated scaling factor for the given segment may approach a value corresponding to the average of the reference scaling factors considered as the distance of the given reference segment from the given segment increases. The distance may be temporal or spatial.

[0125] The estimated scaling factor may be based on a weighted sum of a number of reference scaling factors.

[0126] Data indicative of a set of weighting values ​​to use in deriving the weighted sum of the plurality of reference scaling factors may be obtained using a linear regression training model, which may use data indicative of the measured count of vehicles traveling through the at least portion of the navigable section represented by the reference segment for the given time, determined based on data measured by the at least one traffic detector associated with the navigable section.

[0127] Thus, according to these further aspects of the invention, the reference scaling factor associated with each reference segment may be based on a ratio (proportion) of the measured count for the given time based on the traffic detector data to the device count for the given time based on the location data and associated timing data.

[0128] Each reference segment may be associated with data indicative of a time-dependent reference scaling factor profile, said reference scaling factor profile indicative of the variation of said reference scaling factor for said reference segment with respect to time.

[0129] The reference scaling factor profile may be based at least in part on a probe profile indicative of changes in counts of devices associated with a vehicle moving through at least a portion of the navigable section represented by the given reference segment over time, determined based on position data and associated timing data relating to movement of a plurality of devices associated with a vehicle along at least a portion of the navigable section represented by the reference segment.

[0130] The given time may be the current time, and the reference scaling factor may be based at least in part on live data.

[0131] The method may further include determining one or more subsets of the reference segments associated with the given segment, the estimated scaling factor being based on the reference scaling factor associated with each of the one or more subsets of the reference segments.

[0132] The subset of one or more reference segments may be determined based at least in part on a comparison of a probe profile associated with the given segment with reference probe profiles associated with some of the reference segments, the probe profile indicating a change in the count of devices associated with a vehicle moving through at least a portion of the navigable section represented by the given segment over time, determined based on position data and associated timing data regarding the movement of multiple devices associated with a vehicle along at least a portion of the navigable section represented by the segment, and the reference probe profile indicating a change in the count of devices associated with a vehicle moving through at least a portion of the navigable section represented by the reference segment over time, determined based on position data and associated timing data regarding the movement of multiple devices associated with a vehicle along at least a portion of the navigable section represented by the reference segment.

[0133] The subset of one or more reference segments may include one or more reference segments having a reference probe profile determined to be most similar to the probe profile of the given segment.

[0134] The selection of the subset of one or more reference segments may be based at least in part on the proximity of the reference segments to the location of the given segment.

[0135] The selection of the subset of one or more reference segments may be based at least in part on the similarity of the characteristics of the reference segments to the given segment, for example, the characteristics including Functional Road Class (FRC).

[0136] The estimated scaling factor for the given segment can be estimated using data indicating a similarity of a probe profile associated with the given segment to each of a set of one or more reference probe profiles, each reference probe profile being associated with a respective one of the one or more reference segments for which a reference scaling factor is used in determining the estimated scaling factor, the probe profiles indicating changes in the count of devices associated with a vehicle moving through at least a portion of the navigable section represented by the given segment over time, determined based on position data and associated timing data regarding the movement of a plurality of devices associated with a vehicle along at least a portion of the navigable section represented by the segment, and the reference probe profile indicating changes in the count of devices associated with a vehicle moving through at least a portion of the navigable section represented by the reference segment over time, determined based on position data and associated timing data regarding the movement of a plurality of devices associated with a vehicle along at least a portion of the navigable section represented by the reference segment.

[0137] The estimated scaling factor for the given segment may be based on a plurality of the reference scaling factors, and the contribution of a given reference scaling factor to the estimated scaling factor for the given segment is based at least in part on the similarity of the reference probe profile associated with the reference segment to which the reference scaling factor is associated with the probe profile associated with the given segment.

[0138] The estimated scaling factor for the given segment may be based on a plurality of the reference scaling factors, the contribution of each reference scaling factor to the estimated scaling factor being based at least in part on the proximity of the reference segment associated with the reference scaling factor to the given segment, and optionally, a greater weight being assigned to a reference scaling factor associated with a reference segment that is closer to the given segment.

[0139] The estimated scaling factor may be based on a weighted sum of a number of reference scaling factors.

[0140] Data indicative of a set of weighting values ​​to use in obtaining the weighted sum of the plurality of reference scaling factors may be obtained using a linear regression training model.

[0141] The linear regression training model can use data indicative of a measured count of vehicles traveling through at least the portion of the navigable section represented by the reference segment for the given time, determined based on data measured by the at least one traffic detector associated with the navigable section.

[0142] The method may further include receiving data indicative of the given segment for which traffic data is required and data indicative of a time of interest, and using the data indicative of the time of interest to identify the given time.

[0143] The given time may be the present time or a time in the future.

[0144] The given time may be a time interval, optionally a recurring time interval, such as a time interval for a given day of the week.

[0145] The method may further include associating data indicative of the estimated traffic volume with data indicative of the given segment to which it relates, and may optionally further include at least one of transmitting data indicative of the obtained estimated traffic volume for the given segment and displaying data indicative of the obtained estimated traffic volume for the given segment to a user.

[0146] The method may further include storing the estimated traffic volume and / or traffic density for later display and / or displaying the estimated traffic volume and / or traffic density to a user.

[0147] The method according to any of its aspects or embodiments may be repeated for one or more additional segments of interest and / or one or more additional times of interest. Accordingly, the method may include determining an estimated traffic volume for at least one further time of interest for the (same) segment of interest using the estimated scaling factor determined for a different given time selected based on the additional time of interest. The method may extend to estimating the scaling factor for a different given time. The applicant also recognizes that, to obtain a more accurate estimate of traffic volume for a segment, the (value of the) scaling factor should vary depending on the location of the (interest) segment. Preferably, the scaling factor (used or estimated according to the present invention in any of its aspects or embodiments) is location-dependent, and the (value of the) scaling factor used is applicable to the location of the segment under consideration. The value of the scaling factor may be specific to the segment under consideration, or in other embodiments, the same scaling factor may be applicable to more than one segment, e.g., segments within a given geographic area. The scaling factor is again preferably an estimated scaling factor.

[0148] Such an embodiment is considered advantageous in itself, regardless of whether the scaling factor is time dependent or not.

[0149] From a further aspect of the present invention, there is provided a method for generating traffic data indicative of traffic volume in a navigable network in an area covered by an electronic map, said electronic map including a plurality of segments representing navigable elements of said navigable network, said method comprising the steps of: obtaining data indicative of a count of devices associated with a vehicle moving along the navigable element represented by the segment with respect to a given time, the count of devices being based on position data and associated timing data relating to movement of a plurality of devices associated with a vehicle along the navigable element represented by the segment; using the determined count data and a scaling factor to obtain data indicative of an estimated traffic volume for the segment for the given time, the scaling factor being a location-dependent scaling factor, the method including using the scaling factor related to a location associated with the segment when obtaining the estimated traffic volume for the segment.

[0150] The invention in this aspect may include any or all of the features described in relation to the previous aspects of the invention, and vice versa, to the extent not mutually inconsistent.

[0151] The estimated traffic volume is for a given time, which is preferably a time interval, as described in the previous embodiment.

[0152] Preferably, the scaling factors for the segments are estimated scaling factors.

[0153] Preferably, the scaling factor is additionally time dependent, and thus may be relative to the given time, as in the previously described embodiments, it being understood that in at least embodiments the above time dependent embodiments also provide position dependent, i.e. segment dependent, estimated scaling factors.

[0154] The method for estimating location-dependent scaling factors may be performed using a set of reference scaling factors, similar to the timing-dependent embodiments described above, whether or not they are additionally time-dependent. However, in these further embodiments, the reference scaling factors need not be related to the given time to which the count data used in determining the estimated traffic volume pertains. In other words, they need not be time-dependent.

[0155] In an embodiment, the scaling factor is based on a set of one or more reference scaling factor, each reference scaling factor being a scaling factor determined for one of a set of one or more reference segments, each reference segment being a segment representing a navigable element associated with at least one traffic detector, and the reference scaling factor for the reference segment being obtained using data obtained from the at least one traffic detector associated therewith.

[0156] Each reference scaling factor can be determined using a count of devices associated with vehicles moving through the navigable element represented by the reference segment based on position data and associated timing data relating to the movement of multiple devices associated with vehicles along the navigable element represented by the reference segment, and a measured count of vehicles moving through the reference segment determined based on data measured by the set of one or more traffic detectors associated with the reference segment. Each probe and measured count can be for any reference time (i.e., time interval), preferably corresponding to the given time. However, it will be understood that in time-independent embodiments, the reference time can be a time different from the given time. For example, the same reference time can be obtained to determine the reference scaling factor used to determine an estimated scaling factor for a segment of interest for different given times.

[0157] Location-dependent embodiments may or may not also include considering similarity between a probe profile of the (interest) segment and a reference probe profile associated with each of one or more, or preferably multiple, reference segments associated with at least one traffic detector. Similarity-considering embodiments may be performed as described above, for example, using a kernel function to evaluate similarity. In this case, the set of reference segments considered in connection with the similarity evaluation may or may not correspond to reference segments considered in connection with the location.

[0158] As in the previous embodiment, the position-dependent scaling factor may be estimated based on contributions from multiple reference scaling factors, for example, the estimated scaling factor may be based on a weighted sum of the multiple reference scaling factors.

[0159] The contribution of each reference scaling factor to the overall estimated scaling factor for the (interest) segment may be based on the distance between the location associated with the reference segment associated with the reference scaling factor and the location of the segment under consideration. A greater weight may be assigned to a reference scaling factor associated with a reference segment that is closer to the location of the segment under consideration. The contribution of a reference scaling factor associated with a given reference segment in a given electronic map area to the estimated scaling factor for the segment may approach a value corresponding to the average of the reference scaling factors associated with reference segments in the given electronic map area as the distance of the location associated with the given reference segment from the location of the (interest) segment increases. This may be achieved using a decay function. The decay function may decay to an average reference scaling factor contribution value for the area. When evaluating distance in these embodiments, any suitable reference distance may be used, such as, for example, a Euclidean distance measured between reference points such as the start, middle, or end of a road segment; a routing distance such as the shortest or fastest distance; or a distance that depends on road class.

[0160] The contribution of each reference scaling factor to the overall estimated scaling factor for the (segment of interest) may alternatively or additionally be based on the similarity of a reference probe profile associated with the reference segment to that of the (segment of interest).

[0161] The contribution of each reference scaling factor to the overall estimated scaling factor for the segment (of interest) may alternatively or additionally be based at least in part on the results of running a linear regression training model. In any of these embodiments, the scaling factors may be time-dependent. For example, this may be done using appropriate time-dependent data in a linear regression training model such as those described above.

[0162] The model may use data indicative of a measured count of vehicles moving through the navigable element represented by the reference segment for a given time, determined based on data measured by the at least one traffic detector associated with the navigable element represented by the reference segment.

[0163] The method according to any of its aspects or embodiments may be repeated for one or more additional segments of interest and / or one or more additional times of interest. Accordingly, the method may include determining estimated traffic volumes for at least one further time of interest for the (same) segment of interest using a scaling factor determined for a different given time selected based on the additional times of interest. The method may extend to estimating the scaling factor for the different given time.

[0164] According to the present invention in any of its aspects or embodiments, the set or sets of segments of the electronic map from which data indicative of a count of devices associated with vehicles moving through the navigable element represented by the segments with respect to a given time is obtained, the count of devices being based on position data and associated timing data relating to the movement of a plurality of devices associated with vehicles along the navigable element represented by the segments, may be a set of a plurality of the segments of the electronic map within a given map area.

[0165] The segments for which the scaling factor is determined according to the present invention in any of its aspects or embodiments are preferably segments for which measured traffic data obtained by measuring traffic moving through the element represented by the segment is unavailable, for example, to a server performing the methods described herein. The segments may also be segments representing navigable elements not associated with any traffic detector. Such segments therefore do not form part of the set of reference segments associated with a traffic detector.

[0166] References herein to traffic detectors associated with elements represented by segments (or for ease of reference to traffic detectors associated with segments) refer to traffic detectors that form part of the fixed infrastructure of said navigable network, such as inductive loops, traffic cameras, infrared, radar, photoelectric sensors, or any type of sensor. Such traffic detectors that form part of the fixed infrastructure are distinct from floating vehicle or probe data, in which vehicles themselves are used as sensors as they travel through said network of navigable elements.

[0167] Using the determined probe count data and a scaling factor to obtain the estimated traffic volume for the segment according to any aspect or embodiment of the present invention may include multiplying the count by the scaling factor, in accordance with Equation 1 above.

[0168] The invention extends to a system for carrying out the steps of the method according to any of the aspects of the invention.

[0169] According to a further aspect of the present invention, there is provided a system for generating traffic data indicative of traffic volume in a navigable network in an area covered by an electronic map, the electronic map including a plurality of segments representing navigable elements of the navigable network, the system comprising, for one or more segments of the electronic map: obtaining data indicative of a count of devices associated with a vehicle moving along the navigable element represented by the segment with respect to a given time, the count of devices being based on position data and associated timing data relating to movement of a plurality of devices associated with a vehicle along the navigable element represented by the segment; and using the determined count data and a scaling factor to obtain data indicative of an estimated traffic volume for the segment for the given time; a set of one or more processors for executing a method comprising the steps of: A system is provided, wherein the scaling factor is a time-dependent scaling factor, and the method includes using the scaling factor for the given time when obtaining the estimated traffic volume for the segment.

[0170] The invention in this aspect may include any or all of the features described in connection with other aspects of the invention, and vice versa, to the extent not mutually inconsistent. Thus, even if not explicitly stated herein, a system of the invention may comprise means, or a set of one or more processors, or circuitry for performing any of the steps of the methods or inventions described herein.

[0171] According to a further aspect of the present invention, there is provided a system for generating traffic data indicative of traffic volume in a navigable network in an area covered by an electronic map, said electronic map including a plurality of segments representing navigable elements of said navigable network, said system comprising: receiving data indicative of a segment of interest, the segment for which estimated traffic volume data is desired, and data indicative of a time of interest; obtaining, for each of a set of a plurality of the segments of the electronic map, data indicative of a count of devices associated with vehicles moving through the navigable element represented by the segment for a given time selected based on the time of interest, the count of devices being based on position data and associated timing data relating to movement of a plurality of devices associated with vehicles along the navigable element represented by the segment; identifying a reference subset of the set of segments of the electronic map, each segment of the reference subset of segments being associated with at least one traffic detector; obtaining, for each of the reference subset of the set of segments of the electronic map, data indicative of a measured count of vehicles moving through the navigable element represented by the segment for the given time based on data obtained from the at least one traffic detector associated therewith; obtaining a reference scaling factor for each of the reference subset of segments, each reference scaling factor being a scaling factor determined for the given time based on the measured count of vehicles traveling on the segment for the given time, determined based on data measured by the at least one traffic detector associated with the segment, and a count of devices associated with vehicles traveling on the segment for the given time based on position data and associated timing data regarding movement of the devices along the segment; determining an estimated scaling factor for the segment of interest for the given time based on the or each reference scaling factor obtained; using the estimated scaling factor for the given time based on the position data and associated timing data regarding the movement of a plurality of devices with respect to time along the navigable element represented by the segment of interest in obtaining the estimated traffic volume for the segment, and the data indicative of the count of devices associated with vehicles moving along the navigable element represented by the segment of interest with respect to the given time; A system is provided comprising a set of one or more processors for performing a method including:

[0172] The invention in this aspect may include any or all of the features described in connection with other aspects of the invention, and vice versa, to the extent not mutually inconsistent. Thus, even if not explicitly stated herein, a system of the invention may comprise means, or a set of one or more processors, or circuitry for performing any of the steps of the methods or inventions described herein.

[0173] According to yet another aspect of the present invention, there is provided a system for estimating traffic volume for a given time period for a given segment of an electronic map representing a navigable network in an area, the electronic map including a plurality of segments representing navigable elements of the navigable network in the area, the navigable network in the area including navigable sections associated with at least one traffic detector and navigable sections not associated with any traffic detector, the given segment representing at least a portion of a navigable section of the navigable network in the area not associated with any traffic detector, the electronic map further including a plurality of reference segments, each reference segment representing a navigable section of the navigable network in the area associated with a traffic detector. wherein each reference segment is associated with data indicative of a respective reference scaling factor for the given time, the scaling factor being based on a measured count of vehicles traveling along at least a portion of the section represented by the reference segment for the given time and a count of devices associated with vehicles traveling along at least a portion of the section represented by the reference segment for the given time, the measured count of vehicles being based on data measured by the at least one traffic detector associated with the section, and the count of devices associated with vehicles being based on position data and associated timing data regarding movement of a plurality of devices along at least a portion of the section represented by the reference segment, estimating the traffic volume for the given segment for the given time using data indicative of a count of devices associated with vehicles traveling along at least a portion of the section of the navigable network represented by the given segment for the given time and an estimated scaling factor for the given time for the given segment, wherein the estimated scaling factor for the segment is based on the reference scaling factor associated with each of a subset of one or more reference segments of the electronic map associated with the given segment, and the count of devices is based on position data and associated timing data regarding movement of a plurality of devices along at least a portion of the navigable network section represented by the given segment for the given time; and generating data indicative of the estimated traffic volume for the given segment for the given time; A system is provided, comprising a set of one or more processors for executing a method including the steps of:

[0174] The invention in this aspect may include any or all of the features described in connection with other aspects of the invention, and vice versa, to the extent not mutually inconsistent. Thus, even if not explicitly stated herein, a system of the invention may comprise means, or a set of one or more processors, or circuitry for performing any of the steps of the methods or inventions described herein.

[0175] According to yet another aspect of the present invention, there is provided a system for generating traffic data indicative of traffic volume in a navigable network in an area covered by an electronic map, the electronic map including a plurality of segments representing navigable elements of the navigable network, the system comprising, for one or more segments of the electronic map: obtaining data indicative of a count of devices associated with a vehicle moving along the navigable element represented by the segment with respect to a given time, the count of devices being based on position data and associated timing data relating to movement of a plurality of devices associated with a vehicle along the navigable element represented by the segment; using the determined count data and a scaling factor to obtain data indicative of an estimated traffic volume for the segment for the given time; a set of one or more processors for executing a method comprising: the scaling factor is a location-dependent scaling factor, and the method includes using the scaling factor for a location associated with the segment when obtaining the estimated traffic volume for the segment. A system is provided.

[0176] The invention in this aspect may include any or all of the features described in connection with other aspects of the invention, and vice versa, to the extent not mutually inconsistent. Thus, even if not explicitly stated herein, a system of the invention may comprise means, or a set of one or more processors, or circuitry for performing any of the steps of the methods or inventions described herein.

[0177] In preferred embodiments, the system of any of these further aspects of the invention comprises one or more servers. The method may be performed by a server operating on suitable sources of measured traffic data and vehicle probe data, for example by filtering to obtain data for applicable times. However, arrangements are envisaged where the steps may be performed by one or more servers and / or a distributed system which may include one or more computing devices of any type, for example navigation devices.

[0178] The various functions described herein may be performed in any desired and suitable manner. For example, the present invention may generally be implemented in hardware or software, as desired. Thus, for example, unless otherwise indicated, the various functional elements, stages, units, and "means" of the techniques described herein may comprise one or more processors, one or more controllers, functional units, circuits, processing logic, microprocessor configurations, etc., operable to perform various functions, etc., as appropriate, including dedicated hardware elements (processing circuitry) and / or programmable hardware elements (processing circuitry) that can be programmed to operate in a desired manner.

[0179] The means (processing circuitry) for performing any of the steps of the method may include a set of one or more processors configured, e.g., programmed, to do so. A given step may be performed using the same or a different set of processors as any other step. Any given step may be performed using a combination of sets of processors. The system may further comprise data storage means, such as a computer memory, for storing the generated traffic volume and / or traffic density data. The system may further comprise display means, such as a computer display, for displaying the generated traffic volume and / or traffic density data.

[0180] The methods described herein are computer-implemented methods.

[0181] The methods of the present invention are, in preferred embodiments, performed by a server. Thus, embodiments include a server that includes means (processing circuitry) for performing the various steps described, and the steps of the methods described herein are performed by the server.

[0182] The navigable network may include a road network, with each navigable element representing a road or a portion of a road. For example, a navigable element may represent a road between two adjacent intersections of the road network, or a navigable element may represent a portion of a road between two adjacent intersections of the road network. However, it will be understood that the navigable network is not limited to road networks and may include, for example, networks of walking paths, bicycle paths, rivers, etc. Note that the term "segment" as used herein takes its ordinary meaning in the art. A segment of an electronic map is a navigable link connecting two points or nodes. While embodiments of the present invention are described with particular reference to road segments, it will be understood that the present invention is also applicable to other navigable segments, such as segments of paths, rivers, canals, bicycle paths, towpaths, railroad lines, etc. Accordingly, any reference to a "road segment" may be replaced with a reference to a "navigable segment" or any particular type(s) of such segment.

[0183] The network is represented by electronic map data, which may be stored by or accessible to the server in embodiments in which the method is implemented using a server. In its simplest form, an electronic map (or mathematical graph, as it is sometimes known) is effectively a database containing data representing nodes, most commonly representing road intersections, and lines between the nodes, representing the roads between those intersections. In more detailed digital maps, the lines may be divided into segments defined by start and end nodes. These nodes may be "real," in that they represent road intersections where a minimum of three lines or segments intersect, or they may be "artificial," in that they serve as anchors for segments not defined at one or both ends by real nodes, particularly to provide shape information about a particular section of road or a means of identifying some characteristic of that road, such as a location along the road where a speed limit changes. In virtually all modern digital maps, nodes and segments are further defined by various attributes, which are again represented by data in the database. For example, each node typically has geographic coordinates defining its real-world location, e.g., latitude and longitude. A node also typically has maneuvering data associated with it indicating whether it is possible to move from one road to another at an intersection, while the segments also have associated attributes such as maximum allowed speed, lane size, number of lanes, whether there is a median between them, etc.

[0184] In various aspects and embodiments, the present invention includes obtaining and / or using position data and associated timing data relating to the movement of multiple devices along navigable elements of the navigable network represented by the electronic map data. The position data may provide data indicative of the movement of multiple devices along the navigable element over time. The position data used in accordance with the present invention is position data relating to the movement of multiple devices along the or each navigable element. The method may include obtaining position data and associated timing data relating to the movement of multiple devices within the navigable network and filtering the position data to obtain position data and associated timing data relating to the movement of multiple devices along the or each given navigable element. The step of obtaining the position data relating to the movement of devices along the or each navigable element may be performed by referencing the electronic map data representing the navigable network. The method may include matching position data relating to the movement of devices within a geographic region including the navigable network to at least one of the or each navigable element considered in accordance with the present invention.

[0185] In some configurations, the step of obtaining the location data may include accessing the data, i.e., the data previously received and stored. Preferably, however, the method may include receiving the location data from the device. In embodiments in which the step of obtaining data includes receiving the data from the device, it is envisioned that the method may further include storing the received location data and, optionally, filtering the data before proceeding to perform other steps of the invention. The step of receiving the location data need not occur simultaneously with or at the same location as one or more other steps of the method.

[0186] The location data used in accordance with the present invention is collected from one or more, preferably multiple, devices and relates to the movement of the devices over time. Thus, the devices are mobile devices. It will be understood that at least some of the location data is associated with temporal data, e.g., a timestamp. However, for purposes of the present invention, it is not necessary that all location data be associated with temporal data, as long as it may be used to provide information about the device's movement along a navigable segment in accordance with the present invention. However, in a preferred embodiment, all location data is associated with temporal data, e.g., a timestamp. It will be understood that timing data may be associated with a "trace" comprising a set of location data "fixes" acquired by a device, rather than directly associated with each individual location data fix. For example, each location data "fix" may be associated with an offset relative to the time associated with the trace.

[0187] The location data and associated timing data relate to the movement of the device and can be used to provide a location "trace" of the path taken by the device. As described above, the data may be received from the device(s) or may be initially stored. The device may be any mobile device capable of providing the location data and sufficient associated timing data for purposes of the present invention. The device may also be any device with positioning capabilities. For example, the device may include means for accessing and receiving information from a cellular communications network, such as a WiFi access point or GSM device, and using this information to determine its location. However, in a preferred embodiment, the device preferably includes a Global Navigation Satellite System (GNSS) receiver, such as a GPS receiver, for receiving satellite signals indicative of the receiver's position at a particular time, and receives updated location information at regular intervals. Such devices may include navigation devices, mobile telecommunications devices with positioning capabilities, location sensors, etc.

[0188] The device is associated with a vehicle. In these embodiments, the location of the device corresponds to the location of the vehicle. References to location data obtained from a device associated with a vehicle can be replaced with references to location data obtained from the vehicle, and references to the movement of one or more devices can be replaced with references to the movement of the vehicle, unless explicitly stated otherwise, and vice versa. The device may be integrated with the vehicle or may be a separate device associated with the vehicle, such as a portable navigation device. Of course, the location data may be obtained from a combination of different devices or a single type of device.

[0189] The location data obtained from the plurality of devices is commonly known as "probe data." Data obtained from devices associated with a vehicle may be referred to as vehicle probe data (or sometimes floating car data). Accordingly, references to "probe data" herein should be understood to be interchangeable with the term "location data," and the location data may be referred to herein as probe data for brevity.

[0190] The sample volume, together with the selected average penetration rate, may be used to estimate either traffic volume or traffic density (or both) for a segment (or preferably multiple segments) within the region. Both traffic volume and traffic density are important parameters for characterizing traffic conditions within the network and can be used for various traffic planning and control applications.

[0191] The present invention enables traffic volume estimation for any segment in a network for which an appropriate scaling factor can be obtained according to the methods described herein. That is, as long as there is sufficient probe data, it is possible to reliably estimate the scaling factor, and therefore traffic volume, over a relatively wide area of ​​the network at a lower cost than is typically possible with conventional methods. For example, preferably, scaling factors, and therefore traffic volume, can be determined for multiple (or all) segments in a region so that a region-wide traffic volume picture can be provided. Thus, the methods described herein can be repeated for one or more additional segments representing navigable elements of the navigable network. For example, the methods may be performed at least in association with each of a subset of segments for which traffic detector data is unavailable, i.e., each of a subset of segments not associated with at least one traffic detector.

[0192] The method includes obtaining data indicative of estimated traffic volume for the segment for the given time interval. The method may include generating data indicative of the estimated traffic volume for output. The method may include associating the data indicative of the estimated traffic volume with data indicative of the segment. Thus, the traffic volume data may be associated with the electronic map data. In a preferred embodiment, the present invention includes transmitting and / or storing and / or displaying the traffic volume data to a user. That is, the traffic volume data may be provided as an output to a user. When the method described herein is performed by a server, the method may include the server transmitting the data indicative of the estimated traffic volume for the segment to a device associated with a user and / or a vehicle, such as a navigation device. As mentioned above, a navigation device refers to a device that runs a navigation application.

[0193] It will be appreciated that the methods according to the invention can be implemented at least in part using software. Viewed from further aspects and embodiments, the invention will therefore be understood to extend to a computer program product comprising computer-readable instructions configured to perform any or all of the methods described herein when executed on suitable data processing means. The invention also extends to a computer software carrier comprising such software. Such a software carrier may be a physical (or non-transitory) storage medium, or may be a signal such as an electrical signal over a wire, an optical signal, or a radio signal to a satellite or the like.

[0194] Any reference to comparing one item to another item can include comparing either item to the other item in any way.

[0195] Note that the phrase "associated with" with respect to one or more segments or elements should not be construed as requiring any particular restriction on data storage location. The phrase requires only that the feature be identifiably related to the element. Thus, association may be achieved, for example, by reference to a side file potentially located on a remote server.

[0196] Unless expressly stated otherwise, it will be understood that the invention, in any of its aspects, may include any or all of the features described with respect to other aspects or embodiments of the invention, to the extent that they are not mutually exclusive. In particular, although various embodiments of operations that may be performed in the method and by the system or device have been described, it will be understood that any one or more or all of these operations may be performed in the method and by the system or device in any combination, as desired and required.

[0197] Advantages of these embodiments are set forth below, and further details and features of each of these embodiments are defined in the accompanying dependent claims and elsewhere in the following detailed description. [Brief explanation of the drawings]

[0198] Embodiments of the present invention will now be described, by way of example only, with reference to the accompanying drawings, in which: [Figure 1] 1 is a schematic illustration of an exemplary portion of a Global Positioning System (GPS) usable by a navigation device. [Figure 2] 1 is a schematic diagram of a communication system for communication between a navigation device and a server. [Figure 3] 3 is a schematic illustration of the electronic components of the navigation device of FIG. 2 or any other suitable navigation device. [Figure 4] 1 is a schematic diagram of an arrangement for mounting and / or docking a navigation device. [Figure 5] 4 is a schematic representation of an architectural stack employed by the navigation device of FIG. 3; [Figure 6] 1 illustrates the various shapes that a navigation device can take. [Figure 7] 1 illustrates various devices that may be associated with a vehicle. [Figure 8] 1 illustrates another exemplary navigation system. [Figure 9] 1 indicates the number of road segments with traffic movement detectors in a given map area. [Figure 10] A constant scaling factor is used to represent the observed error between the estimated and measured traffic flow. [Figure 11] The error observed when using a scaling factor that is time-dependent but location-independent is shown. [Figure 12] 1 shows an exemplary probe profile illustrating the weekly pattern of probe data. [Figure 13] It shows the degree of similarity between a probe profile (which may be related to the segment of interest) and four (reference) probe profiles. [Figure 14] FIG. 1 is a functional diagram illustrating one system for implementing generation of a set of scaling coefficients according to embodiments described herein. [Figure 15] 15 is a flowchart illustrating one method for obtaining time-dependent scaling factors using the system of FIG. 14. [Figure 16] 10 illustrates the errors observed when using scaling factors obtained in accordance with certain preferred embodiments of the present invention. [Figure 17] 10 shows an example of an attenuation function that may be used in obtaining the position-dependent scaling factor. [Figure 18] 10 shows a map area showing values ​​of the location dependent scaling factors. DETAILED DESCRIPTION OF THE INVENTION

[0199] 1-5, a system that may be used to facilitate understanding of the context of the present invention will now be described. Embodiments will now be described with particular reference to portable navigation devices (PNDs). However, it should be remembered that the teachings of the present invention are not limited to PNDs, but are instead generally applicable to devices capable of transmitting probe data samples to a server, including, but not limited to, any type of processing device configured to execute navigation software in a portable manner to provide route planning and navigation functionality. Some exemplary such devices are described below with reference to FIGS. 6 and 7. Accordingly, in the context of the present application, a navigation device is intended to include (without limitation) any type of route planning and navigation device, regardless of whether the device is embodied as a PND, and would include a device integrated into a vehicle, such as an automobile, or indeed a portable computing resource that executes route planning and navigation software, such as a portable personal computer (PC), mobile phone, or personal digital assistant (PDA). The present invention is also applicable to devices that may transmit probe data samples, which may not necessarily be configured to execute navigation software, but which are configured to transmit probe data samples and perform other functions described herein.

[0200] Furthermore, embodiments of the present invention are described with reference to road segments. It should be understood that the present invention may also be applicable to other navigable segments, such as segments of roads, rivers, canals, bike paths, towpaths, railroad tracks, or the like. For ease of reference, these will generally be referred to as road segments.

[0201] It will become clear from below that when route planning is performed, this can also occur in situations where the user does not want instructions on how to navigate from one point to another, but simply wants to be provided with a view of a given location. In such situations, the "destination" location selected by the user does not necessarily have a corresponding start location from which the user wants to begin navigation, and consequently, references herein to a "destination" location, or indeed a "destination" view, should not be interpreted as implying that generation of a route is mandatory, that travel to the "destination" must occur, or indeed that the existence of a destination necessitates the specification of a corresponding start location.

[0202] With the above provisos in mind, global positioning systems (GPS), such as that of FIG. 1, are used for a variety of purposes. Generally, GPS is a satellite radio-based navigation system capable of determining continuous position, velocity, time, and in some instances, direction information for an unlimited number of users. GPS, formerly known as NAVSTAR, incorporates multiple satellites that orbit the Earth in highly precise orbits. Based on these precise orbits, the GPS satellites can relay their positions, as GPS data, to any number of receiving units. However, it will be understood that other global positioning systems, such as GLOSNASS, the European Galileo Positioning System, the COMPASS positioning system, or IRNSS (Indian Regional Navigation Satellite System), may also be used.

[0203] A GPS system is implemented when a device specially equipped to receive GPS data begins scanning radio frequencies for GPS satellite signals. Upon receiving a radio signal from a GPS satellite, the device determines the satellite's precise location through one of several different conventional methods. In most instances, the device continues scanning for signals until it acquires at least three distinct satellite signals (note that, although not common, location can be determined with only two signals using other triangulation techniques). Performing geometric triangulation, the receiver uses the three known positions to determine its own two-dimensional position relative to the satellites. This can be done in a known manner. Additionally, by acquiring a fourth satellite signal, the receiving device can calculate its own three-dimensional position through the same geometric operations, in a known manner. Position and velocity data can be continuously updated in real time by an unlimited number of users.

[0204] As shown in FIG. 1, a GPS system 100 includes a plurality of satellites 102 orbiting the Earth 104. A GPS receiver 106 receives GPS data from multiple satellites 102 as spread spectrum GPS satellite data signals 108. The spread spectrum data signals 108 are transmitted continuously from each satellite 102, and each transmitted spread spectrum data signal 108 includes a data stream that includes information identifying the particular satellite 102 from which the data stream originates. A GPS receiver 106 typically requires spread spectrum data signals 108 from at least three satellites 102 to be able to calculate a two-dimensional position. Receipt of a fourth spread spectrum data signal allows the GPS receiver 106 to calculate a three-dimensional position using known techniques.

[0205] 2 , a navigation device 200 (e.g., a PND) including or coupled to a GPS receiver device 106 may optionally establish a data session with the network hardware of a “mobile” or telecommunications network via a mobile device (not shown), e.g., a mobile phone, PDA, and / or any device with mobile phone technology, to establish a digital connection, e.g., via known Bluetooth® technology. Thereafter, through its network service provider, the mobile device can establish a network connection (e.g., over the Internet) with server 150. In this way, a “mobile” network connection may be established between navigation device 200 (which may be, and often is, mobile due to movement alone and / or in a vehicle) and server 150 to provide a “real-time” or at least very “up-to-date” gateway for information.

[0206] Establishment of a network connection between a mobile device and another device, such as server 150 (via a service provider), for example, using the Internet, may be performed in a known manner. In this regard, any number of suitable data communications protocols may be employed, for example, the TCP / IP layered protocol. Furthermore, the mobile device may utilize any number of communications standards, such as CDMA2000, GSM, IEEE 802.11a / b / c / g / n, etc.

[0207] It will therefore be appreciated that internet connectivity may be utilized, which may be achieved via a data connection, for example via a mobile phone or mobile phone technology within the navigation device 200 .

[0208] Although not shown, the navigation device 200 may of course include its own mobile phone technology within the navigation device 200 itself (e.g. including an antenna or optionally using an internal antenna of the navigation device 200). The mobile phone technology within the navigation device 200 may include internal components and / or may include, for example, an insertable card (e.g. a Subscriber Identity Module (SIM) card) with the necessary mobile phone technology and / or antenna. In this manner, the mobile phone technology within the navigation device 200 may similarly establish a network connection between the navigation device 200 and the server 150, for example via the Internet, in the same manner as any mobile device.

[0209] For phone settings, a Bluetooth enabled navigation device may be used to operate correctly with the ever-changing spectrum of mobile phone models, manufacturers, etc., and model / manufacturer specific settings may be stored, for example, in the navigation device 200. The stored data for this information may be updated.

[0210] In Figure 2, the navigation device 200 is shown communicating with the server 150 via a general purpose communication channel 152, which may be implemented by any of several different devices. The communication channel 152 collectively represents the propagation medium or path connecting the navigation device 200 and the server 150. The server 150 and the navigation device 200 can communicate when a connection via the communication channel 152 is established between the server 150 and the navigation device 200 (note that such a connection may be a data connection via a mobile device, a direct connection via a personal computer over the Internet, etc.).

[0211] The communication channel 152 is not limited to a particular communication technology. Furthermore, the communication channel 152 is not limited to a single communication technology. That is, the channel 152 may include several communication links using various technologies. For example, the communication channel 152 may be adapted to provide a path for electrical, optical, and / or electromagnetic communication, etc. As such, the communication channel 152 may include, but is not limited to, one or a combination of electrical circuits, electrical conductors such as wires and coaxial cables, fiber optic cables, converters, radio frequency (RF) waves, the atmosphere, free space, etc. Furthermore, the communication channel 152 may include intermediate devices such as, for example, routers, repeaters, buffers, transmitters, and receivers.

[0212] In one exemplary configuration, communication channel 152 includes telephone and computer networks. Additionally, communication channel 152 may be capable of accommodating wireless communications, e.g., infrared communications, radio frequency communications such as microwave frequency communications, etc. Additionally, communication channel 152 may accommodate satellite communications.

[0213] Communication signals transmitted over communication channel 152 include, but are not limited to, signals that may be necessary or desired for a given communication technology. For example, the signals may be adapted for use in cellular communication technologies such as time division multiple access (TDMA), frequency division multiple access (FDMA), code division multiple access (CDMA), Global System for Mobile Communications (GSM), General Packet Radio Service (GPRS), etc. Both digital and analog signals may be transmitted over communication channel 152. These signals may be modulated, encrypted, and / or compressed as may be desirable for the communication technology.

[0214] Server 150 includes, in addition to other components that may not be shown, a processor 154 operably connected to memory 156 and further operably connected to a mass data storage device 160 via a wired or wireless connection 158. Mass storage device 160 includes storage of navigation data and map information and may again be a separate device from server 150 or may be incorporated into server 150. Processor 154 is further operably connected to a transmitter 162 and a receiver 164 for transmitting information to or receiving information from navigation device 200 over communication channel 152. The transmitted and received signals may include data, communication, and / or other propagated signals. Transmitter 162 and receiver 164 may be selected or designed according to the communication requirements and communication technology used in the communication design of navigation system 200. It should be further noted that the functions of transmitter 162 and receiver 164 may be combined into a single transceiver.

[0215] As mentioned above, the navigation device 200 may be configured to communicate with the server 150 over the communication channel 152 using the transmitter 166 and the receiver 168 to send and receive signals and / or data over the communication channel 152, and it should be noted that these devices may also be used to communicate with devices other than the server 150. Furthermore, the transmitter 166 and the receiver 168 are selected or designed according to the communication requirements and communication technologies used in the communication design for the navigation device 200, and the functionality of the transmitter 166 and the receiver 168 may be combined into a single transceiver as described above in connection with Figure 2. Of course, the navigation device 200 may comprise other hardware and / or functional parts, which will be described in more detail later herein.

[0216] Software stored in server memory 156 provides instructions to processor 154, enabling server 150 to provide services to navigation device 200. One service provided by server 150 includes processing requests from navigation device 200 and transmitting navigation data from mass data storage 160 to navigation device 200. Another service that may be provided by server 150 includes processing navigation data using various algorithms for a desired application and sending the results of these operations to navigation device 200.

[0217] The server 150 constitutes a remote source of data accessible by the navigation device 200 over a wireless channel. The server 150 may include a network server located on a local area network (LAN), a wide area network (WAN), a virtual private network (VPN), etc.

[0218] The server 150 may comprise a personal computer, such as a desktop or laptop computer, and the communication channel 152 may be a cable connected between the personal computer and the navigation device 200. Alternatively, a personal computer may be connected between the navigation device 200 and the server 150 to establish an internet connection between the server 150 and the navigation device 200.

[0219] The navigation device 200 may be provided with information from the server 150 via information downloads, which may be updated automatically, from time to time, or when a user connects the navigation device 200 to the server 150, and / or may be more dynamic when there is a more constant or frequent connection between the server 150 and the navigation device 200, for example via a wireless mobile connection device and a TCP / IP connection. For many dynamic calculations, the processor 154 in the server 150 may be used to handle most of the processing needs, although the processor of the navigation device 200 (not shown in FIG. 2) may also handle more processing and calculations, often independently of a connection to the server 150.

[0220] With reference to FIG. 3 , it should be noted that the block diagram of the navigation device 200 is not inclusive of all components of the navigation device, but merely represents many exemplary components. The navigation device 200 is disposed within a housing (not shown). The navigation device 200 includes processing circuitry including, for example, the processor 202 described above, which is coupled to an input device 204 and a display device, such as a display screen 206. While the input device 204 is referred to in the singular herein, those skilled in the art will understand that the input device 204 represents any number of input devices, including a keyboard device, a voice input device, a touch panel, and / or any other known input device utilized to input information. Similarly, the display screen 206 may include any type of display screen, such as, for example, a liquid crystal display (LCD).

[0221] In one configuration, the input device 204, touch panel, and one aspect of the display screen 206 are integrated to provide an integrated input / output device including a touchpad or touchscreen input 250 (FIG. 4), allowing both the input of information (via direct entry, menu selection, etc.) and the display of information through the touch panel screen, such that a user need only touch a portion of the display screen 206 to select one of multiple display options or to activate one of multiple virtual or "soft" buttons. In this regard, the processor 202 supports a graphical user interface (GUI) that operates in conjunction with the touch screen.

[0222] In the navigation device 200, the processor 202 is operatively connected to an input device 204 via a connection 210 to receive input information, and is operatively connected to and outputs information to at least one of a display screen 206 and an output device 208 via a separate output connection 212. The navigation device 200 may also include an output device 208, such as an audible output device (e.g., a loudspeaker). It should also be understood that the input device 204 may also include a microphone and software for receiving input voice commands, as the output device 208 may generate audible information for a user of the navigation device 200. Furthermore, the navigation device 200 may also include any additional input device 204 and / or any additional output device, such as an audio input / output device.

[0223] The processor 202 is operatively connected to memory 214 via connection 216 and is further configured to receive / transmit information from / to an input / output (I / O) port 218 via connection 220, which is connectable to an I / O device 222 external to the navigation device 200. The external I / O device 222 may include, but is not limited to, an external listening device such as, for example, an earphone. The connection to the I / O device 222 may further be a wired or wireless connection to any other external device such as a car stereo unit for hands-free operation and / or for connection to, for example, an earphone or headphones, and / or for voice-activated operation, for example, for connection to a mobile phone, which may for example be used to establish a data connection between the navigation device 200 and the Internet or any other network, and / or for establishing a connection to a server via, for example, the Internet or some other network.

[0224] The memory 214 of the navigation device 200 includes a portion of non-volatile memory (e.g., for storing program code) and a portion of volatile memory (e.g., for storing data when the program code is executed). The navigation device also includes a port 228 in communication with the processor 202 via a connection 230 to allow a removable memory card (commonly referred to as a card) to be added to the device 200. In the described embodiment, the port is configured to allow for the addition of an SD (Secure Digital) card. In other embodiments, the port may allow for the connection of other types of memory, such as a CompactFlash (CF) card, a Memory Stick, an xD memory card, a USB (Universal Serial Bus) flash drive, an MMC (Multimedia) card, a SmartMedia card, a Microdrive, etc.

[0225] Figure 3 further illustrates an operable connection between processor 202 and antenna / receiver 224 via connection 226, which may be, for example, a GPS antenna / receiver, and thus function as GPS receiver 106 of Figure 1. While the antenna and receiver indicated by reference numeral 224 are schematically combined for purposes of illustration, it should be understood that the antenna and receiver may be separately located components, and the antenna may be, for example, a GPS patch antenna or a helical antenna.

[0226] Of course, those skilled in the art will understand that the electronic components shown in FIG. 3 are powered by one or more power sources (not shown) in a conventional manner. Such power sources may include an internal battery and / or an input for a low-voltage DC power source or any other suitable configuration. As will be understood by those skilled in the art, different configurations of the components shown in FIG. 3 are contemplated. For example, the components shown in FIG. 3 may communicate with each other via wired and / or wireless connections, etc. Thus, the navigation device 200 described herein may be a portable or handheld navigation device 200.

[0227] 3 may be connected or "docked" in a known manner to a vehicle, such as a bicycle, motorcycle, automobile, or boat. Such a navigation device 200 may then be removed from the docked location for portable or handheld navigation use. Indeed, in other embodiments, the device 200 may be configured to be handheld to enable user navigation.

[0228] Referring to FIG. 4, the navigation device 200 may be a unit that includes an integrated input and display device 206 and the other components of FIG. 2 (including, but not limited to, an internal GPS receiver 224, a processor 202, a power supply (not shown), a memory system 214, etc.).

[0229] The navigation device 200 may sit on an arm 252, which may itself be secured to a vehicle dashboard, window, etc. using a suction cup 254. The arm 252 is an example of a docking station to which the navigation device 200 may be docked. The navigation device 200 may be docked or otherwise connected to the arm 252 of the docking station, for example, by snap-connecting the navigation device 200 to the arm 252. The navigation device 200 may then be rotatable on the arm 252. To disconnect the navigation device 200 from the docking station, for example, a button (not shown) on the navigation device 200 may be pressed. Other equally suitable configurations for coupling and uncoupling the navigation device 200 to a docking station will be known to those skilled in the art.

[0230] Of course, the navigation device need not be provided by a PND-type device as described: as will be described below, a wide range of general computing devices, when running a navigation client, may provide the functionality described with reference to the navigation device 200 and may communicate with the server in the same manner.

[0231] 5, the processor 202 and memory 214 cooperate to support a BIOS (basic input / output system) 282, which serves as an interface between the functional hardware components 280 of the navigation device 200 and the software executed by the device. The processor 202 then loads an operating system 284 from the memory 214, which provides an environment in which application software 286 (which implements some or all of the described route planning and navigation functions) can run. The application software 286 provides an operating environment, including a graphical user interface, that supports the core functionality of the navigation device, such as map viewing, route planning, navigation functions, and any other functionality related thereto. In this regard, part of the application software 286 includes a view generation module 288.

[0232] In the described embodiment, the navigation device's processor 202 is programmed to receive GPS data received by the antenna 224 and, when triggered according to the methods described herein, store the GPS data in memory 214 along with a timestamp of when the GPS data was received to build a record of the navigation device's position. Each data record so stored may be considered a GPS fix, i.e., a fix of the navigation device's position, and includes latitude, longitude, and a timestamp. Such data is referred to herein as a probe data sample.

[0233] Furthermore, the processor 202 is configured to upload each probe data sample (i.e., GPS data and timestamp) to the server 150. The navigation device 200 may have a permanent, or at least commonly present, communication channel 152 connecting it to the server 150.

[0234] In the described embodiment, the probe data samples provide one or more traces, each trace representing the movement of a given navigation device 200 within an applicable time period, e.g., while traversing a given route. The server 150 is configured to receive the received probe data samples and store them as records of device locations in the mass data storage 160 for processing. Thus, over time, the mass data storage 160 accumulates multiple records of the location of the navigation device 200 that uploaded the probe data samples. The server may reconstruct the probe data samples that form a trace by associating common elements, such as a device identifier value or the time period to which the data pertains, with the entire trace rather than with each of the probe data samples that make up the trace. After moving the common elements down to the trace level, the individual probe data samples within the trace may include at least a position value and a time offset (e.g., a time from the start of the time period or a sequence number) within the time period to which the trace pertains.

[0235] As mentioned above, mass data storage 160 also includes map data, which provides information regarding the locations of road segments, points of interest, and other such information commonly found on a map.

[0236] As noted above, the term "navigation device" as used herein should be understood to encompass any form of device that executes a suitable navigation client, and is not limited to use with special-purpose PND-type devices such as those illustrated in FIG. 4. A navigation client is a software application that executes on a computing device. A navigation device may be implemented using a wide range of computing devices. Some exemplary such devices are shown in FIG. 6.

[0237] All of the devices in Figure 6 include a navigation screen to assist a user in navigating to a desired destination. These include a personal navigation device (PND), which is a single-purpose computing device (top left), a general-purpose computing device in the form of a mobile phone (top right), a laptop (bottom left), and an in-vehicle integrated computing device (bottom right). Of course, these are only a few examples of a wide range of common computing devices that may be used to run a navigation client. For example, a tablet or a wearable device such as a watch may also be used.

[0238] A vehicle may have multiple computing devices and multiple displays to assist the driver, as shown in Figure 7. Figure 7 shows the interior of a car with a steering wheel 300, a first display area 320 behind the steering wheel, a head-up display 330 projected onto the windshield, a center display 340, and multiple controls (buttons, touchscreen) 350. Additionally, the car may support the adoption of mobile terminals into the car's computing environment.

[0239] A functional diagram of another exemplary navigation system 400 is shown in Figure 8. The system includes a navigation client 402, which may be provided by a software application executing on any suitable computing device, such as those illustrated with reference to Figures 6 and 7. The system 400 also includes a map server 404, a traffic information server 406, and an in-vehicle control system 408. These components are described in more detail below. It will be understood that the navigation system includes multiple navigation clients 402 in communication with the map server 404 and the traffic information server 406.

[0240] Navigation Client 402 The navigation client is provided by a navigation application running on a computing device. The navigation client 402 provides user input / output devices 410, 412, as is common to most computing devices. The navigation client also provides a map data controller 414 that retrieves map data and stores the map data in non-volatile memory of the computing device on which the navigation application running the client runs. The navigation device on which the navigation application runs also includes a position sensor 416, in addition to traditional computing device components such as a processing unit, memory, display, long-term storage (flash memory), and networking interfaces. Such more traditional components are not shown in FIG. 8, which shows components associated with supporting navigation functionality.

[0241] The navigation client 400 operates using an electronic map of a geographic area. The map information may be stored locally on the device (e.g., in non-volatile solid-state memory) or may be retrieved from a navigation server. The navigation client uses the electronic map to generate a map view of the geographic area of ​​interest on the display of the computing device. Typically, the geographic area is centered around the current location of the computing device running the navigation client software application.

[0242] The current location is determined using a position sensor 416, which may use any of a wide range of location sensing techniques, such as satellite positioning (GPS, GNSS, ...), Wi-Fi (radio triangulation), mobile phone tracking, Bluetooth beacons, image analysis (examples of which are described in the applicant's International Patent Applications PCT / EP2016 / 068593, PCT / EP2016 / 068594, PCT / EP2016 / 068595, and PCT / IB2016 / 001198, the entire contents of which are incorporated herein by reference), map matching, dead reckoning, and other location sensing techniques. In the presence of location sensing errors, map matching may be used to adjust the measured location to best match a road segment on a map.

[0243] The navigation client 402 may assist a user in navigating from a current location to a destination location. The destination may be input using a destination selection module 418. A route determination module 420 of the navigation client calculates a route to the selected destination. The route determination module 420 also retrieves current traffic information in addition to the electronic map to determine an estimated travel time or estimated arrival time. The current traffic information describes the current conditions on the road network in the geographic area of ​​the electronic map. This may include current average speed, current traffic density, current road closures, etc. The route determination module 420 may present a preferred route as well as alternative routes to allow the end user to select a preferred route.

[0244] The navigation client's guidance module 422 uses the selected preferred route to guide the end user to the selected destination. This may use a display showing a map and a portion of the route to the destination. Guidance may also take the form of additional graphical indications on the display. Most navigation clients also support voice guidance with turn instructions.

[0245] An active navigation client generates location probes and uses these probes to calculate and update traffic information and provides them to the traffic information server. To perform these functions, the navigation client 402 includes a location probe generator 418 and a probe interface 420.

[0246] The navigation client 402 also includes an HTTPS client for communicating with the map server 404 and the traffic information server 424 .

[0247] Map Server 404 The map server 404 is an infrastructure for storing, managing, and creating large amounts of information for creating electronic maps and using electronic maps for navigation. The map server may be provided by a cloud server system.

[0248] The map server 404 includes a map compiler 430 that receives map data from an appropriate cartographer 432. The cartographer 432 receives map source data from a map data source 434 and converts it into a format suitable for inclusion in an electronic map. For example, the map compiler 430 may sort the map data into individual layers and tiles for the electronic map. The map server 404 also includes a map data service 436 and a map metadata service 438. The combination of the map data service 436 and the map metadata service 438 may together be referred to as a "cloud service." The HTTPS client 424 may retrieve map metadata from the map metadata service 436 and then use the metadata to retrieve map data from the map data service 438, if necessary.

[0249] A typical navigation server provides 10 7 ~10 8It manages map information related to kilometers of road networks. Because the map information needs to be high quality, the server infrastructure processes updates to the map information at an update rate that averages about 1000 updates per second. Furthermore, the map information needs to be distributed to a global infrastructure of navigation clients. Distribution requires a sophisticated content delivery network in addition to cloud computing systems to generate the distributed map information. The navigation server also aggregates, processes, and distributes real-time traffic information.

[0250] Traffic information server 406 The traffic information server 406 includes a traffic information compiler 440 that compiles traffic information using data obtained from a probe data source 442. The probe data source 442 receives data from a probe data service 446, which is in turn configured to receive probe data from navigation clients. The traffic information compiler 440 provides traffic information to a traffic information service 444, which communicates with the HTTPS client 424 to provide the traffic information.

[0251] The traffic information server 406 provides road and traffic information to the navigation client 402 .

[0252] The map information typically includes static traffic information based on historical data. For more dynamic traffic information such as traffic density, parking space availability, accidents, road closures, updated road signs, and points of interest, the traffic information server receives position probe data from navigation clients. The traffic information compiler uses the current position probe data obtained from multiple navigation clients to generate current traffic information.

[0253] Location probe data During normal operation, the navigation client 402 periodically sends location probe data to the traffic information server 406. The location probe data includes information about the navigation client's recent or current location. The location probe data may be combined into a collection of probe data elements, commonly called a trace. The traffic information server 406 uses the trace or probe data to estimate current traffic information. This information includes parameters for road segments, such as current average speed and current traffic density. The traffic information server 406 processes the location probe data to provide real-time traffic information to the navigation client 402, enable better route generation, and improve estimated travel times to destinations.

[0254] The term location probe (or "probe") refers to a data sample containing at least location information that indicates the location of a navigation client, i.e., a device implementing the client. Typically, location data includes longitude and latitude values ​​(both typically with an accuracy of about 10 meters). A probe data sample may also include other data, such as a time value. The time value provides a time associated with the location data and may be received from a positioning system to correspond to the time the location data was generated or to correspond to the transmission time of the probe data sample. A probe data sample may also include a device identifier value (uniquely associated with the end user device and user).

[0255] The term trace refers to a collection of location probes that are associated with the same device, user, and common time period. Trace data can be reconstructed at the server by associating common elements, such as a device identifier value or the time period to which the probe data pertains, with each of the probes that make up the trace. After moving the common elements to the trace level, each individual probe in the trace includes at least a location value and a time offset within the time period (e.g., a time from the start of the time period or a sequence number).

[0256] In preferred embodiments, the present invention is directed to at least a method for generating data indicative of traffic volume within a navigable network. Accurate generation of such traffic data is important for many traffic management and control applications. Accordingly, the present invention provides improved methods for generating such traffic data. In particular, the present invention provides methods for generating such traffic data from probe data. Preferred embodiments will now be described with respect to estimating traffic volume. Such techniques may be performed by a server having access to vehicle probes and measured traffic count data, as described below. For example, the server may be a traffic server operating in a system of the type described with reference to FIG. 8.

[0257] Traffic volume (also called traffic flow) is defined as the number of vehicles ΔN passing through a cross section at a location x within a time interval Δt. That is, traffic volume Y is generally given by the following equation:

number

[0258] The time interval Δt can generally be set or selected as desired depending on the application, e.g., the required time resolution (and accuracy). For example, in the context of dynamic traffic phenomena such as traffic congestion, a typical aggregation time interval Δt may be in the range of about 1 minute to 1 hour. However, in other applications, such as traffic signal calibration and traffic planning, much longer time intervals, e.g., days, weeks, or even months, may need to be considered.

[0259] Traffic volume cannot usually be measured directly from probe data, since only a fraction of all vehicles on the road network report data; that is, probe data only represents an arbitrary sample of the total traffic in the road network. Although the percentage of probes is generally increasing, currently coverage (or "penetration level") is typically only around 10% (e.g., in Germany or the Netherlands), and even less in some areas.

[0260] Thus, embodiments of the present invention provide an improved method by which traffic volume can be estimated using probe data from a sample of floating vehicles. The basic idea underlying this concept is that, given an appropriate scaling factor (which may inversely indicate penetration levels), the observed probe counts for a segment in a given time interval can be projected or extrapolated to give the total traffic volume for the segment in that time interval.

[0261] The present invention relates to improved techniques for estimating such scaling factors. The scaling factors are time-dependent and / or location-dependent. As shown with reference to Figure 9 above, current techniques rely on scaling factors that are constant across all segments of the map and across all time periods considered. However, as Figure 10 shows, this can introduce significant errors.

[0262] Some embodiments of the present invention will now be described with reference to methods implemented by a traffic information server.

[0263] The traffic information server uses vehicle probe data and measured traffic data obtained from traffic detectors associated with road elements represented by segments of an electronic map to provide an improved estimate of the scaling factor of a segment at a given time interval. Such measured traffic data is, for simplicity, referred to as "inductive loop data", but it is understood that the measured traffic data may be obtained from any other type of traffic detector associated with a road element, i.e., forming part of a fixed road infrastructure such as a camera, rather than probe data.

[0264] The traffic information server receives vehicle probe data for a set S of road segments s within a map area A. The set S of road segments includes a first subset L = {s i | 0 ≤ i < N} of N road segments s for the map area A, and each road segment s i in this first set is associated with a traffic flow detector. i

[0265] The remaining road segments of this map area form a second subset M = {s r | 0 ≤ i < R} of R road segments s that have no association with a traffic flow detector. Thus, L = {s r | 0 ≤ i < N}, M = {s i | 0 ≤ i < R}, S = L ∨ M, and |S| = N + R. r

[0266] The traffic information server receives measured traffic flow data Y(s i , t) from the traffic detector of road segment s i . Further, the traffic information server receives probe data X(s i , t) from a navigation device associated with a vehicle moving on road segment s i . The navigation device may be any device that executes a navigation application as described above.

[0267] Since both traffic flow data sources are time - dependent and location (road segment) - dependent, the traffic information server is a function of time t and location s i to determine the scaling coefficient k(s i ,t). The scaling coefficient indicates the penetration level.

Number

[0268] The k(s i ,t) coefficient links the received measured traffic flow data obtained by traffic detectors associated with the segment and the received probe data obtained from navigation devices.

[0269] However, the traffic information server does not receive measured traffic flow data for a second set of road segments s r . Instead, the traffic information server only receives probe data X(s r ,t) from navigation devices associated with vehicles moving on road segment s r .

[0270] In Equation 4, time t represents the time in fixed time units of Δt. This means that a high - resolution time T (for example, counting microseconds from a reference time) may be converted to t as t = trunc(T / ΔT). Thus, t counts the number of time intervals Δt from the reference time. Using 1 hour as Δt, time t is a time (hour) indicator with an accuracy of 1 hour. Since traffic patterns are very similar at the same time and on weekdays, a common simplification is to replace time with a discrete time index t k covering 24×7 hours in a week. The time index t k counts Z time intervals ΔT in a week. The formula t k = t mod Z uses modulo arithmetic to convert time t to a time index t in the range 0,..,Z - 1 (0≦k<Z)k If Δt is 10 minutes, the value of Z is 24*7*(60 / 10) = 1008. If Δt is 1 hour (60 minutes), Z is equal to 168 (24 x 7).

[0271] What is needed is road segments for which there is no measured traffic information. r k(s r ,t k ) is a method for determining

[0272] In a simple embodiment, through a comparison of traffic count data based on probe data and measured inductive loop data, the time-dependent values ​​of the scaling factors determined for the segments associated with the inductive loops can be used to infer time-dependent scaling factor values ​​for use in determining traffic volume for segments for which no measured traffic information exists, which provides some improvement in the accuracy with which traffic volume for such segments can be determined compared to using a constant scaling factor across all segments.

[0273] In experiments with two sets of inductive loops, k(t k ) coefficients were calculated from the first set of inductive loops. These k(t k ) coefficients are discrete time dependent but location independent, i.e., used for all road segments. Then, measured traffic data for a second set of inductive loops is calculated using the probe data and k(t k ) coefficient is used to compare with the traffic estimate. k ), the mean relative prediction error (MRE) was reduced to 10.5%, compared to 12.9% MRE for a constant coefficient k.

[0274] Figure 11 shows the constant coefficient k and the discrete time-dependent coefficient k(t k) shows the median relative error of the coefficients. Increasing the granularity of the coefficients therefore improves the accuracy of the traffic estimation. Further refinement of the granularity ideally involves taking into account the location of the road segment to estimate its traffic flow. It is therefore desirable for the estimation of the scaling coefficients to depend on the location, e.g., the position of the segment being considered. Several techniques for doing this are described below.

[0275] The traffic information server receives measured traffic data Y(s i ,t k ) and road segments s i Probe data X(s) from a navigation device associated with the vehicle i ,t k ) is received.

[0276] Traffic information is the received measured traffic data Y(s i ,t k ) into the received probe data X(s i ,t k ) and calculate k(s i ,t k ) is calculated for road segment s i k(s i ,t k ) coefficients are calculated from the measured traffic data Y(s i ,t k ), probe data X(s i ,t k ) to road segment s i This may later be used to estimate the traffic flow.

[0277] The traffic information server determines the road segment s r Probe data X(s) from a navigation device associated with a vehicle moving r ,t k ) is received. The purpose of the present invention is not related to the traffic flow detector, but the traffic information server receives only the probe data X(s r ,t k ) to receive such road segments s rk(s r ,t k ) is to determine

[0278] The traffic information server receives probe data X(s r ,t k ) and X(s i ,t k ) is used to generate a traffic pattern profile P(s), referred to herein as a "probe profile." The probe profile describes the probe count versus time over a week, i.e., the change in vehicle count due to probe data. This profile is obtained by aggregating probe data over recurring weekly time intervals, e.g., one-hour intervals. Thus, the probe profile describes the weekly pattern of probe data, as shown in FIG. 12.

[0279] Figure 12 shows the time index t that counts the number of time intervals per day and per day of the week. k The profile shows the number of probes in a road segment as a function of . <ZであるプローブX(s r ,t k ) and Z is the number of time intervals in a week. r and s i Upon receiving probe data, a probe profile P(s) for all these road segments can be generated.

number

[0280] Equation 5 indicates that there are Z elements in the probe profile. For a map area, there are N+M road segments for which the traffic information server receives probe data. The number of probe counts for each element in the probe profile is the average for each different week considered in constructing the profile.

[0281] The traffic information server determines a parameter indicating profile similarity using the probe profile for the road segment. The similarity value is for the road segment of interest s r and establishes a link with all road segments {s i | i = 1... M} having traffic detectors. The road segments having traffic detectors can be called "reference segments". In other words, the probe profile of a road segment not associated with a traffic flow detector is matched to each of the probe profiles (reference probe profiles) of the road segment s i having a traffic detector. By the matching, k(s r , t k ) is estimated.

[0282] Finding the similarity between the probe profiles P(s1) and P(s2) is performed using a non - negative kernel function K(a, b) (do not confuse with the coefficient k(s, t) function) that takes two vector arguments and outputs a single real number in the range (0... 1] or 0 < K(a, b) ≤ 1. The result of the kernel function being 1 is generated only when a = b (the same vector) or when a value in the range (0... 1] is returned. The K(P(s)1, P(s2) kernel function maps the similarity of the two probe profiles P(s1) and P(s2) to the result of a real - valued function. The profile P(s j [[ID=]16]) is a vector having dimensions Z: P(s j ) = [X(s<00o1090>, t0), X(s j , t1), ···, X(s[[ID=]23] j , t Z ).

[0283] As an example, the kernel function can be a radial basis function, that is, <00o1094>TIFF0007789866000005.tif1258. In particular, the kernel function is a radial basis function kernel TIFF0007789866000006.tif1781.

[0284] The similarity between profiles and how similarity parameters can be determined is shown in FIG.

[0285] Figure 13 shows the similarity between the leftmost probe profile (which may be associated with a segment of interest) and four (reference) probe profiles on the right. The left probe profile has a clear peak near the end of the day (indicating evening traffic congestion). The similarity is used to obtain similarity parameters that indicate the similarity between the probe profile associated with the segment of interest and each of the reference profiles. These similarity parameters range from a maximum range of 0...1 to 0.3...0.9. In other words, the parameters are normalized.

[0286] Road segments r and a reference probe profile from set L (representing the road segments associated with the traffic flow detector) to determine the road segment s r Coefficient k(s r ,t k ) is estimated.

number

[0287] Equation 6 expresses the similarity measure K(P(s i ), P(s r )) and weighting coefficient α i and k(s i ,t k ) coefficient (see Equation 3).

[0288] The weighting coefficients are calculated based on the ground truth data Y(s i ,t k ), i.e., road segment s iAdvantageously (by way of example and not limitation), an L2 penalized ridge regression model is used for the regression function Q(α).

number

[0289] In Equation 7, the constant C relates to a regularization term found by cross-validation. Regression models are fitted with a standard quadratic loss function TIFF0007789866000009.tif1157 is used.

[0290] The above is the case where the traffic information server receives probe data X(s,t k ) for a road segment s in the map area receiving the traffic volume estimate Y(s,t k ;α) is generated from the received probe data X(s,t k ) and the traffic volume estimate Y(s,t k ;α) and the coefficient k(s,t k ) is linked to.

[0291] Equations 4 and 6 yield the following expressions for the coefficients:

number

[0292] Equation 8 and Equation 6 are Y(s,t k ;α)=k(s,t k ;α)X(s,t k ) is used to link the coefficients k(s,t k ;α) can also be formulated.

[0293] FIG. 14 is a functional diagram illustrating one system for implementing generation of a set of scaling coefficients according to embodiments described herein.

[0294] A preferred embodiment of a method for obtaining time-dependent scaling factors using such a system is described below with reference to FIG.

[0295] FIG. 14 shows the probe data and traffic detector data received by the traffic information server.

[0296] In step 1, the traffic information server selects a set S of road segments s within a map area A for which it will receive probe data (600 in FIG. 14). These may be segments within a given map area of ​​interest. In step 3, the server identifies (similar to the description above) a subset L of road segments from set S for which traffic detector data is also available, and a subset M of road segments for which only probe data is available. The probe data and traffic detector data for segments within subset L are labeled 610 and 620 in FIG. 14. The associated probe data and traffic detector data may be stored in respective databases.

[0297] In step 5, the traffic information server determines (using probe profile builder module 630) a probe profile describing probe data for a one-week time interval for each road segment in set S and (using probe profile builder module 640) for each road segment in set L. In step 7, the traffic server compares (using profile similarity comparison module 650) the probe profiles for the road segments from set S with the probe profiles for all road segments in set L and stores the results in similarity parameter module (660) - step 9.

[0298] In step 11, the traffic volume estimation module 670 of the traffic information server calculates the similarity parameters of the road segments in L and α (value α i) to obtain traffic volume estimates for the road segments in L. Regression analysis module 680 compares the traffic volume estimates with the observed traffic data for all road segments in L and updates a weighting vector α (the weighting values ​​are stored in weighting value module 690). Once the best match is found, the final weighting vector from module 690 and the similarity parameters from module 660 are used to generate a set of coefficients k for each road segment in S that describe, for each time interval in the probe profile, the value that converts received probe counts to estimated traffic volumes.

[0299] While this method can be performed for any segment of interest from S, it is particularly useful for obtaining traffic volume estimates for any segment for which traffic detector data is not available, i.e., forming part of subset M. Traffic detector data from subset L is used to verify the accuracy of the estimation function in the linear regression module. Based on Equation 8 above, it can be seen that the estimated scaling coefficients for a segment of interest within a time interval of interest forming part of subset M for which traffic detector data is not available are based on the similarity between the probe profile of the segment of interest and a reference probe profile associated with each one of a plurality of reference segments corresponding to segments in subset L for which traffic detector data is available. The estimated scaling coefficients are also based on a count of the segment's movements according to the probe data and a reference scaling coefficient for each of the reference segments based on measured traffic count data, i.e., the measured count divided by the probe count (Equation 4) for a given time interval of interest. The weight assigned to each such reference scaling coefficient is determined using a linear regression model trained using measured traffic detector data associated with segments in subset L.

[0300] Referring to FIG. 16, an experiment having a map area in which a set of road segments L is divided into two subsets, a first subset for generating traffic volume estimates and a second subset for verifying the accuracy of the estimation function, has an accuracy as shown in FIG. 16.

[0301] The figure shows the distributions for the previous results (baselines 1, 2) and the estimation method using probe profile matching and weighted vector estimation (neighborhood). The volume (quantity) estimates have a median relative prediction error (MRE) of 5.78% (compared to 12.9% and 10.5% MRE for each of baseline 1 (FIG. 1) and baseline 2 (FIG. 11)).

[0302] As described above, in order to further improve the accuracy, it is also desirable for the scaling factor to be position-dependent, i.e., dependent on the position of the segment.

[0303] Such a method can be carried out as follows. As described above, the following terms are used. · S is the set of road segments s within the map area A · L = {s i | 0 ≤ i < N} is the set of road segments s within the map area A where traffic information is available. Each road segment s i is associated with a traffic flow detector i <000117​​​​​​​​​​​​​​​In this further embodiment, the difference is that the estimated scaling factors (functions) use the reference scaling factors (obtained using measured traffic detector data for the reference segment where such data is available) in a different way.

number

[0305] In the above equation, the kernel function of equation (8) is replaced by a decay function that represents the contribution of the reference (measurement) scaling coefficients that decreases as a function of the distance between the reference road segment and the road segment for which the estimated coefficients are being determined. i scales the estimated coefficients to the average coefficients for map area A.

[0306] Thus, rather than considering the similarity between the probe profiles of the segment and the reference segment, a linear regression training model is used to refine the weighting of the reference scaling factors calculated for segments with traffic detector data, and in these embodiments the scaling factors for the segment of interest are based on the reference scaling factors for the reference segment, which is the segment for which measured traffic data exists, and different weights are assigned to the reference scaling factors depending on the distance between the reference segment and the segment of interest.

[0307] This approach spreads the difference between the measured reference scaling factor and the average factor over several nearby road segments.

[0308] The distance function can be based on any suitable distance measure, for example, Euclidean distance measured between (start / end / center) points of road segments, (shortest, fastest) routing distance between these points, or a distance that depends on road class. The distance function can be any function that decays to an average coefficient contribution value.

[0309] In this example, the obtained scaling factor for the segment of interest is further time dependent, and the reference scaling factor, and therefore the scaling factor for the segment of interest, is for a given time interval, however, it is envisaged that the obtained scaling factor may alternatively be dependent only on position.

[0310] This embodiment can be combined with the previous embodiment, for example, to further consider the similarity between the probe profiles of the segments, or to use a linear regression model based on measured (reference) scaling factor data.

[0311] More specifically, Equation 9 calculates the reference (measurement) scaling factor k(s) based on the measurement data and the normalization value α. i ,t k ) to the scaling factor k(s r ,t k The decay function D(s ;α) is used to estimate i ,s r ) is introduced. The decay function aims to spread the reference scaling factor over a given map area (e.g., a range of locations). At longer distances, the result of the decay function should approach the average value of the reference scaling factor (based on measured data).

[0312] An example of the decay function is shown in FIG.

[0313] The figure shows two reference segments s i and s i+1 , and the map locations associated with these reference segments. For clarity, the map locations are shown as a one-dimensional map, e.g., based on the distance between two reference segments. The figure also shows the scaling coefficients k(s,t) of these map locations. k ) for all reference segments s i For the average coefficient k av (or the time-dependent coefficient k av (t kFor two reference positions, the road segment s r Coefficient k(s r ,t k ;α) is calculated for road segment s r is the road segment s i and s i+1 Since the reference segments are close to both the road segment s and the reference segments s, the reference coefficients from these reference segments are applied to the road segment s using a decay function. r is propagated to.

[0314] The decay function in Equation 8 combines the reference scaling factors to obtain an estimated scaling factor for the road segment. For locations far from the reference road segment, the contributions of all the reference scaling factors are multiplied by the average value of the reference scaling factors k av It should be close to.

[0315] The propagation model described in Equation 8 includes a vector of scaling values ​​that scale the propagation. For example, the scaling is r The average value over all reference road segments s i The propagation can be adjusted to ensure that the average scaling factor is close to The decay function shown is merely an example. Other forms of decay functions can be used, such as exponential decay, step decay, hyperbolic decay, or an inverse distance weighting function. The decay should preferably occur over a relatively short range.

[0316] Figure 18 shows the 2D results overlaid on the map area.

[0317] Such an embodiment is illustrated in more detail with reference to FIG. 18. FIG. 18 shows a map area including road segments (light gray) and traffic detectors (black dots). The map also shows the k(s) function for the map area, which shows the results of distributing the measured k values ​​of the reference road segments (i.e., the road segments associated with the traffic detectors). The figure shows the k(s) function as a "heat map" overlay of the map area. In practice, the k(s) scaling factor is determined based on the location associated with each road segment within the map area. For example, the location may be the start, end, or center point of the road segment. The k(s) value for that location of the road segment is then associated with the entire road segment.

[0318] The heatmap in Figure 18 depicts the estimated coefficients k(s), which are determined using a decay function. The k(s) coefficients may also be obtained using probe profile similarity methods. In either of these methods, the estimated coefficients are calculated based on the time or time index t k may also depend on the estimated coefficients k(s,t) and k(s,t), respectively. k ) The method for determining the estimated coefficients as a function of (map) location typically involves determining a weighting value α for scaling the contributions to the estimated coefficients.

[0319] According to various embodiments of the present invention, electronic map data may be provided that includes reference segments, which are segments for which measured count data is available, and "non-reference" segments for which measured count data is not available. Each "reference segment," i.e., a segment for which measured count data is available, is associated with data indicating a reference scaling factor for a given time of interest. Each reference segment may be associated with a reference scaling factor profile that represents the change in the reference scaling factor with respect to time, from which a scaling factor for the given time may be obtained.

[0320] The baseline scaling factor may be based on the ratio of measured traffic counts to probe counts for a segment for the applicable time period or times.

[0321] The reference scaling factors for the reference segments may be based at least in part on live data. For example, this may be particularly applicable when estimated scaling factors for non-reference segments are needed for the current time. Alternatively or additionally, the reference scaling factors (or scaling factor profiles) may be based on historical data, for example, based on historical probe profiles. This may be applicable when the time of interest for which estimated scaling factors are needed is a past or future time, but is also applicable to the current time.

[0322] Each non-reference segment may be associated with one or more reference segments. These reference segments are a subset of the reference segments of the electronic map in the considered area that are determined to be relevant for obtaining traffic data for the given non-reference segment. In other embodiments, no association exists in the map data and the method may extend to determining the associated reference segment(s) for the given segment.

[0323] The subset of reference segments associated with a given non-reference segment may be determined as desired. For example, the relevant reference segments may be determined by comparing a time-dependent, e.g., weekly, probe profile for the segment with the corresponding probe profile for the reference segment. The subset of reference segments may be selected based on the similarity of their probe profiles to those of the non-reference segments. For example, the most similar reference segments may be selected, or a similarity value may be assigned to each reference segment, where these reference segments have a similarity value above a predetermined value, or a predefined number of the most similar reference segments may be selected, etc.

[0324] The selection of reference segments for inclusion in the subset may alternatively or additionally be based on the proximity (proximity) of the reference segments to non-reference segments for which traffic data is required. Proximity may be spatial or temporal proximity, such as straight-line distance through the navigable network between the reference segment and the non-reference segments, or distance in terms of travel time or spatial distance. For example, only reference segment(s) within a predetermined distance or travel time, or only a predetermined number of the closest reference segments, may be considered.

[0325] Alternatively or additionally, the selection may be based on the similarity of characteristics between the reference and non-reference segments, for example by considering functional road class.

[0326] Any of these techniques can be used alone or in any combination to enable identification of a set of one or more reference segments that are related to a non-reference segment of interest, which can aid in obtaining more accurate estimated scaling factors for the non-reference segments that are related to the segment.

[0327] This step of identifying a subset of reference segments, which may be a single reference segment or multiple reference segments, may be performed before the aforementioned step of determining estimated scaling factors based on multiple reference scaling factors. Thus, the determination of estimated scaling factors may be based on multiple reference scaling factors and may include weighting contributions from the multiple reference scaling factors based on, for example, probe profile similarity, and / or proximity of associated reference and non-reference segments, and / or any other criteria as described above.

[0328] It will be appreciated that, from a broader perspective, the present invention enables traffic volume to be determined for segments for which absolute vehicle count data does not exist (at least for a given time of interest) using reference scaling factors associated with one or more reference segments for which such absolute vehicle count data does exist.

[0329] Those skilled in the art will appreciate that an apparatus provided for carrying out the methods as described herein may comprise hardware, software, firmware, or any combination of two or more of these.

[0330] Those skilled in the art will understand that the term GPS data has been used to refer to positioning data derived from the GPS Global Positioning System. Other positioning data may be processed in a manner similar to that described herein. Thus, the term GPS data may be interchangeable with the phrase positioning data.

[0331] All of the features disclosed in this specification, and / or all of the steps of any method or process so disclosed, may be combined in any combination, except combinations in which at least some of such features and / or steps are mutually exclusive.

[0332] Each feature disclosed in this specification, unless stated otherwise, may be replaced by alternative features serving the same, equivalent, or similar purpose. Thus, unless stated otherwise, each feature disclosed is only an example of a generic series of equivalent or similar features.

[0333] The invention is not limited to the details of the foregoing embodiments. The invention extends to any novel or any novel combination of features disclosed herein, or any novel or any novel combination of steps of any method or process so disclosed. The claims should not be construed to cover only the foregoing embodiments, but also any embodiment that falls within the scope of the claims.

Claims

1. 1. A method for estimating traffic volume for a given time period for a given segment of an electronic map representing a navigable network in an area, the method being computer-implemented, the electronic map including a plurality of segments representing navigable elements of the navigable network in the area, the navigable network in the area including navigable sections associated with at least one traffic detector and navigable sections not associated with any traffic detector, the given segment being a section not associated with any traffic detector that represents at least a portion of a navigable section of the navigable network in the area, the electronic map further including a plurality of reference segments, each reference segment being a section associated with a traffic detector that is a navigable section of the navigable network in the area. segments representing at least a portion of a gateable section, each reference segment associated with data indicative of a respective reference scaling factor for the given time, the reference scaling factor being based on a measured count of vehicles traveling on the at least a portion of the section represented by the reference segment for the given time, and a count of devices associated with vehicles traveling on the at least a portion of the section represented by the reference segment for the given time, the measured count of vehicles being based on data measured by the at least one traffic detector associated with the section, and the count of devices associated with vehicles being based on position data and associated timing data relating to movement of a plurality of devices along the at least a portion of the section represented by the reference segment; estimating the traffic volume for the given segment for the given time using data indicating a count of devices associated with vehicles traveling along at least a portion of the section of the navigable network represented by the given segment for the given time and an estimated scaling factor for the given segment for the given time, wherein the estimated scaling factor for the segment is based on the reference scaling factor associated with each of a subset of one or more reference segments of the electronic map associated with the given segment, and the count of devices is based on position data and associated timing data regarding movement of a plurality of devices along at least a portion of the navigable network section represented by the given segment for the given time; generating data indicative of the estimated traffic volume for the given segment for the given time; Including, the estimated scaling factor for the given segment is estimated using data indicative of a similarity of a probe profile associated with the given segment to each of a set of one or more reference probe profiles, each reference probe profile associated with a respective one of the one or more reference segments for which a reference scaling factor is used in determining the estimated scaling factor; The method, wherein the estimated scaling factor is based on a weighted sum of a plurality of reference scaling factors.

2. 2. The method of claim 1, wherein the reference scaling factor for the given time associated with each reference segment is based on a ratio of the measured count for the given time based on the traffic detector data to the count of devices for the given time based on the location data and associated timing data.

3. 3. The method of claim 1, wherein each reference segment is associated with data indicative of a time-dependent reference scaling factor profile, the reference scaling factor profile indicating changes in the reference scaling factor for the reference segment over time.

4. 4. The method of claim 3, wherein the reference scaling factor profile is based at least in part on a probe profile indicative of changes in counts of devices associated with a vehicle traversing the at least portion of the navigable section represented by the given reference segment with respect to a time determined based on position data and associated timing data relating to movement of a plurality of devices associated with a vehicle along the at least portion of the navigable section represented by the reference segment.

5. The method of claim 1 , wherein the given time is a current time and the reference scaling factor is based at least in part on live data.

6. The method of claim 1 , wherein the electronic map further comprises, for each segment that is not a reference segment, data indicating the subset of one or more reference segments associated with the segment.

7. 7. The method of claim 1, further comprising determining the subset of one or more of the reference segments associated with the given segment, and optionally storing data indicative of the determined subset of one or more reference segments with the associated segments in the electronic map.

8. 8. The method of claim 7, wherein the subset of one or more reference segments is determined based at least in part on a comparison of a probe profile associated with the given segment to reference probe profiles associated with some of the reference segments, the probe profiles indicating changes in counts of devices associated with vehicles traversing the at least portion of the navigable section represented by the given segment with respect to a time determined based on position data and associated timing data regarding the movement of multiple devices associated with vehicles along the at least portion of the navigable section represented by the reference segment, and the reference probe profiles indicating changes in counts of devices associated with vehicles traversing the at least portion of the navigable section represented by the reference segment with respect to a time determined based on position data and associated timing data regarding the movement of multiple devices associated with vehicles along the at least portion of the navigable section represented by the reference segment.

9. The method of claim 8 , wherein the subset of one or more reference segments includes one or more reference segments having a reference probe profile determined to be most similar to the probe profile of the given segment.

10. 10. The method of claim 7, wherein the subset of one or more reference segments is determined based at least in part on the proximity of the reference segments to the location of the given segment.

11. 11. The method of claim 7, wherein the subset of one or more reference segments is determined based at least in part on a similarity of characteristics, e.g., functional road class (FRC), of the reference segments to the given segment.

12. 10. The method of claim 1, wherein the probe profile indicates a change in count of devices associated with a vehicle moving through at least a portion of the navigable section represented by the given segment with respect to a time determined based on position data and associated timing data relating to the movement of a plurality of devices associated with a vehicle along at least a portion of the navigable section represented by the segment, and the reference probe profile indicates a change in count of devices associated with a vehicle moving through at least a portion of the navigable section represented by the reference segment with respect to a time determined based on position data and associated timing data relating to the movement of a plurality of devices associated with a vehicle along at least a portion of the navigable section represented by the reference segment.

13. 13. The method of claim 12, wherein the estimated scaling factor for the given segment is based on a plurality of the reference scaling factors, and the contribution of a given reference scaling factor to the estimated scaling factor for the given segment is based at least in part on the similarity of the reference probe profile associated with the reference segment with which the reference scaling factor is associated to the probe profile associated with the given segment.

14. 14. The method of claim 1, wherein the estimated scaling factor for the given segment is based on a plurality of the reference scaling factors, the contribution of each reference scaling factor to the estimated scaling factor being based at least in part on the proximity of the reference segment associated with the reference scaling factor to the given segment, and optionally a greater weight being assigned to a reference scaling factor associated with a reference segment that is closer to the given segment.

15. 15. The method of claim 1, wherein data indicative of a set of weighting values ​​for use in obtaining the weighted sum of the plurality of reference scaling factors is obtained using a linear regression training model.

16. 16. The method of claim 15, wherein the linear regression training model uses data indicative of a measured count of vehicles traveling through the at least a portion of the navigable section represented by the reference segment for the given time period determined based on data measured by the at least one traffic detector associated with the navigable section.

17. 17. The method of claim 1, comprising receiving data indicative of the given segment for which traffic data is required and data indicative of a time of interest, and using the data indicative of the time of interest to identify the given time.

18. 18. The method of any one of claims 1 to 17, wherein the given time is the current time or a future time.

19. 19. The method of any one of claims 1 to 18, wherein the given time is a time interval, optionally a recurring time interval.

20. 20. The method of claim 19, wherein the recurring time interval is a time interval for a given day of the week.

21. 21. The method of claim 1, further comprising associating data indicative of the estimated traffic volume with data indicative of the given segment to which it relates, and optionally transmitting the data indicative of the estimated traffic volume for the given segment and / or displaying the data indicative of the estimated traffic volume for the given segment to a user.

22. 22. The method of any one of claims 1 to 21, further comprising storing the estimated traffic volume and / or traffic density for later display and / or displaying the estimated traffic volume and / or traffic density to a user.

23. 1. A system for estimating traffic volume for a given time period for a given segment of an electronic map representing a navigable network in an area, the electronic map including a plurality of segments representing navigable elements of the navigable network in the area, the navigable network in the area including navigable sections associated with at least one traffic detector and navigable sections not associated with any traffic detector, the given segment being a section not associated with any traffic detector that represents at least a portion of a navigable section of the navigable network in the area, the electronic map further including a plurality of reference segments, each reference segment being a section associated with a traffic detector that represents at least a portion of a navigable section of the navigable network in the area. and wherein each reference segment represents at least a portion of the section represented by the reference segment, and each reference segment is associated with data indicative of a respective reference scaling factor for the given time, the reference scaling factor being based on a measured count of vehicles traveling along at least the portion of the section represented by the reference segment for the given time, and a count of devices associated with vehicles traveling along at least the portion of the section represented by the reference segment for the given time, the measured count of vehicles being based on data measured by the at least one traffic detector associated with the section, and the count of devices associated with vehicles being based on position data and associated timing data regarding movement of a plurality of devices along the at least the portion of the section represented by the reference segment. estimating the traffic volume for the given segment for the given time using data indicating a count of devices associated with vehicles traveling along at least a portion of the section of the navigable network represented by the given segment for the given time and an estimated scaling factor for the given segment for the given time, wherein the estimated scaling factor for the segment is based on the reference scaling factor associated with each of a subset of one or more reference segments of the electronic map associated with the given segment, and the count of devices is based on position data and associated timing data regarding movement of a plurality of devices along at least a portion of the navigable network section represented by the given segment for the given time; generating data indicative of the estimated traffic volume for the given segment for the given time; a set of one or more processors configured to cause the estimated scaling factor for the given segment is estimated using data indicative of a similarity of a probe profile associated with the given segment to each of a set of one or more reference probe profiles, each reference probe profile associated with a respective one of the one or more reference segments for which a reference scaling factor is used in determining the estimated scaling factor; The system, wherein the estimated scaling factor is based on a weighted sum of a plurality of reference scaling factors.

24. 23. A computer program comprising instructions that, when read by a computing device, cause the computing device to operate in accordance with the method of any one of claims 1 to 22, said computer program optionally stored on a non-transitory computer readable medium.

Images