METHOD FOR ASCERTAINING AN INITIAL POSE OF A VEHICLE

Abstract

A method for ascertaining an initial pose of a vehicle using a control device. Measured data ascertained by a GNSS sensor system and/or an odometry sensor system are received and evaluated to ascertain an approximate pose of the vehicle with a margin of uncertainty. At least one trajectory of road users is extracted from a trajectory map for the ascertained margin of uncertainty. Test points are positioned along the extracted trajectory. An optimization algorithm is performed for each test point along the trajectory. The optimization algorithm ascertains poses having corresponding cost functions. A pose having the greatest cost function is determined as the initial pose of the vehicle from the poses ascertained by the optimization algorithm. A control device, a computer program, and a machine-readable storage medium are also provided.

Claims

1-10. (canceled)

11. A method for ascertaining an initial pose of a vehicle using a control device, the method comprising the following steps: receiving and evaluating measured data ascertained by a GNSS sensor system and/or an odometry sensor system, to ascertain an approximate pose of the vehicle with a margin of uncertainty; extracting at least one trajectory of road users from a trajectory map for the ascertained margin of uncertainty; positioning test points along the extracted trajectory, and performing an optimization algorithm for each of the test points along the trajectory, the optimization algorithm ascertaining poses having corresponding cost functions; and ascertaining, as the initial poser of the vehicle, a pose having a greatest cost function from the poses ascertained by the optimization algorithm.

12. The method as recited in claim 11, wherein an orientation of the vehicle is determined from the ascertained approximate pose of the vehicle and compared with driving directions of the extracted trajectories, and a filter function is implemented for an exclusive consideration of trajectories featuring driving directions, and at least one trajectory that agrees with the orientation of the vehicle is taken into account.

13. The method as recited in claim 11, wherein a hill-climbing algorithm is performed as the optimization algorithm.

14. The method as recited in claim 11, wherein the optimization algorithm is performed along the at least one extracted trajectory within the ascertained margin of uncertainty.

15. The method as recited in claim 11, wherein the method for ascertaining the initial pose of the vehicle is carried out repeatedly at defined time intervals.

16. The method as recited in claim 15, wherein a plurality of initial poses of the vehicle is ascertained at different points in time, and a deviation of the vehicle from a traffic lane or a traffic lane change is determined by detecting inconsistencies between the ascertained initial poses.

17. The method as recited in claim 11, wherein measured data ascertained by at least one LiDAR sensor and/or radar sensor are received and evaluated, and features are ascertained based on the measured data and compared with data of a feature map in order to determine a pose of the vehicle along at least one extracted trajectory in the feature map, and a maximum cost function of a pose of the vehicle is calculated by the optimization algorithm based on a minimum deviation of the ascertained features and features stored in the feature map.

18. A control device configured to ascertain an initial pose of a vehicle using a control device, the control device configured to: receive and evaluate measured data ascertained by a GNSS sensor system and/or an odometry sensor system, to ascertain an approximate pose of the vehicle with a margin of uncertainty; extract at least one trajectory of road users from a trajectory map for the ascertained margin of uncertainty; position test points along the extracted trajectory, and performing an optimization algorithm for each of the test points along the trajectory, the optimization algorithm ascertaining poses having corresponding cost functions; and ascertain, as the initial poser of the vehicle, a pose having a greatest cost function from the poses ascertained by the optimization algorithm.

19. A non-transitory machine-readable memory medium on which is stored a computer program for ascertaining an initial pose of a vehicle using a control device, the computer program, when executed by a computer, causing the computer to perform the following steps: receiving and evaluating measured data ascertained by a GNSS sensor system and/or an odometry sensor system, to ascertain an approximate pose of the vehicle with a margin of uncertainty; extracting at least one trajectory of road users from a trajectory map for the ascertained margin of uncertainty; positioning test points along the extracted trajectory, and performing an optimization algorithm for each of the test points along the trajectory, the optimization algorithm ascertaining poses having corresponding cost functions; and ascertaining, as the initial poser of the vehicle, a pose having a greatest cost function from the poses ascertained by the optimization algorithm.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0035] FIG. 1 shows a schematic representation of a traffic lane with a margin of uncertainty of an approximate pose and a multitude of points for performing an optimization algorithm, according to an example embodiment of the present invention.

[0036] FIG. 2 shows a schematic representation of a traffic lane with a margin of uncertainty of a pose and a multitude of points for performing an optimization algorithm along trajectories, according to an example embodiment of the present invention.

DETAILED DESCRIPTION OF EXAMPLE EMBODIMENTS

[0037] FIGS. 1 and 2 show schematic representations to illustrate a method for ascertaining an initial pose A of a vehicle 2, which is able to be used in a localization of vehicle 2 (not described further). For instance, initial pose A may be determined with the aid of a vehicle-side control device 4 which carries out the method.

[0038] FIG. 1 shows a schematic representation of a traffic lane 1 with a margin of uncertainty U of an approximate pose P and a plurality of test points 5 for performing an optimization algorithm. Vehicle 2 travels in traffic lane 1 in driving direction F.

[0039] Vehicle 2 has an odometry sensor system and/or a GNSS sensor system 6 and an additional sensor system 8 for a feature-based localization. Additional sensor system 8 may be developed as a LiDAR sensor system, a radar sensor system and/or a camera sensor system, for instance.

[0040] In the illustrated exemplary embodiment, the odometry sensor system and GNSS sensor system 6 collect measured data while the vehicle is in motion. An approximate pose P is determined from the measured data of the odometry sensor system and GNSS sensor system 6. Since approximate pose P includes errors, a margin of uncertainty U of approximate pose U is schematically illustrated.

[0041] Test points 5 may be distributed across entire margin of uncertainty U. Starting from the positions of test points 5, an optimization algorithm is able to be performed. However, in order to accelerate the optimization algorithm, trajectories 10 that run through margin of uncertainty U are extracted from a trajectory map.

[0042] Next, test points 5 are positioned along extracted trajectories 10 so that the optimization algorithm is performed for each test point 5 and along trajectories 10. This makes it possible to reduce the calculation work of the optimization algorithm from a two-dimensional to a one-dimensional problem. This step is shown in FIG. 2.

[0043] FIG. 2 shows a schematic representation of traffic lane 1 including a margin of uncertainty U of a pose, and a multitude of test points 5 for carrying out the optimization algorithm along trajectories 10.

[0044] Based on the historical data of the trajectory map, the probability for vehicle 2 along one of extracted trajectories 10 is the most likely. As a result, the optimization algorithm is performed along extracted trajectories 10.

[0045] When the optimization algorithm is performed, measured data preferably received from environment sensor system 8 are received and evaluated.

[0046] Statistical features are ascertained based on the measured data and compared to data of a feature map. The optimization algorithm is predominantly used for adapting the statistical features to features of the feature map and for maximizing a cost function.

[0047] The cost function reaches its maximum when the extracted statistical features optimally agree with the features of the feature map.

[0048] One of multiple possible poses of vehicle 2 that has a maximum cost function is ascertained as an initial pose A and used for a further localization of vehicle 2.