DE102021104640A1 - Procedure for GNSS-based localization of a vehicle - Google Patents

Abstract

The invention relates to a method for GNSS-based localization of a vehicle (1) using a localization device (2) of the vehicle (1), comprising at least the following steps: a) receiving at least one GNSS signal (3) from at least one GNSS satellite ( 4) by means of at least one GNSS receiver (5) of the vehicle (1),b) reading in at least one piece of statistical information from a GNSS profile for the reception situation on which the reception in step a) is based, the at least one piece of statistical information being suitable, to describe at least one property of the received GNSS signal (3),c) determining at least one piece of localization information using the localization device (2) taking into account the at least one GNSS signal (3) received in step a) and the one read in step b). at least one piece of statistical information.

Description

The invention relates to a method for GNSS-based localization of a vehicle using a localization device of the vehicle, a computer program for carrying out the method, a machine-readable storage medium on which the computer program is stored and a localization device for a vehicle, set up for carrying out the method. The invention can be used in particular in autonomous driving.

State of the art

Automated or autonomous driving places very high demands on location. GNSS (Global Navigation Satellite System, such as GPS or GLONASS) plays a particularly important role here, as this is currently the most important positioning system to enable direct absolute positioning. In addition to position accuracy, automated or autonomous driving places particularly high demands on the availability and integrity (or correctness) of the location information, for example on the correctness of the accuracy specification.

GNSS correction data services are known which can determine the influence of the GNSS errors in orbit (essentially satellite orbit errors, satellite clock errors, code and phase biases, as well as ionospheric and tropospheric refraction influences). With the help of such existing correction data services, it is possible to take into account the error influences mentioned in the GNSS-based positioning, so that the accuracy of the GNSS-based positioning result can be increased.

The existing correction data services enable GNSS-based positioning accuracy in the cm range by correcting the so-called Signal-In-Space (SIS) errors (anomalies in the satellite orbit, satellite clock, atmospheric effect) as long as there is line of sight to the satellites used. In the case of shadowing, e.g. by high buildings in urban areas, signals can be reflected on the surfaces, so that the use of correction data services still compensates for the SIS errors and thus usually achieves an increase in accuracy compared to not using correction data, but it still deteriorates In this case, the localization accuracy (e.g. to an accuracy of the order of meters or 10 meters) due to signal refractions or reflections (so-called multipath effects). This problem is also described as GNSS performance degradation in urban or inner-city areas.

Based on this, the aim is to improve the GNSS position accuracy and, if possible, the integrity of the position result in inner-city areas.

Disclosure of Invention

1. a) Empfangen mindestens eines GNSS-Signals von mindestens einem GNSS-Satelliten mittels mindestens eines GNSS-Empfängers des Fahrzeugs,
2. b) Einlesen mindestens einer statistischen Information aus einem GNSS-Profil zu der dem Empfang in Schritt a) zugrunde liegenden Empfangssituation, wobei die mindestens eine statistische Information geeignet ist, um mindestens eine Eigenschaft des empfangenen GNSS-Signals zu beschreiben,
3. c) Ermitteln mindestens einer Lokalisierungsinformation mittels der Lokalisierungseinrichtung unter Berücksichtigung des in Schritt a) empfangenen mindestens einen GNSS-Signals und der in Schritt b) eingelesenen mindestens einen statistischen Information.

According to claim 1, a method for GNSS-based localization of a vehicle by means of a localization device of the vehicle is proposed here, comprising at least the following steps:

1. a) receiving at least one GNSS signal from at least one GNSS satellite using at least one GNSS receiver of the vehicle,
2. b) reading in at least one item of statistical information from a GNSS profile relating to the reception situation on which the reception in step a) is based, the at least one piece of statistical information being suitable for describing at least one property of the received GNSS signal,
3. c) determining at least one piece of localization information using the localization device, taking into account the at least one GNSS signal received in step a) and the at least one piece of statistical information read in step b).

The specified sequence of steps a), b) and c) is exemplary and can thus occur during a regular operating sequence of the method or can take place at least once in the specified sequence. In addition, steps a), b) and c), in particular at least steps a) and b), can also be carried out at least partially in parallel or simultaneously.

By reading in at least one item of statistical information from a GNSS profile, the method advantageously allows a priori knowledge of the GNSS signal quality to be used in the GNSS-based localization or in the localization device. This describes a new approach as to how the location accuracy and/or the integrity of the localization result can be improved by using GNSS profiles or previously learned GNSS signal characteristics in the localization device. The method thus advantageously contributes to the fact that the GNSS position accuracy and, if possible, also the integrity of the position result can be improved in inner-city areas.

In particular, with the described new approach using one or more (specific) statistical parameters from GNSS profiles, an advantageous utilization of the GNSS-Pro file or GNSS signal characteristics. This can advantageously be done taking into account the use of several different localization methods.

GNSS stands for Global Navigation Satellite System, such as GPS (Global Positioning System), GLONASS, Beidou or Galileo. A GNSS-based localization is to be understood as a localization that is at least also or mainly based on GNSS measurements or GNSS evaluations. In addition, other localization processes that work, for example, with environmental sensor data, inertial or inertial sensor data and/or (digital) map data can also be included in the GNSS-based localization. The vehicle is usually a motor vehicle, such as an automobile. The vehicle is preferably one that is set up for at least partially automated or autonomous driving.

In step a), at least one GNSS signal is received from at least one GNSS satellite by means of at least one GNSS receiver of the vehicle. The GNSS receiver can be part of the localization device or connected to it. The GNSS receiver typically includes at least one GNSS antenna. The GNSS receiver and/or the localization device can carry out transit time measurements with the received GNSS signals in order to determine the current (geodetic) position of the GNSS receiver or the vehicle, in particular by means of trilateration or triangulation.

In step b), at least one piece of statistical information is read from a GNSS profile for the reception situation on which the reception in step a) is based, the at least one piece of statistical information being suitable for describing at least one property of the received GNSS signal. The reception situation is characterized in particular by the (geodetic) position of the GNSS receiver or the vehicle at the time the GNSS signal was received and the (orbital) position of the at least one GNSS satellite at the time the GNSS signal was transmitted. The (geodetic) position of the GNSS receiver or the vehicle can be approximated by an area or an area in which the vehicle is probably located. The (orbital) position of the at least one GNSS satellite can be determined using navigation data that each individual satellite broadcasts (so-called broadcast ephemeris). This also allows the geometric distribution of the satellite constellation to be described as a further quality measure at the point in time when the statistical information of the signal is determined.

The GNSS profile usually includes at least one GNSS signal characteristic. The GNSS profile can be specific to the reception situation and in this context, for example, can include or describe at least one GNSS signal characteristic for a specific reception situation. For example, a model can be provided (inside the vehicle or outside the vehicle) in which a large number of GNSS profiles are stored. This model can be accessed in step b). In this case, the GNSS profile that is stored in the model for the reception situation on which the reception in step a) is based can be accessed. The respective GNSS profile can describe, for example, GNSS received signal properties or GNSS signal characteristics with regard to pseudorange, frequency or Doppler, signal strength and/or carrier phase. The GNSS profiles or GNSS signal characteristics can, for example, be suitable for carrying out a localization method such as GNSS fingerprint, signal corrections, weighting and/or selection of signals and/or amplifying integrity information. For example, the GNSS profiles can be made available in the form of mapped GNSS profiles or GNSS signal characteristics. For example, the GNSS profiles or GNSS signal characteristics can be provided as a data layer in a digital map (HD map) of an (automated) vehicle.

The (respective) GNSS profile usually also includes at least one piece of statistical information. The at least one piece of statistical information is suitable for describing at least one property of the received GNSS signal. This property is in particular a quality property or qualitative property of the received GNSS signal. In other words, this relates in particular to a property that describes the quality of the received GNSS signal. For example, a statistical parameter or a statistical parameter can be read in as statistical information. An example of such a parameter can be a variance. The at least one piece of statistical information or the statistical parameter can be suitable, for example, to represent the uncertainty of at least one GNSS received signal property and/or individual, possibly aggregated GNSS received signal properties that are described by the relevant GNSS profile will or will

In particular, the respective GNSS profiles or GNSS signal characteristics provide statistical ones Parameters for the GNSS received signal properties regarding pseudorange, frequency or Doppler, signal strength, and/or carrier phase. This has the particular advantage that these are now available as a priori knowledge. In an advantageous manner, this can be done via a model, depending on the reception situation, in particular depending on the receiver position and satellite position or reception position and associated azimuth and elevation angle between the satellite, which emits the signal in question, and the antenna of the GNSS receiver, which receives the relevant signal.

The GNSS profiles or GNSS signal characteristics can be obtained by aggregating many individual observations (e.g. using crowd sourcing). Several measurements of a GNSS signal property relating to the same reception situation or constellation of reception and satellite position (which is usually repeatable due to cyclic satellite orbits, so that associated measurements may have been collected on different days, for example) can be used to calculate a statistical parameter of a Determine GNSS received signal properties.

In step c), at least one piece of localization information is determined by means of the localization device, taking into account the at least one GNSS signal received in step a) and the at least one piece of statistical information read in step b). The at least one item of localization information can include, for example, one or more of the following parameters: (own) position of the vehicle, (own) speed of the vehicle, (own) acceleration of the vehicle. The localization information preferably includes at least the (own) position of the vehicle. Furthermore, the localization information can alternatively or cumulatively (in addition to one or more of the aforementioned parameters) include integrity information, such as a confidence range or a confidence interval. A so-called "protection level" can be determined as an example of integrity information. As a rule, this relates to a region or an area in which the true vehicle position lies with a definable minimum probability. The statistical information can also be used to weight individual signals or satellites in order to regulate their influence on the position estimate. It is also possible to use the statistical information for real-time calibration of the internal monitoring, so that, for example, detection threshold values can be selected dynamically according to the situation.

According to an advantageous embodiment, it is proposed that the at least one piece of statistical information is suitable for describing a quality of the received GNSS signal. The at least one piece of statistical information can be stored in the (respective) GNSS profile in the form of at least one parameter, for example, which is suitable for describing a quality of the received GNSS signal. The at least one parameter can include the short-term variance and/or the long-term variance as exemplary parameters.

For example, statistical parameters can be used in the GNSS profiles or GNSS signal characteristics, which represent the uncertainty of the individual aggregated GNSS received signal properties, for example in the form of the variance. On the one hand, the so-called short-term variance can be used for this purpose, which usually represents the typical scatter of a certain signal property (e.g. pseudoorange or carrier to noise ratio) during a measurement (i.e. the scatter over the time of the measurement regarding a measurement period of e.g. 10s). On the other hand, the so-called long-term variance can be used (alternatively or cumulatively), which usually represents the scatter of a specific signal property considering several measurements (i.e. measurements that have been taken on different days, for example).

According to a further advantageous embodiment, it is proposed that the at least one piece of statistical information includes long-term statistical information, such as long-term variance. In particular, it is preferred in this context that the statistical long-term information (in particular long-term variance) was determined on the basis of chronologically different GNSS observations for the same reception situation. The use of long-term information (long-term variance) is particularly advantageous since it can provide information about the repeatability of GNSS received signal properties.

Taking into account one or more of the uncertainties mentioned (in particular the long-term variance), the GNSS profiles or GNSS signal characteristics can advantageously be used in a targeted manner. This can be done in particular by using different methods (e.g. GNSS fingerprint, correction of signals, weighting/selection of signals, PL boosting) depending on the underlying uncertainty.

According to a further advantageous embodiment, it is proposed that in step c) depending on the at least one piece of statistical information tion a localization method for determining the at least one piece of localization information is selected. In particular, in this context, taking into account the uncertainties described by the statistical information (in particular the long-term variance), the GNSS profiles or GNSS signal characteristics can be used in a targeted manner by using a GNSS fingerprint method or a GNSS Trilateration method are used. A so-called GNSS fingerprint method (GNSS fingerprint method) can be selected as the localization method, for example, if the statistical information describes an uncertainty value that is below a predefined limit value.

According to a further advantageous embodiment, it is proposed that in step c) a measure for adapting the determination of the at least one piece of localization information is selected as a function of the at least one piece of statistical information. In particular, in this context, taking into account the uncertainties described by the statistical information (in particular the long-term variance), the GNSS profiles or GNSS signal characteristics can be used in a targeted manner, in that, depending on the underlying uncertainty, a correction of GNSS signals or a weighting ( or selection) of GNSS signals are used. As a measure for adapting the determination of the at least one item of localization information, a correction of GNSS signals can be carried out, for example, if the statistical information describes an uncertainty value that is below a predefinable limit value. Alternatively or cumulatively, a further measure for adapting the determination of the at least one item of localization information can be, for example, a weighting and/or selection of GNSS signals depending on the statistical information or the uncertainty value. If the statistical information describes an uncertainty value that is above a predefinable limit value, the relevant GNSS signal can be correspondingly downgraded, for example.

According to a further advantageous embodiment, it is proposed that the at least one piece of localization information includes integrity information about the integrity of a localization result. The integrity information can be determined, for example, in the form of a confidence range or a confidence interval. A so-called "protection level" can be determined as an example of integrity information. As a rule, this relates to a region or an area in which the true vehicle position lies with a definable minimum probability. In this context, as a measure to adapt the determination of the at least one piece of localization information, the integrity information can be amplified, such as an artificial increase in the integrity information (e.g. as a surcharge on the protection level) if the statistical information describes an uncertainty value that is above a predefined limit value.

Furthermore, the artificial increase in the integrity information can depend on the degree of uncertainty or on the magnitude of the uncertainty value. Furthermore, the artificial increase in the integrity information can be dependent on the number of GNSS profiles that, based on the same place (the same position) or the same area, each contain statistical information that describes an uncertainty value that is above a (the) predefined limit value lies.

According to a further aspect, a computer program for carrying out a method described here is also proposed. In other words, this relates in particular to a computer program (product), comprising instructions which, when the program is executed by a computer, cause the latter to execute a method described here.

According to a further aspect, a machine-readable storage medium is also proposed, on which the computer program is stored. The machine-readable storage medium is usually a computer-readable data carrier.

According to a further aspect, a localization device for a vehicle is proposed, the localization device being set up to carry out a method described here. For example, the storage medium described above can be part of the localization device or connected to it. The localization device is preferably arranged in or on a (motor) vehicle or is provided and set up for installation in or on such a vehicle.

The localization device is preferably a GNSS sensor or includes one. The localization device is also preferably provided and set up for automated or autonomous operation of the vehicle. Furthermore, the localization device can be a combined movement and position sensor. Such is particularly advantageous for autonomous vehicles. The localization device or a computing unit (processor) of the localization device can, for example, access the computer program described here in order to execute a method described here.

The details, features and advantageous configurations discussed in connection with the method can accordingly also occur in the computer program presented here and/or the storage medium and/or the localization device and vice versa. In this respect, full reference is made to the statements there for a more detailed characterization of the features.

• 1: einen beispielhaften Ablauf des hier beschriebenen Verfahrens, und
• 2: eine bespielhafte Lokalisierungseinrichtung in einem Fahrzeug.

The solution presented here and its technical environment are explained in more detail below with reference to the figures. It should be pointed out that the invention should not be limited by the exemplary embodiments shown. In particular, unless explicitly stated otherwise, it is also possible to extract partial aspects of the facts explained in the figures and to combine them with other components and/or findings from other figures and/or the present description. They show schematically:

• 1 : an example of the procedure described here, and
• 2 : an exemplary localization device in a vehicle.

1 shows a schematic of an exemplary sequence of the method described here. The method is used for the GNSS-based localization of a vehicle 1 using a localization device 2 of the vehicle 1 (cf. 2 ). The sequence of steps a), b) and c) represented by blocks 110, 120 and 130 is an example and can thus be set in a regular operating sequence.

In block 110, according to step a), at least one GNSS signal 3 is received from at least one GNSS satellite 4 by means of at least one GNSS receiver 5 of the vehicle 1. In block 120, according to step b), at least one piece of statistical information is read in from a GNSS profile for the reception situation on which the reception in step a) is based, the at least one piece of statistical information being suitable for describing at least one property of the received GNSS signal 3 . In block 130, according to step c), at least one piece of localization information is determined by means of the localization device 2, taking into account the at least one GNSS signal 3 received in step a) and the at least one piece of statistical information read in step b).

An embodiment variant of the method is explained as an example using the GNSS profile relating to the pseudoranges (abbreviated: PR). Specifically, this is information, e.g. on the absolute PR and/or the PR error. The corresponding GNSS profile can be provided as an example for each currently receivable satellite (each GNSS profile can describe the signal property of a satellite signal; since several satellites are received at the same time, there can be an associated GNSS profile for each receivable satellite).

An uncertainty σ ² is assigned to the respective GNSS profile via the variance. The variance here is an example of the statistical information that can be read from the GNSS profile. This also represents an example for the fact that and possibly how the at least one piece of statistical information can be suitable for describing a quality of the received GNSS signal 3 .

Furthermore, the at least one piece of statistical information can include long-term statistical information, such as long-term variance. With a view to the embodiment variant described, this means that the variance can be a long-term variance. The long-term variance is usually determined on the basis of chronologically different GNSS observations of the same reception situation.

For example, in the case of a low level of uncertainty, the respective GNSS profile, which describes the PR error, can be used to correct the PR measurement and/or the respective GNSS profile, which describes the absolute PR, can be used for a GNSS fingerprint method be used.

This also represents an example of the fact that, and possibly as in step c), a localization method for determining the at least one piece of localization information can be selected as a function of the at least one piece of statistical information.

In the case of a high level of uncertainty associated with a GNSS profile, this speaks for a less reliable GNSS signal, especially in view of the long-term variance, which is associated with a complex reception condition for the satellite signal in question. Provision can accordingly be made for such a signal to be taken into account to a lesser extent when determining the position.

With the help of the statistical information available from the GNSS profile (a priori) on the uncertainty of the signal, in this case, for example, a weighting or selection of signals to be taken into account can be carried out (e.g. directly via the factor 1/σ ² or in steps), for example by including the uncertainty as “measurement noise” in the position calculation, for example using a Kalman filter.

This also represents an example of the fact that, and possibly as in step c), a measure for adapting the determination of the at least one piece of localization information can be selected as a function of the at least one piece of statistical information.

A high level of uncertainty associated with a GNSS profile can be an indicator of complex reception conditions or reception situations, particularly in view of the long-term variance, such as can arise in inner-city areas due to complex signal shadowing by tall buildings.

It is correspondingly difficult to determine a reliable uncertainty of the current position estimate using the so-called protection level (PL, e.g. as an error ellipse around the position). In order to take into account the complexity of the reception situation and the associated uncertainty of the GNSS signal at the position level as well, the PL calculation can be calculated more conservatively in this situation (here also referred to as PL amplification or PL boosting), for example Example in which an additional surcharge of the PL is applied in this case.

In a further variant, the surcharge of the PL boosting can also be dependent on the degree of uncertainty.

The situation can be exacerbated if this state of high uncertainty is reached simultaneously for several received satellite signals, i.e. several GNSS profiles indicate high uncertainty. This can also be taken into account in PL boosting. For example, the "surcharge" can be determined depending on the frequency of GNSS signals with a high degree of uncertainty.

In this context, for example, a PL can be added if a significant number of GNSS profiles show a high degree of uncertainty.

This also represents an example of the fact that and possibly how the at least one piece of localization information can include integrity information about the integrity of a localization result.

The facts illustrated on the basis of the GNSS profile or GNSS signal characteristic for PR or PR errors can be transferred in an analogous manner to further elements of the GNSS signal characteristics or further GNSS profiles for describing further GNSS received signal properties.

2 shows a schematic of an exemplary localization device 2 in a vehicle 1. The localization device 2 is set up to carry out the method described here.

The method described advantageously contributes to improving the GNSS position accuracy and, if possible, also the integrity of the position result.

Claims (10)

Method for GNSS-based localization of a vehicle (1) using a localization device (2) of the vehicle (1), comprising at least the following steps: a) receiving at least one GNSS signal (3) from at least one GNSS satellite (4) by means of at least one GNSS receiver (5) of the vehicle (1), b) reading in at least one item of statistical information from a GNSS profile relating to the reception situation on which the reception in step a) is based, the at least one piece of statistical information being suitable for describing at least one property of the received GNSS signal (3), c) determining at least one piece of localization information using the localization device (2), taking into account the at least one GNSS signal (3) received in step a) and the at least one piece of statistical information read in step b). procedure after claim 1 , wherein the at least one piece of statistical information is suitable for describing a quality of the received GNSS signal (3). procedure after claim 1 or 2 , wherein the at least one piece of statistical information comprises long-term statistical information. procedure after claim 3 , whereby the statistical long-term information was determined on the basis of chronologically different GNSS observations of the same reception situation. Method according to one of the preceding claims, wherein in step c) a localization method for determining the at least one piece of localization information is selected as a function of the at least one piece of statistical information. The method according to any one of the preceding claims, wherein in step c) depending on the at least one piece of statistical information, a measure for adapting the determination of the at least one piece of localization information is selected. Method according to one of the preceding claims, wherein the at least one piece of localization information comprises integrity information about the integrity of a localization result. Computer program for carrying out a method according to one of the preceding claims. Machine-readable storage medium on which the computer program claim 8 is saved. Localization device (2) for a vehicle (1), set up to carry out a method according to one of Claims 1 until 7 .

Images