METHOD FOR PROCESSING POSE INFORMATION IN AN AT LEAST PARTIALLY AUTOMATED VEHICLE AND/OR ROBOT

Abstract

A method (100) for processing vehicle and/or robot pose information in an at least partially automated vehicle (50), a driving assistance system (60) of the vehicle (50), and/or a robot (70), comprising the steps of: determining (110), based at least in part on measurement data (1) gathered by at least one sensor that is carried by the vehicle (50) and/or robot (70), a pose (2) of the vehicle (50) and/or robot (70), as well as maximum expected errors (2a) of at least the pose (2); querying (120), based at least in part on the position comprised in the determined pose (2), an alert limit service (3) for position-dependent, and optionally also orientation-dependent, maximum permissible errors (4); determining (130) whether the maximum expected errors (2a) are within the maximum permissible errors (4); and if the maximum expected errors (2a) exceed the maximum permissible errors (4), initiating (160) at least one remedial action.

Claims

1. A method for processing vehicle and/or robot pose information in an at least partially automated vehicle, a driving assistance system of the vehicle, and/or a robot, comprising the steps of: determining, based at least in part on measurement data gathered by at least one sensor that is carried by the vehicle and/or robot, a pose of the vehicle and/or robot, as well as maximum expected errors of at least the pose; querying, based at least in part on the position comprised in the determined pose, an alert limit service for position-dependent, and/or orientation-dependent, maximum permissible errors; determining whether the maximum expected errors are within the maximum permissible errors; and if the maximum expected errors exceed the maximum permissible errors, initiating at least one remedial action.

2. The method of claim 1, further comprising the step of: if the maximum expected errors are within the maximum permissible error, computing, based at least in part on the determined pose, an actuation signal, and actuating the vehicle, the driving assistance system, and/or the robot, with the actuation signal.

3. The method of claim 1, wherein the alert limit service comprises at least one map and/or database in which maximum permissible errors, and/or precursors for the computation of the maximum permissible errors, are stored.

4. The method of claim 3, wherein the at least one map and/or database is located on board the vehicle and/or robot.

5. The method of claim 3, wherein the maximum permissible errors, and/or the precursors, stored in the at least one map and/or database represent: the strictest possible maximum permissible errors that may be rendered more lenient by maximum permissible errors and/or precursors from other sources, or the most lenient possible maximum permissible errors that may be rendered stricter by maximum permissible errors and/or precursors from other sources.

6. The method of claim 1, wherein the alert limit service comprises a cloud service that delivers, based at least in part on the position, maximum permissible errors and/or precursors.

7. The method of claim 6, wherein the alert limit service comprises at least one map and/or database in which maximum permissible errors, and/or precursors for the computation of the maximum permissible errors, are stored; and wherein the at least one map and/or database is located on board the vehicle and/or robot; the method further comprising: performing, based at least in part on measurement data gathered by at least one sensor that is carried by the vehicle and/or robot, a plausibility check as to whether the information obtained from the map and/or database on board the vehicle and/or robot is still accurate; and if the information is found to be still accurate, using it to determine the sought maximum permissible errors; and if the information is found to be no longer accurate, querying the cloud service for up-to-date maximum permissible errors and/or precursors.

8. The method of claim 6, wherein the alert limit service comprises at least one map and/or database in which maximum permissible errors, and/or precursors for the computation of the maximum permissible errors, are stored; and wherein the at least one map and/or database is located on board the vehicle and/or robot; and wherein the cloud service is queried first, and the map and/or database on board the vehicle and/or robot is queried if the cloud service is not available.

9. The method of claim 1, further comprising: modifying the maximum permissible errors based at least in part on the mass, and/or the mass distribution, of the vehicle and/or robot; and/or the dimensions of a load that extends beyond the vehicle and/or robot.

10. A localization module for an at least partially automated vehicle, a driving assistance system of the vehicle, and/or a robot, comprising: an interface configured to read in measurement data gathered by at least one sensor that is carried by the vehicle and/or robot, processing means configured to determine, based at least in part on the measurement data, a pose of the vehicle and/or robot, as well as maximum expected errors of at least the pose, and an integrity monitoring submodule that is configured to: determine maximum permissible errors by querying the maximum permissible errors, and/or precursors for their computation, from a local map and/or database, and/or from a cloud service; compare the determined maximum expected errors with the maximum permissible errors; and in response to determining that the maximum expected errors exceed the maximum permissible errors, cause a disengaging of the autonomous operation of the vehicle, the robot, and/or the driving assistance system.

11. A method for determining maximum permissible errors of at least a pose of a vehicle and/or robot that is to move in an at least partially automated manner, and/or that is to be assisted by a driving assistance system, the method comprising the steps of: providing a map of the area in which the vehicle and/or robot is to be operated, wherein this map comprises at least the geometry of roads and/or paths on which the vehicle and/or robot is to travel; and for each of a set of possible positions that are reachable by the vehicle and/or robot: determining, based at least in part on features from the map, a correlation between a risk that the vehicle and/or robot is implicated in at least one undesired event on the one hand, and maximum expected errors of at least the pose of the vehicle and/or robot on the other hand; and determining, based at least in part on this correlation and a predetermined maximum allowable risk level for the undesired event, the sought maximum permissible errors, and/or precursors for their computation.

12. The method of claim 11, wherein the correlation is based at least in part on a distance of at least of a portion of the vehicle and/or robot to an area where the presence of this portion of the vehicle and/or robot can cause the at least one undesired event.

13. The method of claim 11, wherein the undesired event comprises one or more of: entry of the vehicle and/or robot into an area where other traffic participants have priority; a collision of the vehicle and/or robot with at least one other traffic participant or other object; a mis-association of traffic signs and/or traffic lights that are valid for another lane of traffic to the lane of traffic travelled by the vehicle and/or robot; and a mis-association of a traffic participant that travels in another lane of traffic to the lane of traffic travelled by the vehicle and/or robot.

14. A non-transitory computer-readable medium for storing a computer program, the computer program comprising machine-readable instructions that, when executed by one or more computers and/or compute instances, upgrade the one or more computers and/or compute instances to an integrity monitoring submodule that is configured to: determine maximum permissible errors by querying the maximum permissible errors, and/or precursors for their computation, from a local map and/or database, and/or from a cloud service; compare the determined maximum expected errors with the maximum permissible errors; and in response to determining that the maximum expected errors exceed the maximum permissible errors, cause a disengaging of the autonomous operation of the vehicle, the robot, and/or the driving assistance system; and cause the one or more computers and/or compute instances to perform a method according to claim 1.

14. (canceled)

15. (canceled)

17. A non-transitory computer-readable medium for storing a computer program, the computer program comprising machine-readable instructions that, when executed by one or more computers and/or compute instances, upgrade the one or more computers and/or compute instances to an integrity monitoring submodule that is configured to: determine maximum permissible errors by querying the maximum permissible errors, and/or precursors for their computation, from a local map and/or database, and/or from a cloud service; compare the determined maximum expected errors with the maximum permissible errors; and in response to determining that the maximum expected errors exceed the maximum permissible errors, cause a disengaging of the autonomous operation of the vehicle, the robot, and/or the driving assistance system; and cause the one or more computers and/or compute instances to perform a method according to claim 11.

18. A non-transitory computer-readable medium for storing a computer program, the computer program comprising machine-readable instructions that, when executed by one or more computers and/or compute instances, cause the one or more computers and/or compute instances to perform a method according to claim 1.

19. A non-transitory computer-readable medium for storing a computer program, the computer program comprising machine-readable instructions that, when executed by one or more computers and/or compute instances, cause the one or more computers and/or compute instances to perform a method according to claim 1.

Description

DESCRIPTION OF THE FIGURES

[0048] In the following, the invention is illustrated using Figures without any intention to limit the scope of the invention. The Figures show:

[0049] FIG. 1 Exemplary embodiment of the method 100 for processing vehicle and/or robot pose information in an at least partially automated vehicle 50, a driving assistance system 60 of the vehicle 50, and/or a robot 70;

[0050] FIG. 2 Exemplary illustrations how maximum permissible errors may vary in different situations;

[0051] FIG. 3 Exemplary embodiment of the localization module;

[0052] FIG. 4 Exemplary embodiment of the method 200 for determining maximum permissible errors.

[0053] FIG. 1 is a schematic flow chart of an exemplary embodiment of the method 100 for processing vehicle and/or robot pose information in an at least partially automated vehicle 50, a driving assistance system 60 of the vehicle 50, and/or a robot 70.

[0054] In step 110, based at least in part on measurement data 1 gathered by at least one sensor that is carried by the vehicle 50 and/or robot 70, a pose 2 of the vehicle 50 and/or robot 70, as well as an maximum expected errors 2a of at least the pose 2, are determined.

[0055] In step 120, based at least in part on the position comprised in the determined pose 2, an alert limit service 3 is queried for position-dependent, and optionally also orientation-dependent, maximum permissible errors 4. In particular, this alert limit service 3 may comprise at least one map 30 and/or database in which maximum permissible errors 4, and/or precursors for the computation of the maximum permissible errors 4, are stored. It may also comprise at least one cloud service 31.

[0056] According to block 121, based at least in part on measurement data 1 gathered by at least one sensor that is carried by the vehicle 50 and/or robot 70, a plausibility check may be performed as to whether the information obtained from the map 30 and/or database on board the vehicle 50 and/or robot 70 is still accurate. If the information is found to be still accurate, it may be used, according to block 122, to determine the sought maximum permissible errors 4. By contrast, if the information is found to be no longer accurate, according to block 123, the cloud service 31 may be queried for up-to-date maximum permissible errors 4 and/or precursors for their computation.

[0057] According to block 124, the cloud service 31 may be queried first. If this cloud service 31 is not available, according to block 125, the map 30 and/or database on board the vehicle 50 and/or robot 70 may be queried.

[0058] Irrespective of how exactly the sought maximum permissible errors 4 are obtained, they may be modified, according to block 126, based at least in part on [0059] the mass, and/or the mass distribution, of the vehicle 50 and/or robot 70, and/or [0060] the dimensions of a load that extends beyond the vehicle 50 and/or robot 70.

[0061] The modified version of the maximum permissible errors is labelled with the reference sign 4*.

[0062] In step 130, it is determined whether the maximum expected errors 2a are within the maximum permissible errors 4. If this is the case (truth value 1), in step 140, based at least in part on the determined pose 2, an actuation signal 5 is computed. In step 150, the vehicle 50, the driving assistance system 60, and/or the robot 70, is then actuated with this actuation signal 5. However, if the maximum expected errors 2a are not within the maximum permissible errors 4 (truth value 0 at diamond 130), in step 160, a remedial action is taken. For example, this remedial action may comprise disengaging autonomous operation and entering a system unavailable mode.

[0063] FIG. 2 illustrates on three examples how the determined maximum permissible errors 4, 4, 4 may vary in different traffic situations.

[0064] FIG. 2a shows a first example. In this example, the vehicle 50 travels a road 52. The road 52 is a two-way road; there is another oncoming vehicle 51 in the oncoming lane beyond the central divider of the road 52. This first example illustrates the dependency of the maximum permissible errors 4 on road geometry.

[0065] When the vehicle 50 is in pose 2 on a long, straight stretch of the road 52, a longitudinal uncertainty in the direction of travel is largely inconsequential. Therefore, the maximum permissible error 4 for the longitudinal component of the pose 2 is rather high. The maximum permissible error 4 for the lateral component of the pose 2 perpendicular to the direction of travel is much lower because it is important that the vehicle 50 stays in lane.

[0066] When the vehicle 50 is in pose 2 in a bend of the road 52, the maximum permissible error 4 for the lateral component of the pose 2 is unchanged. The maximum permissible error 4 for the longitudinal component of the pose 2 needs to be much lower because such an uncertainty might cause the vehicle 50 to cross the central divider of the road 52.

[0067] When the vehicle 50 is in pose 2 on a shorter straight stretch of the road 52 before said bend, the maximum permissible error 4 for the longitudinal component of the pose 2 is higher than the corresponding maximum permissible error 4 in pose 2 in the bend. However, it is a lot lower than the maximum permissible error 4 in pose 2 on the much longer straight stretch of the road 52.

[0068] FIG. 2b shows a second example. In this example, the road 52 leads to an intersection 53 that is protected with a stop sign 54. This second example illustrates the dependency of the maximum permissible errors 4 on the distance available for stopping.

[0069] When the vehicle 50 is in pose 2 immediately before the intersection, the maximum permissible error 4 for the longitudinal component of the pose 2 is low because such uncertainty might cause the vehicle to pass the stop line of the stop sign 54. In poses 2 and 2, the maximum permissible error 4, 4 for the longitudinal component of the pose 2, 2 increases with the distance that is still available for stopping before the intersection 53. The maximum permissible error 4, 4, 4 for the lateral component of the pose 2, 2, 2 perpendicular to the direction of travel is always the same, so as to ensure that the vehicle 50 keeps in lane.

[0070] FIG. 2c shows a third example. In this example, the road 52 with two lanes 52a and 52b leads to an intersection 53 again, but the intersection 53 is now protected with traffic lights 54a and 54b that are associated with lanes 52a and 52b, respectively. This third example illustrates how the maximum permissible error for the orientation may vary when approaching the intersection 53.

[0071] When the vehicle 50 is in pose 2, it is critical that it does not mis-interpret the green traffic light 54b that is valid for another vehicle 51 as being valid for vehicle 50, causing this vehicle 50 to run the red light 54a. Therefore, the maximum permissible error 4 for the orientation is very low. Farther away from the intersection 53, when the vehicle 50 is in pose 2, the maximum permissible error 4 for the orientation can be relaxed considerably.

[0072] FIG. 3 is a schematic illustration of an embodiment of the localization module 40. The localization module 40 takes in measurement data 1 and outputs a pose 2 of the vehicle 50 and/or robot 70, including maximum expected errors 2a of at least the pose 2.

[0073] An interface 41 of the localization module 40 is configured to read in measurement data 1 gathered by at least one sensor that is carried by the vehicle 50 and/or robot 70. Processing means 42 of the localization module 40 are configured to determine, based at least in part on the measurement data, a pose 2 of the vehicle 50 and/or robot 70, as well as maximum expected errors 2a of at least the pose 2.

[0074] An integrity monitoring submodule 43 of the localization module 40 is configured to determine maximum permissible errors 4 by querying the maximum permissible errors 4, and/or precursors for their computation, from a local map 30 and/or database, and/or from a cloud service 31. The determined maximum expected errors 2a are compared with the maximum permissible errors 4. If the maximum permissible errors 4 are exceeded, the autonomous operation of the vehicle 50, the robot 70, and/or the driving assistance system 60, is disengaged. This implies that in this case, the pose 2 and maximum expected errors 2a will no longer be provided to downstream systems of the vehicle 50, the robot 70, and/or the driving assistance system 60 for use.

[0075] FIG. 4 is a schematic flow chart of an embodiment of the method 200 for determining maximum permissible errors 4 of at least a pose 2 of a vehicle 50 and/or robot 70 that is to move in an at least partially automated manner, and/or that is to be assisted by a driving assistance system 60.

[0076] In step 210, a map 30 of the area in which the vehicle 50 and/or robot 70 is to be operated is provided. This map 30 comprises at least the geometry of roads and/or paths on which the vehicle 50 and/or robot 70 is to travel.

[0077] For each of a set of possible positions that are reachable by the vehicle 50 and/or robot 70, a correlation 6 between a risk that the vehicle 50 and/or robot 70 is implicated in at least one undesired event on the one hand, and maximum expected errors 2a of at least the pose 2 of the vehicle 50 and/or robot 70 on the other hand, are determined in step 220.

[0078] According to block 221, the correlation 6 may be based at least in part on a distance of at least of a portion of the vehicle 50 and/or robot 70 to an area where the presence of this portion of the vehicle 50 and/or robot 70 can cause the at least one undesired event.

[0079] According to block 222, the undesired event may comprise one or more of: [0080] entry of the vehicle 50 and/or robot 70 into an area where other traffic participants have priority; [0081] a collision of the vehicle 50 and/or robot 70 with at least one other traffic participant or other object; [0082] a mis-association of traffic signs and/or traffic lights that are valid for another lane of traffic to the lane of traffic travelled by the vehicle 50 and/or robot 70; and [0083] a mis-association of a traffic participant that travels in another lane of traffic to the lane of traffic travelled by the vehicle 50 and/or robot 70.

[0084] In step 230, based at least in part on this correlation 6 and a predetermined maximum allowable risk level 7 for the undesired event, the sought maximum permissible errors 4, and/or precursors for their computation, are determined.

LIST OF REFERENCE SIGNS

[0085] 1 measurement data [0086] 2, 2, 2 pose of vehicle 50 and/or robot 70 [0087] 2a maximum expected errors [0088] 3 alert limit service [0089] 30 map of alert limit service 3 [0090] 31 cloud service of alert limit service 3 [0091] 4, 4, 4 maximum permissible errors [0092] 4* modified version [0093] 5 actuation signal [0094] 6 correlation between risk and maximum expected errors 2a [0095] 7 maximum allowable risk level [0096] 40 localization module [0097] 41 interface of localization module 40 [0098] 42 processing means of localization module 40 [0099] 43 integrity monitoring submodule of localization module 40 [0100] 50 vehicle [0101] 51 other vehicle [0102] 52 road [0103] 52a, 52b lanes of road 52 [0104] 53 intersection [0105] 54 stop sign [0106] 54a, 54b traffic lights [0107] 60 driving assistance system [0108] 70 robot [0109] 100 method for controlling vehicle 50, system 60, robot 70 [0110] 110 determining pose 2 and maximum expected errors 2a [0111] 120 querying alert limit service 3 [0112] 121 performing plausibility check [0113] 122 using plausible information to determine maximum permissible error 4 [0114] 123 querying cloud service 31 [0115] 124 querying cloud service 31 first [0116] 125 querying map 30 if cloud service 31 not available [0117] 126 modifying alert limits 4 [0118] 130 determining whether expected error 2a is within permissible error 4 [0119] 140 computing actuation signal 5 [0120] 150 actuating systems 50, 60, 70 with actuation signal 5 [0121] 160 initiating remedial action [0122] 200 method for determining maximum permissible errors 4 [0123] 210 providing map 30 [0124] 220 determining correlation 6 [0125] 221 specific way of determining correlation 6 [0126] 222 specific undesired events [0127] 230 determining maximum permissible errors 4