Determination of a Control Signal for an In-Part-Autonomous Vehicle

Abstract

A system for a vehicle, which drives in an at least in-part-automated manner is configured to determine a control signal for a control system. The system includes a sensor, a planning module, and a monitoring module. The sensor is configured to detect an object in a surrounding area of the vehicle and store a corresponding object representation. The planning module is configured to determine, based to the stored object representation, a first trajectory and a first probability of collision of the first trajectory for the vehicle. The monitoring module is configured to perform one of following actions when the first probability of collision exceeds a predefined probability of collision: determine, using the planning module and based on the stored object representation, a further trajectory having a further probability of collision and a maximum deceleration of the further trajectory; or assess the stored object representation of the object using the sensor.

Claims

1. A system for a vehicle, which is at least in-part automated, for determining a control signal for a control system, comprising: a sensor configured to detect an object in a surrounding area of the vehicle and to store a corresponding object representation; a planning module configured to determine, based on the stored object representation, a first trajectory and a first probability of collision of the first trajectory for the vehicle; a monitoring module configured to perform one of following actions when the first probability of collision exceeds a predefined probability of collision: determine, using the planning module and based on the stored object representation, a further trajectory having a further probability of collision and a maximum deceleration of the further trajectory; or assess the stored object representation of the object using the sensor.

2. The system according to claim 1, wherein the sensor includes at least one of a camera, a plurality of cameras, a radar sensor, a LiDAR sensor, and an ultrasonic sensor.

3. The system according to claim 1, wherein the stored object representation is a verified object representation or a non-verified object representation.

4. The system according to claim 1, wherein when the first probability of collision exceeds the predefined probability of collision, the monitoring module is further configured to: use a second trajectory in lieu of the first trajectory when a second probability of collision of the second trajectory is judged as acceptable by the monitoring module; use a third trajectory having a third predefined probability of collision when the second probability of collision of the second trajectory is judged as unacceptable by the monitoring module, wherein the third trajectory uses only a verified object representation of the object; determine a fourth trajectory having a fourth probability of collision and select the lower probability of collision from one of the third probability of collision and the fourth probability of collision when the third probability of collision of the third trajectory is judged as unacceptable by the monitoring module; and select between a probability of collision having the lowest braking deceleration when the third probability of collision and the fourth probability of collision are approximately identical.

5. A vehicle comprising: a sensor configured to detect an object in a surrounding area of the vehicle and to store a corresponding object representation; a planning module configured to determine, based on the stored object representation, a first trajectory and a first probability of collision of the first trajectory for the vehicle; a monitoring module configured to perform one of following actions when the first probability of collision exceeds a predefined probability of collision: determine, using the planning module and based on the stored object representation, a further trajectory having a further probability of collision and a maximum deceleration of the further trajectory; or assess the stored object representation of the object using the sensor, wherein the vehicle is at least in-part automated.

6. A method for determining a control signal for a control system for a vehicle that is at least in-part-automated, comprising: detecting, with a sensor, an object in a surrounding area of the vehicle; storing a corresponding object representation based on the detected object; determining, using a planning module and based on the stored object representation, a first trajectory and a first probability of collision of the first trajectory for the vehicle; and performing one of the following using a monitoring module when the first probability of collision exceeds a predefined probability of collision: determining, using the planning module and based on the stored object representation, a further trajectory having a further probability of collision and a maximum deceleration of the further trajectory, or checking the stored object representation of the object using the sensor.

7. The method according to claim 6, further comprising: applying a second trajectory, using the monitoring module, in lieu of the first trajectory when a second probability of collision of the second trajectory is judged as acceptable by the monitoring module.

8. The method according to claim 7, further comprising: applying a third trajectory, using the monitoring module, having a third predefined probability of collision and a third maximum deceleration when the second probability of collision of the second trajectory is judged as unacceptable by the monitoring module, wherein the third trajectory uses only a verified object representation of the object.

9. The method according to claim 8, further comprising: determining a fourth trajectory, using the monitoring module, having a fourth probability of collision and a fourth maximum deceleration; and selecting the lower probability of collision from the third probability of collision and the fourth probability of collision when the third probability of collision of the third trajectory is judged as unacceptable by the monitoring module.

10. The method according to claim 9, further comprising: selecting, using the monitoring module, the probability of collision having the lowest braking deceleration when the third probability of collision and the fourth probability of collision are approximately identical.

11. The method according to claim 6, wherein the sensor includes at least one of a camera, a plurality of cameras, a radar sensor, a LiDAR sensor, and an ultrasonic sensor.

12. The method according to claim 6, wherein the stored object representation is a verified object representation or a non-verified object representation.

13. The method according to claim 6, wherein a program element, when executed on a processing unit, is configured to implement the method.

14. The method according to claim 13, wherein the program element is stored in a computer-readable medium.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0033] FIG. 1 illustrates schematically a system in accordance with one embodiment of the disclosure.

[0034] FIG. 2 illustrates schematically an overview of trajectories that have been determined in accordance with one embodiment of the disclosure.

[0035] FIG. 3 illustrates a method in accordance with one embodiment of the disclosure.

DETAILED DESCRIPTION

[0036] FIG. 1 illustrates schematically a system 200 in accordance with one embodiment of the disclosure. In this case, an object 120 that is located in a real surrounding area 100 is detected by the sensor 300. The sensor 300 may comprise one or multiple individual sensors, the sensor 300 may also comprise a subsystem that combines the data of multiple sensors and/or multiple sensor types (for example camera and radar) and provides said data as a combined representation to the system. The object 120 that is detected by means of the sensor is guided to a control device 250 that is used so as to determine a control signal 370. The control signal 370 controls a control system 380 for a vehicle 110 (not illustrated). The control device 250 comprises a storage device 400. The object 120 that is detected by means of the sensor is stored in this storage device 400 as an object representation 420, by way of example in the form of a geometric description or as an attributed object. In at least some cases, it may be that the object 120 that is detected by means of the sensor is a ghost, in other words does not exist in the real world but is stored in the storage device 400 as an object representation 420. In many cases, a plurality of objects 120 is detected by the sensor 300 and is stored as a plurality of object representations 420 in the storage device 400. The object representation 420 may be stored as a verified object representation 425 or a non-verified object representation 422. In the case of a verified object representation 425, the object 120 is detected at least one more time by the sensor 300 or by another sensor and/or by another sensor type.

[0037] A planning module 500 uses the object representation 420 or the plurality of object representations 420 in order to determine therefromusing further information, for example from a navigation systema first trajectory 501 and a first probability of collision 521, where applicable also a first maximum deceleration 541, of the first trajectory 501 for the vehicle 110. The planning module 500 may also determine further trajectories 502, 503, 504 in dependence upon the value of the first probability of collision 521. The further trajectories 502, 503, 504 may be determined for example at the same time as the first trajectory or only after checking the probabilities of collision and/or in accordance with applying the above-mentioned rules. The probabilities of collision 522, 523, 524 corresponding to each of the trajectories 502, 503, 504 are determined. The maximum decelerations 542, 543, 544 corresponding to at least some of the trajectories 502, 503, 504 are determined.

[0038] Furthermore, the system 200 comprises a monitoring module 600 that is configured so as to perform one of the following actions if the first probability of collision 521 exceeds a predefined probability of collision 530: determine by means of the planning module 500 in reaction to the object representation 420 a further trajectory 502, 503, 504 having a further probability of collision 522, 523, 524 and a maximum deceleration 542, 543, 544 of the further trajectory 502, 503, 504, or check the object representation 420 of the object 120 by means of the sensor 300 with the result that the object representation 420 either disappears from the storage device 400 or is converted from a non-verified object representation 422 into a verified object representation 425. The control signal 370 is determined by means of the selection of one of the trajectories 501, 502, 503, 504. This may be used for example by a control system 380 (illustrated as dots) in order to control the vehicle 110.

[0039] FIG. 2 illustrates schematically an overview of trajectories 501, 502, 503, 504 that have been determined in accordance with one embodiment of the disclosure. In this case, the speed v that is to be controlled is plotted over the time t using 501, 502, 503, 504. In this case, it is clearly apparent that the first trajectories 501 (of which a plurality is plotted as an example), permit an acceleration and a deceleration. If one of the trajectories 501 would potentially lead to a collision, in other words if the first probability of collision 521 exceeds a predefined probability of collision 530 then second trajectories 502 are determined. If the probability of collision 522 of said trajectories is judged as unacceptable by the monitoring module 600, then third trajectories 503 are determined; the third trajectories 503 in this case only use the verified object representation 425 of the object 120. If the probability of collision 523 of the third trajectories 503 is judged as unacceptable, then fourth trajectories 504 and the maximum braking decelerations 543, 544 are determined. If the probabilities of collision 523, 524 of the third and the fourth trajectories 503, 504 are approximately identical, then the trajectory having the lower maximum braking deceleration 543, 544 (in other words the trajectory having the lower negative incline) is selected. This increases the safety and comfort of the vehicle because as a consequence as much as possible sudden emergency braking procedures become considerably less frequent.

[0040] FIG. 3 illustrates a method 700 in accordance with one embodiment of the disclosure. The method is executed on a system 200 for a vehicle 110 which drives in an at least in part automated manner. The system 200 may be a part of a processor unit. The method is used to determine a control signal 370 for a control system 380. In one step 701, an object 120 is to detected in a surrounding area 100 by means of a sensor 300 (cf. FIG. 1) and is stored in a storage device 400 as an object representation 420. In one step 702, a first trajectory P1 501 and a first probability of collision 521 of the first trajectory 501 for the vehicle 110 is determined by means of a planning module 500 in reaction to the object representation 420. In one step 703, a check is performed as to whether the first probability of collision P1 521 exceeds a predefined probability of collision P 530. If this is not the case, in one step 704, the first trajectory 501 is selected. If the predefined probability of collision P 530 is exceeded, then one of the following actions is performed by the monitoring module 600 (cf. FIG. 1): in one step 705, a further trajectory 502, 503, 504 having a further probability of collision 522, 523, 524 and a maximum deceleration 542, 543, 544 of the further trajectory 502, 503, 504 is determined by means of the planning module 500 in reaction to the object representation 420. Or in one step 706, the object representation 420 of the object 120 is checked by means of the sensor 300.