Methods for improving position estimation

Abstract

A procedure for improving a position estimate with the following steps within a first period: - Estimating the position of a localization device (14) using a positioning system as a first estimate (S1); - Acquiring an environmental feature which has a known position using a sensing device (13) (S2); - Estimating a directed distance between the sensing device (13) and the environmental feature based on the acquisition of the environmental feature (S3); - Determining a second estimate for the position of the sensing device (13) based on the position of the environmental feature and the directed distance (S4); - Comparing the first estimate with the second estimate by calculating the difference between the first and second estimates and storing the comparison (S5); - Repeating the preceding steps within the first period;and with the following steps within a second period: - Estimating a changed position of the localization device (14) using a position system as a changed first estimate (S6); - Adjusting the changed first estimate based on at least one comparison between a first estimate and a second estimate (S7).;

Description

FIELD OF THE INVENTIONThe present invention relates to a method for improving a position estimate, a computing unit for improving a position estimate, and a data processing device for improving a position estimate. The invention further relates to a computer program product for executing a method according to the invention.BACKGROUND ARTU.S. Pat. No. 9,052,207 B2 shows a method for improving position estimation by means of known environmental features (SLAM). DE 10 2011 112 404 A1 shows a method for determining the position of a motor vehicle. DE 10 2016 002 704 A1 shows a control device and a method for determining a self-position of a motor vehicle. DE 10 2013 015 705 A1 shows a method for determining a current position of a motor vehicle in a geodetic coordinate system and a motor vehicle.SUMMARY OF THE INVENTIONAgainst this background, the object of the invention is to specify a method by means of which position estimation can be reliably improved.According to the invention, this object is achieved by a method having the features of patent claim 1 and / or by a computing unit for improving a position estimate having the features of patent claim 10.Accordingly, the following is provided:a method for improving position estimation, comprising the following steps within a first time period:estimating a position of a locating device by means of a position system as a first estimated value.detecting an environmental feature having a known position by means of a detection device.estimating a directed distance between the detecting means and the environmental feature based on the detection of the environmental feature. determining a second estimate for the position of the detection device based on the position of the environmental feature and the directed distance.comparing the first estimate with the second estimate. In the comparing, a difference is formed between the first and the second estimated values. The comparison is stored.repeating the above steps within the first period. Within a second time period, the following steps are carried out: estimating a changed position of the localization device by means of a position system. This estimate forms a modified first estimate. improving the modified first estimate value on the basis of at least one comparison between the first estimate value and the second estimate value.a computing unit for improving a position estimate, which is configured to receive a first estimated value for a position of a localization device by means of an interface between the computing unit and the localization device. The computing unit is further configured to estimate a directed distance between a detection device for detecting environmental features and a detected environmental feature. The computing unit is further configured to determine a second estimate for a position of the detection device based on a position of the environmental feature and the directed distance. The computing unit is further configured to compare the first estimated value for the position of the localization device with the second estimated value. The computing unit is further configured to receive a modified first estimated value for a position of the localization device by means of the interface. The computing unit is further configured to improve the changed first estimated value on the basis of at least one comparison between the first estimated value and the second estimated value.An estimate is the approximate determination of numerical values, variables or parameters by eye appearance, experience or calculation-based methods. The result of an estimation generally deviates from a true value. The true value can only be determined with countable values. On the other hand, measurements are subject to errors.In this application, a directed distance is understood to mean a directed distance, i.e. a vector in three spatial directions.An interface is a device between at least two functional units, at which an exchange of logical variables, e.g. data, or physical variables, e.g. electrical signals, takes place, either only unidirectionally or bidirectionally. The exchange can be analog or digital. The exchange can furthermore take place in a wired or wireless manner.A control unit is an electronic module for open-loop or closed-loop control. Control units are used in the motor vehicle sector in all conceivable electronic sectors, likewise for controlling machines, systems and other technical processes.Detection of an object is a recording of an object. For example, capturing a building with a camera is a capturing of the building.Environmental features are objects, preferably widely visible objects, that can be used for navigation. For example, traffic signs, buildings in the vicinity of a road or mountain are environmental features. Going back to aircraft and ship travel, an environmental feature is also referred to as a landmark. In particular, an environmental feature is an object located on the surface of the earth.A data processing apparatus is an apparatus that processes incoming information and outputs a result resulting from this processing. Electronic circuits such as central processing units or graphics processors are data processing devices.Position systems are used for position estimation, i.e. for estimating a location with respect to a defined fixed point. An example of a positioning system is GPS (global positioning system).The determination of positions on the basis of satellite positioning systems, such as GPS, is effected by evaluating signal transit times between the satellites to be received and the respective receiver to be positioned on the ground. The measurement of the signal transit times is subject to random and systematic fluctuations or errors. For example, the number and arrangement of the visible satellites, the changing conditions in the atmosphere and the construction on the ground will affect the measurement. As a result, it is not possible to determine an accurate point on the earth, but an area in which this point is likely to be located. The more strongly the negative effects act, the greater is this area too, i.e. the more inaccurate the determination of the sought coordinate becomes. Therefore, the satellite positioning used hitherto is often not sufficient for position estimation.The position determination in relation to environmental features is also referred to as silmutaneous localization and mapping, abbreviated to SLAM.A locating device is a device for estimating a position. For example, a GPS receiver is a locating device.It is understood that the localization device and the detection device do not necessarily have to be contained in a module, but can also be formed separately from one another. In both cases, it is advantageous if a distance between the localization device and the detection device is known and is taken into account for comparisons between position estimates of the localization device and the detection device.The basic idea of the invention is to observe an error estimate with respect to a position estimate within a first period of time and to improve the position estimate due to the observed error estimate within a second period of time in which no error estimate is possible.Advantageous embodiments and refinements emerge from the further dependent claims and from the description with reference to the figures of the drawing.The assignment of a captured environmental feature to the position of a known environmental feature, which is stored in a database, for example, can be effected by the first estimated value, for example.Alternatively, it is conceivable for the environmental feature to have an assignment means, for example a numerical code, by means of which the environmental feature can be assigned to a known position.According to a preferred development of the invention, it is detected within the second time period that an environmental feature having a known position cannot be detected or that the estimated directed distance is subject to an error value, wherein the error value exceeds a threshold value.Accordingly, the method is particularly applicable when a known environmental feature is not detectable or the quality of the detection is too low for an improvement of the position estimation. For example, the method may be applied in driver assistance systems of vehicles in order to improve a position estimate of a vehicle.If a vehicle is in a curve, it is not excluded that a known environmental feature is temporarily concealed and cannot be detected by the detection device.A vehicle is in particular a ground-bound vehicle, for example a car, truck and / or a commercial vehicle. Further vehicles are aircraft, spacecraft, watercraft or the like.A driver assistance system is an electronic additional device in a vehicle for assisting the driver in specific driving situations.Furthermore, a weather-related change in the environment may prevent or impair the detection of a known environmental feature. This can occur, for example, in mist sinks.Thus, a position estimate of a vehicle can be further specified even if temporarily no environmental feature is detectable or the quality of the detection temporarily is not sufficient for a specification.According to a preferred development of the invention, an own movement of the localization device is determined and the adaptation of the changed estimated value in the second time period is based on the determined an own movement of the localization device.Thus, for example, the own movement of a vehicle during a trip can be determined and used for improving a position estimate.According to a preferred development of the invention, a profile of comparison values between the first and the second estimated value is determined and the modified first estimated value is adapted on the basis of the profile.It is particularly advantageous in this case to compare first estimated values with the second estimated values on the basis of differences. However, it is understood that the estimated values can also be compared using other comparison methods, for example weighted differences or the like.Accordingly, it can be provided that a position estimate value, which has been determined, for example, by means of a GPS system, is compared with a position estimate value, which has been determined from the known position of an environmental feature and a directed distance between the detection device and the environmental feature, by forming a distance between the two position vectors.It is furthermore advantageous if the modified estimated value is adapted on the basis of an extrapolation of the differences, i.e. the distances of the position vectors.Thus, an estimated error value of the first estimate of the position system may be observed and extrapolated within the second time period if detection of a known environmental feature is not possible or too inaccurate.Since the position estimation is affected by certain environmental influences by means of position systems, for example GPS, the course of an error of the first estimated value is predictable.According to a preferred development of the invention, a compensation calculation is applied to the determined differences, i.e. distances between the position vectors of the first and the second estimated value, wherein in particular the mean value of the differences is formed, and the changed first estimated value is adapted on the basis of the compensation calculation.Compensation calculation is an optimization method with the aid of which the unknown parameters of a model or the parameters of a predefined function can be determined or estimated for a series of measurement data. One method of compensation calculation is the least squares method. In this case, a curve is sought for a plurality of data points, which curve runs as close as possible to the data points.The method for improving a position estimation can thus be further refined.According to a preferred development of the invention, a method for estimating the directed distance is adapted on the basis of properties, in particular a variance or an accuracy of the detection of the environmental feature. Alternatively, the method for estimating the directed distance can also be adapted on the basis of properties, in particular a variance or an accuracy of the estimated directed distance.A detection is, for example, an image recording.Variance is a measure of the spread of data. Thus, data values of a data set with a greater variance vary more strongly than data values of a data set with a lesser variance.The accuracy of detection is understood to mean the resolution of the detection. If the detection is an image, the accuracy of the image is understood to mean its image resolution. If the detection can be expressed as a numerical value, the accuracy is dependent on the number of the decimal point positions.The variance of a detection is understood to mean the variance of a property over a data record of the detection. For example, in the case of image recordings in gray levels, the variance of the gray level of the image recording is considered.Alternative characteristics of a detection are maximum values or minimum values of a parameter of the detection, for example a maximum value of the gray value of an image recording.This is advantageous because both the detection of the environmental feature and the determination of the directed distance are subject to inaccuracies. For example, the data resolution of the detection of the environmental feature, for example an image quality of an image recording of the environmental feature, is decisive for the quality of the estimated directional distance. Furthermore, a directed distance between the environmental feature and the detection device can also be determined only with errors on the basis of the detection of the environmental feature.Accordingly, the directed distance estimation method may be adjusted based on the data resolution of the detection of the environmental feature or based on an accuracy of the estimated directed distance.For example, it is conceivable to extend the detection duration of the surrounding feature if it is determined that the data resolution of the detection of the surrounding feature is low.According to a preferred development of the invention, properties of the detection of the environmental feature or of the estimated directional distance are used for an adaptation of the changed estimated value of the second time period.Thus, for example, different estimated values of the first time period can be weighted depending on a variance or an accuracy of the detection of a respective environmental feature or the estimated directed distance.According to a preferred development of the invention, a data processing device is configured to improve position estimation. The data processing device has a localization device for estimating a position of the localization device by means of a position system, in particular GPS. The data processing device further has a detection device for detecting an environmental feature, in particular an image sensor, a lidar / and / or radar sensor. The data processing device further comprises a computing unit according to claim 10. The data processing device further comprises a control unit which is configured to generate control signals on the basis of the adapted estimate value. The data processing device further comprises at least one interface which is configured to transmit data between the localization device, the detection device, the computing unit and the control unit.Accordingly, a control unit, in particular in a driver assistance system, is configured to generate control signals which are based on the ascertained estimated values or on properties which were ascertained during the method for improving a position estimate. For example, it is conceivable to control a driving movement on the basis of the estimated values or to direct a signal which is directed to a driver on the basis of estimated values or properties which were determined during the method for improving a position estimate.The computer program product according to one embodiment of the invention executes the steps of a method according to the preceding description when the computer program product runs on a computer, in particular an in-vehicle computer.Computer program products typically comprise a sequence of instructions which, when the program is loaded, cause the hardware to perform a specific method which leads to a specific result. When the program in question is used on a computer, the computer program product causes an effect, namely the improvement of a position estimate of a locating device.CONTENT SPECIFICATION OF THE DRAWINGSThe present invention is explained in more detail below with reference to the exemplary embodiments indicated in the schematic figures of the drawings. The following are shown: FIG. 1 is a schematic block diagram of an embodiment of the invention; FIG. 2 is a schematic block diagram of an embodiment of the invention; FIG. 3 is a schematic block diagram of an embodiment of the invention;The accompanying drawings are intended to provide a further understanding of the embodiments of the invention. They illustrate embodiments and, in conjunction with the description, serve to explain principles and concepts of the invention. Other embodiments and many of the advantages mentioned are evident with reference to the drawings. The elements of the drawings are not necessarily shown to scale with respect to each other.In the figures of the drawings, identical, functionally identical and identically functioning elements, features and components are provided with the same reference numerals, unless stated otherwise.DESCRIPTION OF EMBODIMENTSFIG. 1 shows a schematic block diagram of a method according to an embodiment of the invention. The method according to FIG. 1 comprises steps S 1-S 7. Step S 1 and the sequence of steps S 2-S 4 can proceed sequentially or simultaneously. Step S 1 does not necessarily have to be executed before steps S 2-S 4.In step S 1, the position of a localization device 14 is estimated by means of a position system. The estimate forms a first estimate. The position system can be designed, for example, as a GPS (Global Positioning System). In this case, the locating device is designed as a GPS receiver. Accordingly, the position of a GPS receiver is estimated by GPS. The accuracy of GPS is about a few meters.In step S 2, an environmental feature having a known position is detected by means of a detection device 13. The environmental feature can be, for example, a road marking. In this case, it can be provided that the environmental feature has a key, for example a numerical code, when the environmental feature is detected as an image. The captured environmental feature is assigned to a position of the environmental feature by means of the key. Alternatively, the assignment can be made on the basis of the position estimate according to step S 1 if exactly one known environmental feature is located in a circle around the first estimate.In step S 3, a directed distance between the detection device 13 and the surrounding feature is estimated on the basis of the detection of the surrounding feature S 3. For example, the detection can be embodied as image recording or radar detection. Depending on a quality or data resolution of the detection, provision can be made for the environmental feature to be detected multiple times or for a specific duration.In step S 4, a second estimate for the position of the detection device 13 is determined based on the position of the surrounding feature and the directed distance. The second estimate may be determined, for example, by adding a position vector of the environmental feature to a vector of the directed distance.In step S 5, the first estimated value is compared with the second estimated value. The estimated values can be compared, for example, by forming in particular a difference between the first and the second estimated value. The comparison value between the estimated values is stored.The first estimated value is an estimated value of the position of the localization device. The second estimated value is an estimated value of the position of the detection device. If the locating device and the detection device are not located at an identical position, a distance between the locating device and the detection device can be taken into account.These steps S 1 to S 5 are repeated until a known environmental feature cannot be detected. If a known environmental feature cannot be detected, the method continues with step S 6.In step S 6, a changed position of the localization device is estimated by means of a position system. The estimated value is stored as a modified first estimated value. It can be provided that step S 6 corresponds to step S 1, wherein the position of the localization device has changed.In step S 7, the modified first estimated value is adapted on the basis of at least one comparison between a first estimated value and a second estimated value. This step is based on the assumption that the method by means of steps S 5 and S 6 could already obtain findings about the accuracy of the first estimated value or of the changed first estimated value. These findings are applied to the changed first estimated value in step S 7.FIG. 2 shows a computing unit 11.FIG. 3 shows a data processing device 10 having a computing unit 11, a control unit 12, a detection device 13 and a localization device 14.Reference numerals denote reference numeralsS1-S7 Method steps 10 Data processing apparatus 11 Computing unit 12 Control unit 13 Detection device 14 Localization device 15 Interface

Claims

Method for improving a position estimate, comprising the following steps within a first time period: - estimating a position of a locating device (14) by means of a position system as a first estimated value (S1); - detecting an environmental feature, which has a known position, by means of a detection device (13) (S2); - estimating a directed distance between the detection device (13) and the environmental feature on the basis of the detection of the environmental feature (S3); - determining a second estimated value for the position of the detection device (13) on the basis of the position of the environmental feature and the directed distance (S4); - comparing the first estimated value with the second estimated value by forming a difference between the first and the second estimated values and storing the comparison (S5); - repeating the above steps within the first time period; and having the following steps within a second time period: - estimating a changed position of the localization device (14) by means of a position system as a changed first estimated value (S 6); - adapting the changed first estimated value on the basis of at least one comparison between a first estimated value and a second estimated value (S 7).The method of claim 1, wherein within the second time period it is detected that an environmental feature having a known position is not detectable, or it is detected that the estimated directional distance is subject to an error value that exceeds a threshold value.Method according to one of the preceding claims, wherein an own movement of the locating device (14) is determined and the adaptation of the changed estimated value in the second time period is based on the determined own movement of the locating device (14).Method according to one of the preceding claims, wherein a profile of the comparisons, in particular of the differences, is formed and the modified first estimated value is adapted on the basis of the profile.The method of claim 4, wherein the altered estimate is adjusted based on extrapolation of the comparisons.Method according to one of the preceding claims, wherein a compensation calculation is applied to the determined differences, wherein in particular the mean value of the differences is formed, and the modified first estimated value is adapted on the basis of the compensation calculation.Method according to one of the preceding claims, wherein properties, in particular a variance or an accuracy, of the detection of the environmental feature and / or properties, in particular a variance or an accuracy, of the estimated directed distance are determined and a method for estimating the directed distance is adapted on the basis of these properties.The method of claim 7, wherein a duration of detecting an environmental feature is adjusted based on the accuracy of the estimated directional distance.Method according to one of the preceding claims, wherein properties, in particular a variance or an accuracy, of the detection of the environmental feature and / or properties, in particular a variance or an accuracy, of the estimated directed distance are determined and the changed estimated value of the second time period is adapted on the basis of these properties.Computing unit (11) for improving a position estimate, which is configured to receive a first estimated value for a position of a locating device (14) by means of an interface (15) between the computing unit and the locating device (14), to estimate a directed distance between a detection device (13) for detecting environmental features and a detected environmental feature, to determine a second estimated value for a position of the detection device (13) based on a position of the environmental feature and the directed distance, to compare the first estimated value for the position of the locating device (14) with the second estimated value, to receive a changed first estimated value for a position of the locating device (14) by means of the interface (15), and to improve the changed first estimated value on the basis of at least one comparison between the first estimated value and the second estimated value.Data processing device (10) for improving a position estimate, having a locating device (14) for estimating a position of the locating device (14) by means of a position system, in particular GPS, a detection device (13) for detecting an environmental feature, in particular an image sensor, a lidar / and / or radar sensor, a computing unit (11) according to Claim 10, a control unit (12) which is configured to generate control signals on the basis of the adjusted estimate, and having at least one interface (15) which is configured to transmit data between the locating device (14), the detection device (13), the computing unit and the control unit.A computer program product executing a method according to any one of claims 1 to 9 when the computer program product is run on a computing unit according to claim 10.

Images