Method for error evaluation in position determination

Abstract

The invention relates to a method for error evaluation in position determination, comprising time-synchronous recording of first and second position values, wherein the second position values are recorded by a different measuring method than the first position values; forming a first and a second trajectory (A, B) from the first and the second position values respectively; forming differential vectors (D) between first and second position values recorded at the same time; parallel-shifting the second trajectory (B) along a displacement vector (s) such that the amounts of the differential vectors (D) are minimized on average; evaluating the faultiness of the position determination on the basis of one or more amounts of the differential vectors (D) created as a consequence of the parallel shift.

Claims

1. A method for error evaluation in position determination, comprising: recording of first and second position values, wherein the second position values are recorded by a different measuring method than the first position values; temporal synchronizing of the first and second position values and forming a first and a second trajectory (A, B) each from the synchronized first and second position values; forming differential vectors (D) between the respectively synchronized first and second position values; parallel-shifting the second trajectory (B) along a displacement vector (s) such that the amounts of the differential vectors (D) are minimized on average; evaluating the faultiness and/or integrity of the position determination on the basis of one or more amounts of the differential vectors (D) created as a consequence of the parallel shift; and wherein the respective measuring method corresponds to one or a combination of the following: runtime measurement, Doppler measurement, carrier phase measurement, and/or motion estimation by means of cameras, radar, and/or lidar.

2. The method according to claim 1, wherein the measuring methods for recording the first and second position values differ in that they have a different influenceability of the measurement by errors, in particular by errors in GNSS satellites and/or by GNSS multipath propagation and/or by a lack of available satellites and/or by drifts and/or offsets in inertial and/or odometry sensor technology and/or by errors of environment sensor technology.

3. The method according to claim 2, wherein the respective measuring method corresponds to at least: acceleration, rotation rate and/or steering angle measurement.

4. The method according to claim 3, wherein the faultiness of the position determination is evaluated on the basis of the amount of the differential vector which extends between the first and second position values recorded at the last time.

5. The method according to claim 4, wherein the faultiness of the position determination is evaluated on the basis of the amount of the differential vector which has the largest amount value among multiple weighted differential vectors (D).

6. The method according to claim 5, wherein the faultiness of the position determination is evaluated on the basis of the sum or quadratic sum of the amounts of multiple weighted differential vectors (D) or of a specified number of most recently formed differential vectors (D).

7. The method according to claim 6, wherein the evaluation of the faultiness is made available as an input variable to a fusion filter, in particular Kalman filter, for continual determination of a position from multiple measuring variables.

8. A control device for evaluating error in position determination, by performing operations comprising; recording of first and second position values, wherein the second position values are recorded by a different measuring method than the first position values; temporal synchronizing of the first and second position values and forming a first and a second trajectory (A, B) each from the synchronized first and second position values; forming differential vectors (D) between the respectively synchronized first and second position values; parallel-shifting the second trajectory (B) along a displacement vector(s) such that the amounts of the differential vectors (D) are minimized on average; evaluating the faultiness and/or integrity of the position determination on the basis of one or more amounts of the differential vectors (D) created as a consequence of the parallel shift; and wherein the respective measuring method corresponds to one or a combination of the following: runtime measssurement, Doppler measurement, carrier phase measurement, and/or motion estimation by means of cameras, radar, and/or lidar.

9. The control device according to claim 8, having a non-transitory computer-readable memory and a processor, wherein the computer-readable instructions are saved in the non-transitory memory in the form of a computer program, and the processor is configured to carry out the computer-readable instructions when the computer program is loaded from the memory into the processor.

10. The control device according to claim 9, wherein the measuring operations for recording the first and second position values differ in that they have a different influenceability of the measurement by errors, in particular by errors in GNSS satellites and/or by GNSS multipath propagation and/or by a lack of available satellites and/or by drifts and/or offsets in inertial and/or odometry sensor technology and/or by errors of environment sensor technology.

11. The control device according to claim 10, wherein the respective measuring method corresponds to at least: acceleration, rotation rate, and/or steering angle measurement.

12. The control device according to claim 11, wherein the faultiness of the position determination is evaluated on the basis of the amount of the differential vector which extends between the first and second position values recorded at the last time.

13. The control device according to claim 12, wherein the faultiness of the position determination is evaluated on the basis of the amount of the differential vector which has the largest amount value among multiple weighted differential vectors (D).

14. The control device according to claim 13, wherein the faultiness of the position determination is evaluated on the basis of the sum or quadratic sum of the amounts of multiple weighted differential vectors (D) or of a specified number of most recently formed differential vectors (D).

15. The control device according to claim 14, wherein the evaluation of the faultiness is made available as an input variable to a fusion filter, in particular Kalman filter, for continual determination of a position from multiple measuring variables.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 schematically represents a parallel shifting of a second trajectory along a displacement vector.

(2) FIG. 2 schematically depicts differential vectors formed between first position values and the parallel-shifted second position values assigned to the same times.

DETAILED DESCRIPTION

(3) FIGS. 1 and 2 show schematically, and for the sake of clarity, a 2-dimensional representation of a relative comparison of two trajectories.

(4) With the aid of the history of GNSS runtime measurements, an absolute motion trajectory of the vehicle or receiver can be represented in geographic coordinates. The absolute trajectory determined by means of runtime measurement can be transformed into any relative coordinate system. A relative motion trajectory of the vehicle can be calculated both from the history of GNSS Doppler measurements and from the history of measurements of the inertial sensor technology (IMU) and measurements of the odometry (ODO).

(5) In a first step, first and second position values are recorded or scanned, possibly with a different scanning rate. The first position value assigned to a determined time and the second position value assigned to this time are subsequently recorded. This is repeated for a suitable number of times, preferably with the same temporal interval, so that a trajectory can be formed from the thus synchronized first position values and second position values respectively which, depending on the temporal intervals of the samples and under the influence of certain deviations, reproduces the temporal position progression, that is to say the motion, in each case with a certain accuracy.

(6) The first position values and, therefore, the first trajectory A are recorded, for example, by means of runtime measurement. The second position values and, therefore, the second trajectory B are recorded, for example by means of Doppler measurement. Consequently, one and the same multipath propagation has a different effect on the first and the second trajectory. It would additionally be possible to record a further trajectory by a further measuring method and to also apply the method to this, for example to better determine errors in the inertial sensor technology or odometry.

(7) However, FIGS. 1 and 2 are restricted to the first trajectory A, spanned by the first position values a.sub.0, a.sub.−1, . . . , a.sub.−N and the second trajectory B, spanned by the second position values b.sub.0, b.sub.−1, . . . , b.sub.−N. In this case, the points a.sub.i=(a.sub.x,i, a.sub.y,i, a.sub.z,i,), and b.sub.i=(b.sub.x,i, b.sub.y,i, b.sub.z,i,) respectively describe the relative receiver position at time i.

(8) As represented schematically in FIG. 1, a parallel-shifting of the second trajectory B along a displacement vector s is effected in the next step. This can be formulated mathematically for an exemplary second position value as b.sub.i=b.sub.u,i+s=(b.sub.x,u,i, b.sub.y,u,i, b.sub.z,u,i)+(s.sub.x, s.sub.y, s.sub.z), wherein b.sub.u,i denotes one of the second position values prior to the application of the parallel shift. The amount and direction of the displacement vector s is determined from the condition that the amounts of a quantity D of differential vectors, which are described below, are minimized on average, that is to say in other words the average formed from these amounts is as small as possible. Various possible ways of forming an average are known to the person skilled in the art.

(9) As represented in FIG. 2, a quantity D of differential vectors d.sub.0, d.sub.−1, . . . , d.sub.−N, is formed between the first position values and the parallel-shifted second position values assigned to the same times in a further step. For example, the differential vector d.sub.0 runs, starting from point a.sub.0, up to point b.sub.0 and the differential vector d.sub.−1 runs, starting from point a.sub.−1, up to point b.sub.−1.

(10) The quantity D of differential vectors thus obtained is a quality measure for the errors and integrity of the determined motion trajectories A, B. A small or vanishingly low differential vector suggests minor errors or a high integrity. In particular, the difference |d.sub.0| of the current, that is to say, last recorded position value, a maximum difference d.sub.max=max(|D|)=max(|d.sub.0|, |d.sub.−1|, . . . , |d.sub.−N|) and/or the simple or square sum of the individual differences di of the trajectories A, B can be formed as error indicators.

(11) The method according to the invention can be deployed, for example, in a vehicle which is explained in more detail below, but is not represented in the figures. The vehicle comprises, for example, an on-board system with a data bus which can be, for example, a Controller Area Network bus known to the person skilled in the art, called a CAN bus, or a FlexRay bus known to the person skilled in the art. These data buses are standardized and their method of operation can be looked up in the relevant specifications.

(12) Various modules are connected to the data bus, all of which communicate with one another via the data bus. These modules include a control unit for an electronic stability control of the vehicle which is not represented in greater detail, hereinafter referred to as the ESC control unit, a control unit for an electronic steering angle support of the vehicle which is not represented in greater detail, hereinafter referred to as the EPS control unit, a control unit for an airbag of the vehicle which is not represented in greater detail and further control units in the vehicle which is not represented in greater detail. These further control units include a control unit for an electronic braking assistant, a control unit for an electronic steering assistant and a control unit for automatic speed control.

(13) In addition, a display unit, which is to be referred to below as an HMI unit for “Human Machine Interface” unit, is connected to the data bus. It can also feed data into the data bus in a manner not shown if, for example, the HMI unit is configured as a touchscreen.

(14) Finally, a device for distributing data about the vehicle, not shown, is also connected to the data bus, which device is to be referred to below as an M2XPro-device for “motion information to different providers” device.

(15) The M2XPro device is embedded in an M2XPro cluster which provides an electrical power supply, hardware interfaces and protective means and a housing for the M2XPro device.

(16) The M2XPro device has a vehicle dynamics or inertial sensor technology interface, a wheel speed interface, a steering angle interface, a Global Position System interface, hereinafter referred to as a GNSS interface, and an output interface.

(17) Vehicle dynamics data, which are provided by an optional pitch rate sensor, an optional roll rate sensor and a yaw rate sensor in a manner known to the person skilled in the art, are received with the inertial sensor technology. In detail, the vehicle dynamics data therefore comprise an optional pitch rate, an optional roll rate and a yaw rate. A longitudinal, transverse and vertical acceleration can also be received correspondingly by a longitudinal, transverse and vertical acceleration sensor via the vehicle dynamics interface in a manner not represented.

(18) The wheel speeds of the individual wheels of the vehicle, not represented in greater detail, can be received at the wheel speed interface.

(19) A steering angle of a steering wheel of the vehicle, not represented in greater detail, can be received at the steering angle interface.

(20) Finally, position data can be received at the GNSS interface from a GNSS receiver, which can be arranged outside the M2XPro cluster.

(21) The received data received via the various reception interfaces indicated above are supplied to the output interface which can also perform the transformation of the received data into transmission data, yet to be written, beyond the pure delivery of the data to other modules. To this end, the output interface has a fusion filter which receives the received data and, based on these, produces the above-mentioned data about the vehicle, not shown. The fusion filter can make use of the information about the quality measure or the integrity of the position determination, which is obtained with the method according to the invention.

(22) The data produced by the fusion filter can be any information about the vehicle. Thus, for example, this information can comprise verified data which have been checked for their reliability on the basis of redundant information. This can be, for example, a verified yaw rate which has been checked for plausibility based on the steering angle, since the steering angle and the yaw rate are dependent on one another. Alternatively or additionally, this information can also include, for example, position data which cannot be supplied by the GNSS receiver. If the vehicle, not represented in greater detail, is driving for example in locations where no GNSS signals can be received such as in a tunnel, position data which cannot be supplied by the GNSS receiver can be derived, for example, by the fusion apparatus based on the last position data received from the GNSS receiver, the steering angle and a speed of the vehicle, not shown, derived from the wheel speeds. The fusion filter is therefore designed to generate new sensor data from sensor data present in the vehicle, not represented in greater detail, or to verify sensor data already existing, for example through plausibility checks. The fusion filter determines which input variables are the most reliable for determining the data, among other things, with the aid of the method according to the invention. To this end, the M2XPro device has a corresponding control device.

(23) The data generated by the fusion can be fed back into the data bus, on the one hand, or transferred to a filter. The generated data fed into the data bus are then available to other modules for further processing.

(24) The generated data transferred to the filter can be filtered there in a manner not specified in greater detail and transferred to a distributor antenna with a further filter. The distributor antenna is used for communication of the vehicle, not shown, with other vehicles and is therefore referred to below as the Car2X antenna. The communication can take place wirelessly, wherein a so-called automotive WLAN can be deployed, can be based on IEEE 802.11p and can have further protocol layers correspondingly based thereon such as IST-G5 in Europe or IEEE 1609 in America. The filter can therefore transfer filtered generated data to the Car2X antenna or receive received data from the Car2X antenna.

(25) The filtered generated data or the received data can be deployed as input data for a wide variety of applications. These applications can, for example, comprise the derivation of position data previously mentioned if the GNSS signal can no longer be received. Alternatively or additionally, these applications can comprise information applications which provide information, for example, about traffic jams or other traffic situations or traffic-relevant traffic situations. The applications can be combined under the term Car2X applications which can, in turn, feed their output data, that is to say the forwarded position data or the traffic information, into the data bus.

(26) The interfaces mentioned above can be combined in a hardware interface in the M2XPro device. Further hardware elements can be provided to realize the other functions in the M2XPro device.

(27) The output data such as the forwarded position data or the traffic information can be displayed, for example, on the HMI device for navigation.

(28) The input data or the generated data can be distributed to other modules in the vehicle, not represented, which can, for their part, have Car2X applications. In addition, system algorithms which use the generated data directly can be present in the individual control units.

(29) For example, the wheel speeds and the steering angle can accordingly be captured by the ESC control unit and the EPS control unit, which have to capture said data anyway for their intended operation.

(30) It should be pointed out that features which are merely disclosed in the description or features which are disclosed in the description or in a claim only in conjunction with other features can, in principle, be of independent inventive relevance. They can therefore also be included separately in claims to distinguish from the prior art.