USE OF COST MAPS AND CONVERGENCE MAPS FOR LOCALIZATION AND MAPPING

Abstract

A method for ascertaining features in an environment of at least one mobile unit for implementation of a localization and/or mapping by a control unit. In the course of the method, sensor measurement data of the environment are received, the sensor measurement data received are transformed by an alignment algorithm into a cost function and a cost map is generated with the aid of the cost function, a convergence map is generated based on the alignment algorithm. At least one feature is extracted from the cost map and/or the convergence map and stored, the at least one feature being provided in order to optimize a localization and/or mapping. A control unit, a computer program, and a machine-readable storage medium are also described.

Claims

1. A method for ascertaining features in an environment of at least one mobile unit, for implementation of a localization and/or mapping by a control unit, the method comprising the following steps: receiving sensor measurement data of the environment; transforming the received sensor measurement data by an alignment algorithm into a cost function, and generating a cost map using the cost function; generating a convergence map based on the alignment algorithm; ascertaining and storing at least one feature from the cost map and/or the convergence map; and providing the at least one feature to optimize the localization and/or the mapping.

2. The method as recited in claim 1, wherein a number of minima is extracted as at least one feature of the at least one feature from the cost map and/or the convergence map.

3. The method as recited in claim 2, further comprising: ascertaining periodically occurring features via a plurality of detected minima in the cost map and/or the convergence map, and ascertaining features occurring one time via a single minimum in the cost map and/or the convergence map.

4. The method as recited in claim 1, wherein the cost function is utilized to generate a two-dimensional or three-dimensional cost map.

5. The method as recited in claim 1, wherein a sharpness and/or a form of a minimum is determined in the cost map and utilized for processing the sensor measurement data.

6. The method as recited in claim 2, wherein differences are determined between the minima ascertained in the cost map.

7. A control unit configured to ascertaining feature in an environment of at least one mobile unit, for implementation of a localization and/or mapping, the control unit configured to: receive sensor measurement data of the environment; transform the received sensor measurement data by an alignment algorithm into a cost function, and generate a cost map using the cost function; generate a convergence map based on the alignment algorithm; ascertain and store at least one feature from the cost map and/or the convergence map; and provide the at least one feature to optimize the localization and/or the mapping.

8. A non-transitory machine-readable storage medium on which is stored a computer program for ascertaining features in an environment of at least one mobile unit, for implementation of a localization and/or mapping by a control unit, the computer program, when executed by a computer, causing the computer to perform the following steps: receiving sensor measurement data of the environment; transforming the received sensor measurement data by an alignment algorithm into a cost function, and generating a cost map using the cost function; generating a convergence map based on the alignment algorithm; ascertaining and storing at least one feature from the cost map and/or the convergence map; and providing the at least one feature to optimize the localization and/or the mapping.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0025] FIG. 1 shows a schematic flow chart to illustrate the method according to one specific embodiment of the present invention.

[0026] FIG. 2 shows an exemplary representation of a cost map with a minimum in accordance with an example embodiment of the present invention.

[0027] FIG. 3 shows an exemplary representation of a convergence map with a minimum in accordance with an example embodiment of the present invention.

[0028] FIG. 4 shows schematic representations of an environment of a mobile unit for illustrating repeating features in accordance with an example embodiment of the present invention.

[0029] FIG. 5 shows a schematic convergence map which reflects repeating features in accordance with an example embodiment of the present invention.

[0030] FIG. 6 shows a further schematic convergence map which reflects repeating features in accordance with an example embodiment of the present invention.

[0031] FIG. 7 shows schematic representation of sensor measurement data of a radar sensor which are used to create the convergence map in FIG. 6 in accordance with an example embodiment of the present invention.

[0032] FIG. 8 shows a further schematic convergence map which reflects a feature occurring one time in accordance with an example embodiment of the present invention.

[0033] FIG. 9 shows schematic representation of sensor measurement data of a radar sensor which are used to create the convergence map in FIG. 8 in accordance with an example embodiment of the present invention.

DETAILED DESCRIPTION OF EXAMPLE EMBODIMENTS

[0034] FIG. 1 shows a schematic flow chart to illustrate method 1 according to one specific embodiment of the present invention. Method 1 is used to ascertain features in an environment U of at least one mobile unit 2, for implementation of a localization and/or mapping by a control unit 4. Mobile unit 2 has control unit 4, takes the form of a vehicle operable in automated fashion, and is shown in FIG. 4.

[0035] In a first step 10 of method 1, sensor measurement data of environment U are received. For example, the sensor measurement data may be ascertained by a driving-environment sensor system 6, and received and evaluated by control unit 4. Alternatively, already existing map data may be called up.

[0036] In a further step 12, an alignment algorithm is provided and a cost function is generated by the alignment algorithm with the aid of the received sensor measurement data.

[0037] A cost map 14 is then created based on the alignment algorithm and the cost function. In a further step, a convergence map 16 is created based on the alignment algorithm.

[0038] In a further step, at least one feature is extracted from cost map 14 and/or convergence map 16 and stored 18.

[0039] The at least one feature is subsequently provided 20 in order to optimize a localization and/or mapping.

[0040] FIG. 2 shows an exemplary representation of a cost map 14 having one local minimum 8. Minimum 8 has a least value, so that mobile unit 2 is able to travel in the area of minimum 8 without danger of collision.

[0041] FIG. 3 shows an exemplary representation of a convergence map 16 having one minimum 8, which corresponds to an alignment algorithm employed for creating cost map 14 shown in FIG. 2. Points 22 show the starting points for the iteration points. The lines or trajectories 24 correspond to the directions in which respective points 22 converge. The efficiency of alignment algorithms may be illustrated by the use of convergence map 16.

[0042] FIG. 4 shows schematic representations of an environment U of a mobile unit 2 for illustrating repeating features 26. For example, repeating features 26 may be lane markings or guardrail supports.

[0043] As an example, mobile unit 2 takes the form of a vehicle and has a control unit 4. Control unit 4 is connected to a driving-environment sensor system 6 in a manner allowing the transfer of data. Control unit 4 is thereby able to receive sensor measurement data from driving-environment sensor system 6.

[0044] For example, driving-environment sensor system 6 may have camera sensors, radar sensors, LIDAR sensors, ultrasonic sensors and the like, and may provide the ascertained sensor measurement data in analog or digital form to control unit 4.

[0045] Corresponding to repeating features 24, FIG. 5 shows a schematic convergence map 16 which reflects repeating features 26. The number of minima 8 gives information here about the uniqueness of features 26. Unique features 25 lead to a single minimum 8. Periodically occurring features 26 lead to a plurality of local minima 8. This relationship is illustrated in FIGS. 6 through 9.

[0046] FIG. 7 and FIG. 9 show sensor measurement data of a radar sensor. FIG. 7 shows sensor measurement data of a turnpike section having a large number of reflector posts as repeating features 26. FIG. 6 illustrates a convergence map 16 corresponding to that and having a plurality of minima 8.

[0047] FIG. 9 shows sensor measurement data of a radar sensor of a turnpike exit. The turnpike exit represents a unique feature 25 occurring one time. Resulting convergence map 16 is shown in FIG. 8 and has a single minimum 8. The uniqueness of respective features 25, 26 may thus be inferred based on the number of minima 8.