Method for providing and improving position probability distribution for GNSS received data

Through the communication connection between mobile systems, the position probability distribution and spacing data are exchanged, and the correction terms are calculated, which solves the problem of low accuracy in position determination in aerospace and autonomous driving, and realizes positioning without adding hardware Improved accuracy and reduced cost.

CN110945382BActive Publication Date: 2024-12-03ROBERT BOSCH GMBH
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN201880048950.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2017-07-21
Filing Date
2018-06-15
Publication Date
2024-12-03
Estimated Expiration
2038-06-15

AI Technical Summary

Technical Problem

The existing GNSS system has low accuracy in position determination in aerospace and autonomous vehicles, especially in the absence of additional hardware.

Method used

Through communication connections between at least two mobile systems, position probability distribution and spacing data are exchanged, and the existing sensors and GNSS receivers are used to calculate correction terms to improve the accuracy of position measurements, avoiding dependence on DGPS services.

Benefits of technology

It achieves improved the accuracy of GNSS positioning without adding hardware, optimizes the position determination of vehicles or mobile systems, reduces costs, and improves the accuracy of traditional navigation systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN110945382B_ABST
    Figure CN110945382B_ABST
Patent Text Reader

Abstract

A method for correcting a position probability distribution is disclosed, wherein at least two mobile systems respectively determine their own position probability distributions via a GNSS receiver, at least one mobile system determines a distance to at least one second mobile system, the at least two mobile systems exchange the determined position probability distributions with each other via a communication connection, and a correction term for improving the probability distribution is calculated using the at least two determined position probability distributions and the distance between the at least two mobile systems. In addition, a method for providing at least one correction term is disclosed.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to a method for correcting a position probability distribution and to a method for providing at least one correction term. Background Art

[0002] Precise position determination is required for many applications. For example in aerospace and in autonomous vehicles, the position must be determined as accurately as possible. However, the civil mode standard of GNSS (Global Navigation Satellite System) such as the Global Positioning System (GPS) can only achieve relatively low lateral accuracy (about 13m) and vertical accuracy (about 22m).

[0003] In aviation, for example, two GNSS receivers that are offset from one another are used to increase the accuracy of position determination. To this end, one GNSS receiver is arranged at the front of the aircraft and one at the rear of the aircraft, and the accuracy of the position determination is optimized based on the fixed and known spacing of the GNSS receivers from one another.

[0004] Furthermore, the accuracy of the position determination can be improved by means of differential GPS (DGPS). For this purpose, a DGPS station, whose position has been measured in advance, transmits a locally effective correction signal with correction terms. This allows errors, for example, caused by satellite clock drift or atmospheric influences to be compensated. However, DGPS services are usually available at a cost and require receiving equipment provided for this purpose. Summary of the invention

[0005] The object on which the present invention is based can be seen as improving the accuracy of position measurement without additional hardware and providing a solution for such an improvement.

[0006] This object is achieved by means of the respective subject matter of the independent claims. Advantageous embodiments of the invention are the subject matter of the respective dependent claims.

[0007] According to one aspect of the present invention, a method for correcting a position probability distribution is provided. In a first step, a position probability distribution for at least two mobile systems is determined by at least two mobile systems each having a GNSS receiver. In a further step, the distance between the at least two mobile systems is determined. The at least two mobile systems can exchange the determined position probability distributions with each other via a communication connection. Subsequently, the position probability distribution can be improved using the at least two determined position probability distributions and the distance between the at least two mobile systems.

[0008] By determining a fixed or variable distance between a movable object (e.g. a vehicle) and another movable or stationary object, this distance can be used as a more accurate reference variable for comparing or matching position probability distributions. As a result, GNSS-based positioning can be improved, for example, by exchanging data from vehicle to vehicle. Various data determined by one or more vehicles (e.g. distances to other traffic participants, distances of multiple vehicles to objects with unknown positions or to public objects with known positions) can be used in particular to optimize the position determination of vehicles or mobile systems. In this case, sensors and GNSS receivers already present in the mobile system can be used for positioning and improving the position determination. Therefore, no additional hardware is required, for example using DGPS.

[0009] In order to carry out the method, each vehicle has a local estimate of its position in the form of a probability distribution. The position in the form of a probability distribution is a position probability distribution obtained based on a GNSS measurement previously performed by a mobile system. In order to obtain the position probability distribution, other sensors of the mobile system (such as distance sensors or acceleration sensors) can also be used auxiliaryly. Once a communication connection can be established between multiple mobile systems, the mobile system exchanges its position estimate together with the estimated clock drift to the observed satellite and the measured GNSS signal. The GNSS signal or GNSS measurement can, for example, have a three-dimensional position, the time of sending the signal, and the local reception time of the mobile system. The sensor data obtained by each mobile system can also be transmitted via the communication connection. For example, the mobile systems can respectively obtain the distance between each other and then exchange them with each other. In this way, the individual measurement uncertainties of the sensors can be compensated. The distance between the mobile systems can also be obtained by the first mobile system and transmitted to another mobile system or the second mobile system. Subsequently, the exchanged or transmitted data can be used together with the obtained position probability distribution of each mobile system to calculate a correction term for correcting or optimizing the position probability distribution. In a further step, the ascertained correction term can be used for a subsequent calculation or measurement of the position of the mobile system.

[0010] The improvement of the GPS position enables new driving functions that require accurate positioning of the lane, for example. In addition, the accuracy of conventional navigation systems can be improved by this method.

[0011] According to an embodiment of the method, the position probability distribution is determined by a GNSS receiver of a vehicle or a portable device. The mobile system can be, for example, a vehicle, a portable device or a fixedly arranged device. The portable device can also be fixedly arranged, for example. In addition, the portable device can also be arranged in a vehicle and used to perform the method. For example, a navigation device, a mobile radio device, a computer, a tablet or the like can be used as a portable device.

[0012] According to another advantageous embodiment of the method, the position probability distribution determined by the position estimator is improved by at least one sensor. Each mobile system can use its existing sensors to detect its external surroundings. The sensor data thus determined can be used to improve the position probability distribution. If, for example, a communicating mobile system detects another mobile system or a second mobile system by means of its sensor device (e.g., radar or camera), the communicating mobile system can accurately determine the relative position of the GNSS receiver of the second mobile system relative to its own GNSS receiver, and use this distance and the received GNSS measurements of at least one other mobile system and its own GNSS measurements to estimate the GNSS correction term. The correction term thus determined can be shared with all mobile systems within the range of action. This is possible because it is assumed that the GNSS errors in the immediate surroundings of the mobile system are almost constant.

[0013] According to another embodiment of the method, at least two mobile systems determine the distance between each other or estimate the distance between each other by means of at least one sensor. For example, the distance from a mobile system to a second mobile system can be determined by means of a radar sensor, a LIDAR (English: licht detection and ranging) sensor, an ultrasonic sensor or the like. In this case, the distance between the at least two mobile systems can change during the measurement or remain unchanged.

[0014] According to another embodiment of the method, locally determined position data are exchanged via a communication connection of at least two mobile systems and used to calculate correction terms. Each mobile system in the communication connection can measure or estimate the distance to another mobile system or to a defined object in its surroundings, for example, via its own sensors. With these sensor data, the mobile system can improve its own position probability distribution according to the calculated correction terms. The determined sensor data or correction terms can be provided to the other mobile system within a limited range.

[0015] According to another advantageous embodiment of the method, the communication connection is a radio connection. The communication connection can be in particular a mobile radio connection, a wireless vehicle-to-vehicle connection or the like. For example, UMTS, LTE, WLAN, Bluetooth or the like can be used as a transmission standard.

[0016] According to another embodiment of the method, in order to determine the position probability distribution and the distance by at least two vehicles, the distance between each other is adjusted. Alternatively or in addition to measuring the distance between at least two mobile systems, the defined distance between the mobile systems can also be adjusted, for example, by a distance keeping assistant (Abstandshalteassistenten). For example, vehicles communicating with each other can automatically match their relative distance and their speed, so that the calculation of local correction items can be optimized. Therefore, a fixed distance value can be assumed when calculating the correction items. For example, the cooperative function of the vehicles can be improved in this way. The cooperative function is designed to automatically implement potentially dangerous common driving maneuvers of multiple vehicles, for example, due to the automatic merging of traffic on the main lane into the highway or the automatic widening of the gap.

[0017] According to another aspect of the invention, a method for providing at least one correction term is disclosed. Here, the correction term is calculated by the method according to the previous aspect of the invention. In a first step, at least one correction term is transferred to a central server unit. The transferred correction term is then assigned to a location or updated at the corresponding location. Depending on the location of the mobile system, the correction term corresponding to the location or location of the mobile system can be transmitted from the central server unit to the mobile system.

[0018] The correction items obtained by multiple mobile systems can be transmitted to the central server unit in real time or at a certain time interval. In addition to the correction items transmitted, the position of the corresponding mobile system can also be transmitted to the server unit. Based on these data, a map can be created, which can provide position-related correction items for any or certain mobile systems. This can be achieved in particular by associating the correction items with the corresponding positions. For example, this method can be used to provide the following service: distribute correction values ​​to other traffic participants through a central server. The correction items and the corresponding positions or position probability distributions can be transmitted to the central server unit, for example, via a mobile radio network or the Internet. It is also possible to provide correction items to other participants or mobile systems in the form of wireless data transmission through a cellular network or through a wired communication network. Compared with the DGPS service, local obstacles and obstructions (Abschattung) caused by buildings, for example, can be reduced or preferably avoided by this method. For inquiries of mobile systems, the correction items stored in the central server unit can be transmitted to the mobile system that is inquiring. In particular, correction items can be provided based on the position of the mobile system. Compared to conventional DGPS systems, no dedicated receiving equipment is required in a vehicle or mobile system, so the GNSS receiving hardware in the vehicle can be kept cost-effective, and the operation of a fixed-position DGPS base station can be omitted.

[0019] According to an advantageous embodiment of the method, correction items for at least one second location are estimated based on at least one correction item transmitted for a first location. A limited number of locally valid correction items or correction items valid only at one location point can also be interpolated to other areas of the map. Once the estimation or calculation of the GNSS correction items is achieved, these GNSS correction items can be transmitted to a central server. The central server unit collects correction items from different mobile systems and uses these correction items to form a local distribution of GNSS correction items. Various schemes can be implemented for this purpose. For example, a Gaussian process can be used. A Gaussian process trained with the data obtained can be used to estimate the GNSS correction items for locations where correction items have not yet been transmitted. The great advantage of using a Gaussian process is that the values ​​it provides for the GNSS correction items are Gaussian distributions, which in addition to the mean also provide covariances, i.e., quantitative numbers of the quality of the interpolated GNSS correction items. Changes in the correction items over time can also be taken into account here. This can be reflected, for example, in the following way: the covariance of the received correction items increases over time, which can explain the reduction in the weight of the received correction items when training the Gaussian process. As an alternative to using a Gaussian process, the problem under consideration of estimating the local distribution of the GNSS correction terms can be understood as a distribution parameter phenomenon, the parameters of which can be ascertained by means of Bayesian estimation.

[0020] According to another embodiment of the method, at least one calculated correction term is transmitted from at least one vehicle to a server unit. The mobile system that transmits the correction term and the corresponding position of the correction term to the central server unit can preferably be a vehicle. The vehicle can, for example, have a wireless Internet connection to the central server unit and can send and receive the correction term via the wireless Internet connection.

[0021] According to another embodiment of the method, at least one correction term and at least one position are calculated by at least one distance to at least one landmark. Instead of or in addition to the currently described scheme for collaboratively improving GNSS positioning by means of estimating GNSS correction terms and providing a map of locally valid GNSS correction terms, the static surroundings of the mobile system can advantageously be taken into account. For example, passive and stationary landmarks with known positions (such as radar mirrors or optical markers, such as ArUco markers) can be used to determine the distance and position to calculate the correction terms. Alternatively or additionally, even without using a GNSS receiver, the own position of the mobile system or vehicle can be determined, for example, by triangulation based on the known position of the landmark. This position does not require further correction and can be used, for example, in the context of a communication connection with other mobile systems to calculate the correction terms. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The preferred embodiment of the present invention is described in detail below based on an extremely simplified schematic diagram. It is shown here:

[0023] Figure 1 A schematic flow chart of a method for providing at least one correction term according to a first embodiment is shown;

[0024] Figure 2 A schematic flow chart showing a method for correcting position probabilities according to a first embodiment;

[0025] Figure 3 A schematic flow of a method for correcting position probability distribution according to the second embodiment is shown.

[0026] In the figures, identical components are provided with the same reference numerals. DETAILED DESCRIPTION

[0027] Figure 1 The schematic flow of a method 1 for providing at least one correction term according to a first embodiment is shown. A first mobile system 2 or a first vehicle 2 communicates with a second mobile system 4 or a second vehicle 4 via a wireless communication connection. The two vehicles 2, 4 determine their respective position probability distributions via GNSS. The second vehicle 4 determines a distance 12 to the first vehicle 2 by means of a distance keeping aid and transmits this distance 12 to the first vehicle 2. The two vehicles 2, 4 can thus correct their respective position probability distributions with respect to each other based on their position probability distributions and the more accurately determined distance 12 of the two vehicles 2, 4.

[0028] According to this embodiment, the second vehicle 4 measures the relative position of the first vehicle 2 by means of an onboard sensor device. Subsequently, the position of the first vehicle 2 is determined by the calculated position of the second vehicle 4 and the measured relative position. Consider the following case: the deviation between the signal propagation speed and the nominal speed of light due to the influence of the atmosphere is the only error source. This assumption is reasonable because this type of influence is usually the most significant. In addition, multipath propagation can also be modeled as a variable of the propagation speed. Now let are the unknown coordinates of the second vehicle 4, is the measured relative position of the first vehicle 2, is the coordinate of satellite i, b1 and b2 are the local vehicle clock drifts, d ij is the measured time difference between the transmission of the GNSS signal by satellite j and the reception of the signal by vehicle i, and finally e j is the deviation between the GNSS signal transmission speed of satellite j and the nominal light speed c. Then the following equations can be established:

[0029]

[0030]

[0031] In the case of two vehicles communicating with each other via a communication link, the equation is overdetermined for five observed satellites and can be solved, for example, by means of the least squares method. In principle, the following inequality can be obtained:

[0032] mn≥m+n+3

[0033] Where m is the number of cooperating vehicles and n is the number of observed satellites. If this inequality is satisfied, the set of equations listed above can be solved. For example, if four vehicles cooperate with each other via corresponding communication connections, two satellites are sufficient for GNSS position determination and for calculating the clock drift of all participating vehicles and for calculating the correction terms. If not enough satellites are observed, either the correction terms from the previous calculation can be used or the conventional GNSS equations without correction terms can be used. Furthermore, the use of recursive and non-recursive estimation methods can improve the results.

[0034] The vehicles 2, 4 can use the determined correction terms to optimize their position determination and can transmit this position determination to the central server unit 6, for example, via an Internet connection 8. The calculated correction terms are assigned to the corresponding positions or position probability distributions of the vehicles 2, 4 on the central server unit 6. Based on a plurality of correction terms and the corresponding positions, a map for correcting the position probabilities can be created and made available. Depending on the position of another vehicle 10, for example, the determined correction values ​​can be sent to this vehicle 10 in order to optimize the position determination of this vehicle.

[0035] Figure 2 The schematic flow of a method 11 for correcting position probabilities according to a first exemplary embodiment is shown. The two vehicles 2, 4 exchange the determined position probabilities with one another via a communication connection. The second vehicle 4 measures the distance 12 between the first vehicle 2 and the second vehicle 4 via a distance sensor. This distance 12 can also be transmitted to the first vehicle 2 via a communication connection. Based on the accurate measurement of the distance 12, the vehicles 2, 4 can improve or calibrate their position determination and can limit or optimize the determined position probability distribution.

[0036] Figure 3A schematic flow chart of a method for correcting a position probability distribution according to a second embodiment is shown. In particular, a landmark 14 recognized by a first vehicle 2 is identified, and the distance between the first vehicle 2 and the landmark 14 is measured. In this example, the second vehicle 4 is not yet able to recognize the landmark 14 because the distance from the vehicle 4 to the landmark 14 is too large. Here, the distance from the landmark 14 to the first vehicle 2 can improve the position probability distribution of the first vehicle 2. Based on the distance from the second vehicle 4 to the first vehicle 2 and then based on another distance to the landmark 14, the second vehicle 4 can also improve its position probability distribution. The second vehicle can, in particular, indirectly recognize the landmark 14 via a communication connection with the first vehicle 2 and thus increase its range of action.

Claims

1. A method (11) for correcting a position probability distribution, wherein: At least two mobile systems (2, 4) obtain position probability distributions respectively through a GNSS receiver each; At least one movement system (4) determines a distance (12) to at least one second movement system (2); The at least two mobile systems (2, 4) exchange the determined position probability distributions with each other via a communication connection; It is characterized in that Using the at least two determined position probability distributions and the distance (12) between the at least two mobile systems (2, 4), improving the position probability distributions of the at least two mobile systems (2, 4), The at least two mobile systems correct the position probability distribution of each of the at least two mobile systems based on the position probability distribution of each of the at least two mobile systems and the obtained distance between the mobile systems.

2. The method according to claim 1, wherein: The position probability distribution is ascertained by a GNSS receiver of the vehicle or of the portable device.

3. The method according to claim 1 or 2, wherein: The position probability distribution ascertained by the position receiver is improved by at least one sensor.

4. The method according to claim 1 or 2, wherein: The at least two mobile systems (2, 4) determine or estimate a distance (12) from one another by means of at least one sensor.

5. The method according to claim 1 or 2, wherein: The at least two mobile systems (2, 4) exchange locally ascertained position data via the communication connection and use the position data for calculating correction terms.

6. The method according to claim 1 or 2, wherein: The communication link is a vehicle-to-vehicle communication link or a radio link.

7. The method according to claim 1 or 2, wherein: In order to determine the position probability distribution and the distance (12) by the at least two vehicles, the distances to one another are set.

8. A method (1) for providing at least one correction term, wherein: The correction term is calculated by a method (11) for correcting a position probability distribution according to any one of the preceding claims, the method comprising the following steps: transmitting the at least one correction term to a central server unit (6); assigning the at least one correction term to a location or updating the at least one correction term at the location; The at least one correction term is transmitted to at least one mobile system (10) depending on the position of the mobile system (10).

9. The method according to claim 8, wherein: At least one correction term for at least one second location is estimated based on the communicated at least one correction term for the first location.

10. The method according to claim 8 or 9, wherein: The calculated at least one correction term is transmitted from at least one vehicle to the central server unit (6).

11. The method according to claim 8 or 9, wherein: The at least one correction term is calculated using at least one position probability distribution and using at least one distance to at least one landmark (14).

Citation Information

Patent Citations

  • Measurement Level Integration of GPS and Other Range and Bearing Measurement-Capable Sensors for Ubiquitous Positioning Capability

    US20100164789A1

  • Method and system for improving accuracy of position correction data in differential global positioning system using vehicle to vehicle communication

    US20130093618A1

  • Vehicle positioning apparatus and method

    US20160018527A1