Method for providing an assistance signal and/or a control signal for an at least partially automated vehicle

Abstract

A method for providing an assistance signal and/or a control signal for an at least partially automated vehicle includes receiving surroundings data, in particular an acoustic signal; recognizing a warning signal emitted by a further road user based on the received surroundings data; determining whether a hazardous situation relating to the vehicle is or was indicated by the warning signal; and providing the assistance signal and/or the control signal based on the result of the determination.

Claims

1. A method for providing at least one of an assistance signal and a control signal for a vehicle that is at least partially automated, the method comprising: receiving surroundings data; recognizing a warning signal emitted by a warning signal actuator of a further road user based on the received surroundings data; determining whether a hazardous situation relating to the vehicle is or was indicated by the warning signal, the hazardous situation including at least one of (i) a hazardous state of the vehicle and (ii) a hazardous action of the vehicle that is planned or already carried out, the determining including: ascertaining whether damage to the vehicle is present that promotes the hazardous situation; ascertaining whether a state of the vehicle is present that promotes the hazardous situation; and ascertaining whether a driving maneuver of vehicle that is planned or already carried out promotes the hazardous situation; and providing the at least one of the assistance signal and the control signal based on whether the hazardous situation is or was indicated by the warning signal.

2. The method according to claim 1, wherein the surroundings data comprise an acoustic signal, and the method further comprises: determining a direction from which the acoustic signal was received.

3. The method according to claim 2, further comprising: determining a distance to a source of the acoustic signal.

4. The method according to claim 1, further comprising: ascertaining the further road user which has output the warning signal.

5. The method according to claim 1, wherein the surroundings data comprise sensor data of imaging sensors, and the method further comprises: ascertaining the further road user which has output the warning signal based on the sensor data of the imaging sensors.

6. The method according to claim 1, further comprising: emitting the assistance signal to a teleoperator; and operating the vehicle based on a signal received from the teleoperator.

7. The method according to claim 1, further comprising: terminating a driving action of the vehicle that is planned or already started, based on the control signal.

8. The method according to claim 1, wherein a device is configured to execute the method.

9. The method according to claim 1, wherein the signal actuator of the further road user is at least one of (i) an optical warning signal actuator and (ii) an acoustic warning signal actuator.

10. The method according to claim 9, wherein the warning signal is at least one of (i) a flashing light and (ii) a horn sound.

11. The method according to claim 1, wherein the hazardous state of the vehicle at least one of (i) a fire of the vehicle, (ii) an open door of the vehicle, (iii) an open tank lid of the vehicle, (iv) a flat tire of the vehicle, (v) a destroyed tire of the vehicle, (vi) a defective light of the vehicle, and (vii) liquids leaking from the vehicle.

12. The method according to claim 1, wherein the hazardous action of the vehicle includes at least one of (i) changing lanes, (ii) passing, (iii) stopping, (iv) braking, (v) exiting, (vi) turning, and (vii) accelerating.

13. A non-transitory machine-readable storage medium that stores a computer program for providing at least one of an assistance signal and a control signal for a vehicle that is at least partially automated, the computer program including commands that, when executed by a computer, cause the computer to: receive surroundings data; recognize a warning signal emitted by a warning signal actuator of a further road user based on the received surroundings data; determine whether a hazardous situation relating to the vehicle is or was indicated by the warning signal, the hazardous situation including at least one of (i) a hazardous state of the vehicle and (ii) a hazardous action of the vehicle that is planned or already carried out, the determination including: ascertaining whether damage to the vehicle is present that promotes the hazardous situation; ascertaining whether a state of the vehicle is present that promotes the hazardous situation; and ascertaining whether a driving maneuver of vehicle that is planned or already carried out promotes the hazardous situation; and provide the at least one of the assistance signal and the control signal based on whether the hazardous situation is or was indicated by the warning signal.

Description

BRIEF DESCRIPTION OF THE DRAWING

(1) The FIGURE shows a schematic method diagram.

DETAILED DESCRIPTION

(2) According to the German Road Traffic Act, the horn can only be used as a warning instrument in hazardous situations. Against this background, it is useful for an at least partially automated vehicle to be able to perceive, interpret, and react to these items of information. In principle, such warning signals can be used by the following traffic for multiple reasons:

(3) A first reason can be, for example, warning of the leading driver about a hazard directly relating to the leading driver (automobile smoking, tire is flat, lights are not working, planned lane change represents a hazard, vehicle is departing from the lane, . . . ).

(4) A further reason can be, for example, that the following driver wishes to draw attention to themselves due to their own urgency, for example, to go forward faster (for example, emergency in the vehicle or a tailgater).

(5) The present patent application primarily relates to the presence of the first mentioned reason, the presence of a hazardous situation relating to the vehicle itself.

(6) In a first exemplary embodiment, a vehicle operated in a fully automated manner, i.e., a vehicle operated at least temporarily without an intervention of a driver, is equipped with multiple surroundings sensors, among them cameras, lidar sensors, radar sensors, ultrasonic sensors, and microphones. The vehicle is operated by a central control unit based on the data acquired by means of the surroundings sensors.

(7) The method for providing an assistance signal and/or a control signal outlined in the FIGURE, which starts in step 101, runs in the vehicle.

(8) In step 102, surroundings data are received by the vehicle, which comprise in particular camera data and acoustic signals.

(9) In step 103, a warning signal emitted by a further road user is recognized based on the surroundings data. The acoustic signal is analyzed for this purpose and checked for the presence of acoustic warning signals, in particular signals generated by horns. An algorithm which has been trained to recognize horn signals by a machine learning method is used to recognize these horn signals.

(10) In step 104, it is determined whether the warning signal indicates or indicated a hazardous situation relating to the vehicle. For this purpose, two things are checked in this exemplary embodiment.

(11) On the one hand, it is determined whether a change of the vehicle control has occurred in a defined period before the acquisition of the warning signal, that is to say whether a change of the lateral or longitudinal acceleration or speed has occurred and/or whether a planned driving maneuver was initiated or is to be initiated. In this case, it is also checked whether a vehicle light or the like was already activated to initiate the maneuver.

(12) If this check is positive, i.e., a change has occurred, it is ascertained based on an evaluation of the current traffic situation on the basis of the surroundings data whether a hazard exists to the vehicle and/or further road users due to the change.

(13) Moreover, a vehicle diagnosis is carried out in which a review of possibly open doors, tank lids, damaged tires, the temperature, the brake pressure, and the like is carried out. The review comprises in particular a check of the functionality of the sensors, in particular the tire pressure sensors. Moreover, image data from the interior are evaluated to recognize fires or smoke development in the vehicle interior.

(14) If it is determined in this case that a defect or a hazardous situation exists, the determination of whether a hazardous situation was indicated is positive. This also applies if it is determined that a hazard to the vehicle and/or further road users exists due to the change or planned change of the vehicle control.

(15) In alternative embodiments, a check of only one of the points/things mentioned can also be carried out.

(16) Based on the result of the determination, in step 105, the assistance signal and/or the control signal is provided.

(17) If the determination is positive, a corresponding signal is thus output in dependence on the present hazardous situation. If it is determined that a hazard could arise due to a planned driving maneuver, a corresponding control signal is output to terminate the planned maneuver.

(18) If it was determined that the vehicle control was changed or a change is planned and it cannot be ascertained whether a hazardous situation was indicated, the planned change is thus initially not carried out or already initiated changes are terminated, if they are not necessary to ensure the safety of the operation of the vehicle. A corresponding control signal is also provided for this purpose.

(19) If it is determined, as in this exemplary embodiment, that smoke has developed in the vehicle interior, an assistance signal to a teleoperator is provided in order to examine the situation in greater detail. Moreover, based on the assistance signal, a message is output to any vehicle occupants present in the vehicle, in which the smoke development is indicated.

(20) The method ends in step 106.

(21) Further embodiments and designs of the method for providing an assistance signal and/or a control signal for an at least partially automated vehicle are described by way of example hereinafter.

(22) In a vehicle operated in a partially automated manner, referred to as ego hereinafter, which is equipped with different surroundings sensors, among them cameras, lidar sensors, radar sensors, ultrasonic sensors, and microphones, the following method steps are carried out.

(23) An acoustic warning signal output by a further vehicle is perceived by means of the microphones of the ego. Via the attachment of at least two microphones on the ego, a direction determination of the acoustic signal can be achieved via the correspondingly recorded signal characteristic. A probability can thus be ascertained with which the evaluated signal can be associated with the following traffic. A probability of a distance of the signal-triggering vehicle from the ego can also be ascertained via the signal strength and stored table values.

(24) Furthermore, the surroundings of the ego are sensed and following vehicles are detected and classified by means of the supplementary sensors (lidar, video, radar).

(25) With the aid of the distance and items of direction information of the acoustic sensors and the further object detection, a data fusion is performed and the probability is ascertained as to which of the following vehicles has triggered the acoustic signal.

(26) In an alternative embodiment, it is ascertained using imaging sensors whether one of the following vehicles also activates a light signal (low beams, warning flashing light). These items of information are supplemented with the above-described sensor data fusion and a probability is ascertained that one of the objects, which has possibly activated an acoustic warning signal, has also simultaneously activated a light signal or has activated and deactivated it repeatedly or at high frequency.

(27) If it should have been established with a certain probability in the case of a situation with a following vehicle that at least the directly following vehicle or the vehicle traveling directly in the parallel lane has triggered an acoustic warning signal, a probability with which the following vehicle wishes to notify the ego of a hazardous situation is furthermore ascertained.

(28) Possible causes of an existing hazardous situation are divided into at least two cause categories for the ascertainment of such a probability:

(29) a) ascertaining whether following traffic wishes to indicate a hazardous situation on the ego (flat tire, mechanical defect of another type, smoke development, open luggage compartment, . . . )

(30) b) ascertaining whether following traffic wishes to indicate a hazardous situation which is triggered by the planned/current driving action of the ego (for example, commenced lane change provokes uncomfortable behavior in the follower, . . . )

(31) The probability for a) can be described on the basis of the evaluation of the currently upcoming traffic situation and further vehicle sensors. Known methods for modeling and/or machine learning are suitable as methods. For example, the following steps can be carried out: ascertaining that the ego behaves conforming to the rules in accordance with the current traffic conditions (maintains applicable maximum speed, travels in the correct lane, no further vehicle is traveling ahead, no traffic junction is upcoming, . . . ) furthermore ascertaining whether a defect has resulted on the vehicle (check error messages to light signals, items of tire pressure information, further items of maintenance information . . . ) ascertaining a probability about an externally visible defect on the vehicle.

(32) The probability for b) is ascertained in an expanded form as to whether the ego has indicated or already started a change of a driving action in the chronological surroundings of the acoustic warning signal. This includes, for example, starting a lane change, traveling through traffic junctions, and the like.

(33) A further evaluation dimension results while driving on the ego lane with respect to a significant change of the longitudinal speed.

(34) In this case, a probability of a hazardous driving situation which could result from the viewpoint of others due to a driving action of the ego which has been initiated or begun is ascertained.

(35) It is possible to react to a hazardous situation according to situation a) using the following measures: A teleoperator is incorporated (by providing an assistance signal) to perform a possible visual check by way of sensor data. A contact to a teleoperator and/or any passenger present is established (by providing an assistance signal), so that they can evaluate the situation. The ego can be steered to a closest parking place, on the basis of a provided control signal. Any passenger present can carry out an external visual check there. Moreover, a maintenance team can be sent to the stopping point by the provision of an assistance signal. The ego can be returned to the depot by a corresponding control signal.

(36) It is possible to react to a hazardous situation according to situation b) using the following measures: The ego ends the initiated or started driving action and returns to the original driving action. The ego reacts in the original driving situation in accordance with the upcoming traffic situation.