Methods for at least partially automated operation of a vehicle and vehicle

Abstract

A method for operating an ego-vehicle is proposed, which is configured to receive information from other vehicles in its vicinity for automated or at least semi-automated operation. According to the invention, monitored areas of the ego-vehicle's environment are determined by means of sensors in the other vehicles in the ego-vehicle's vicinity, and these areas are aggregated to form a representation of the overall monitored area. A condition for activating at least semi-automated operation using the information received from the other vehicles is that the overall area covers a predetermined minimum value of the vehicle's environment.

Description

[0001] The present invention relates to a method for the at least partially automated operation of a vehicle and a vehicle for carrying out the method.

[0002] From DE 10 2021 209 541 A1 a method is known in which an ego vehicle networked with other vehicles receives information, plans, maneuvers, trajectories or driving corridors of other vehicles for its own maneuver coordination.

[0003] DE 10 2019 209 701 A1 describes systems and methods for shared autonomy through cooperative sensing. According to one embodiment, a cooperative sensing system includes a rendezvous module that receives broadcast messages from a plurality of cooperating vehicles on the roadway. The rendezvous module selects a subordinate vehicle from the plurality of cooperating vehicles based on the autonomy level of the subordinate vehicle compared to an autonomy level of a master vehicle. The cooperative sensing system also includes a positioning module that determines a cooperative position of the master vehicle and the subordinate vehicle. The cooperative sensing system further includes a negotiation module that receives at least one cooperating parameter from the subordinate vehicle.The cooperative sensing system includes a perception module that initiates cooperative automation with the subordinate vehicle according to the at least one cooperating parameter when the master vehicle and the subordinate vehicle are positioned in the cooperative position.

[0004] DE 10 2022 003 165 A1 discloses a vehicle assistance system for the at least partially autonomous control of a vehicle, comprising a vehicle computing unit configured to generate a model of the vehicle in spatial relation to objects detected in the sensor data based on objects detected by the vehicle's sensors. The vehicle is part of a vehicle fleet in a vehicle network, which further comprises an off-board server that communicates with fleet vehicles of the vehicle fleet. The fleet vehicles are also equipped with sensors for providing sensor data, including at least the position of the respective fleet vehicle and objects in its surroundings.The server is configured to merge the sensor data received from the fleet vehicles and, based on the merged sensor data, to generate and transmit relative position data of fleet vehicles and other objects in the vehicle's surroundings.

[0005] The object of the present invention is to provide an improved method that enables a host vehicle to operate safely and at least partially automated using information from other vehicles. The host vehicle can then be fully or partially equipped with the sensors required for automated operation.

[0006] The object is achieved by a method having the features of claim 1 and a vehicle according to claim 8. The dependent claims define preferred and advantageous embodiments of the present invention.

[0007] In the method, monitored areas of the ego vehicle's surroundings are determined using sensors of the other vehicles in the ego vehicle's vicinity, and the areas are aggregated to represent a monitored overall area. As a condition for activating at least partially automated operation of the ego vehicle using the information received from the other vehicles, it must be met that the overall area covers a predetermined minimum value of the vehicle's surroundings. The other vehicles networked with the ego vehicle use sensors to determine information received from the ego vehicle. The information includes data required for automated driving of the ego vehicle, for example objects in the ego vehicle's surroundings. In order for at least automated driving to be carried out with sufficient safety, it must be ensured that the areas around the ego vehicle are adequately covered by the other vehicles.To do this, the other vehicles in the vicinity of the ego vehicle each determine the areas in the vicinity of the ego vehicle that are covered by their sensors. The ego vehicle exchanges GPS data with the other vehicles, and based on their relative positions, the other vehicles determine which vehicles in the vicinity need to be monitored. In one variant, data exchange between the vehicles can limit monitoring to those vehicles that are configured for automated driving but have no or insufficient sensor technology.

[0008] The monitored area is the area visible from another vehicle in the vicinity of the ego vehicle. This area is determined by the distance and orientation of the sensors of the other vehicle to the ego vehicle (extrinsic sensor parameters) as well as by the properties of the sensors themselves (intrinsic sensor parameters). Furthermore, when determining the monitored area, the obstruction of the view by the ego vehicle itself and by objects located between the ego vehicle and the other vehicle is taken into account. The areas covered by the detected objects are determined using methods known from environment or image recognition, for example.

[0009] The monitored areas are further aggregated to form a representation of the overall monitored area surrounding the ego vehicle. The representation can be designed as a visibility map similar to a bird's-eye view. The representation indicates which areas around the ego vehicle can be viewed and monitored via the sensors of other vehicles and, if applicable, those of the ego vehicle. Automated driving of the ego vehicle is only permitted if the overall monitored area covers a minimum value of the vehicle's surroundings; otherwise, automated driving cannot be reliably guaranteed because there may be undetected objects in the vicinity of the ego vehicle. The minimum value can, for example, be defined as a portion of a monitored area at a specified distance around the vehicle contour and / or a monitored angular range around the ego vehicle.In other words, the unmonitored area in a defined zone around the vehicle contour must not exceed a threshold. For example, it can be specified that a predetermined percentage of, say, 90% of the area must be monitored as a minimum value in a band of a given width around the vehicle contour, and / or that the area must be monitored at a minimum distance around the vehicle. In other words, this means that the area must be monitored up to a distance of at least 0.3 m from the vehicle contour, for example.

[0010] According to the invention, the ego vehicle receives data on all detected objects in the vicinity of the ego vehicle from the other vehicles as soon as the overall area covers a predetermined minimum value of the vehicle's surroundings. Only when it is ensured that the surroundings of the ego vehicle are sufficiently well monitored will further information on objects in the vicinity be transmitted. As soon as the overall monitored area covers a predetermined minimum value of the vehicle's surroundings, the ego vehicle can, for example, provide feedback to the other vehicles and request and receive data or information on objects in the vicinity, in particular objects that are not networked or do not have any sensors. The load on the data channels is advantageously minimized because data on objects is only transmitted if it is also used for automated driving.

[0011] In a further developed embodiment, the ego vehicle receives the areas monitored by the other vehicles to aggregate the overall area. The ego vehicle is configured to receive the areas received from the other vehicles and generate the aforementioned representation from them. Preferably, the other vehicles transmit the monitored areas upon request from the ego vehicle. By generating the overall monitored area in the ego vehicle, the sensor data of the ego vehicle can advantageously be incorporated with largely zero latency.

[0012] According to one embodiment of the present invention, the information received by the ego vehicle from the other vehicles comprises at least one of the following: - Positions, speeds, acceleration values, trajectories, orientation, or signal outputs of the other vehicles, - Positions, speeds, acceleration values, trajectories, orientation, or signal outputs of objects detected by other vehicles in the vicinity of the ego vehicle. This data enables the ego vehicle to fully observe its surroundings without or without the use of its own sensors, enabling the calculation of a driving trajectory in automated driving mode. By additionally using data from the ego vehicle's own sensors, driving safety can be further increased through plausibility checks.

[0013] In a further embodiment of the invention, at least one of the following conditions must be met for the automated process: - Driving speed of the ego vehicle and / or the other vehicles are within a specified range, - Distance of the ego vehicle to other vehicles or road users is above a specified value, - permissible weather conditions or - Driving on a route approved for at least partially automated driving. The additional conditions advantageously further increase the safety of automated driving.

[0014] In an embodiment of the invention, the sensors for detecting the monitored areas of the host vehicle's surroundings are determined using lidar sensors, ultrasonic sensors, cameras, and / or radar sensors. Advantageously, the data from different sensors is merged and preferably fed into a trained neural network for evaluation.

[0015] In one embodiment, the mutual positions of the other vehicles and the ego vehicle are determined from GPS data, wherein the monitored areas of the ego vehicle's surroundings are determined based on the mutual positions and intrinsic parameters of the sensors of the other vehicles. The GPS data is exchanged among the networked vehicles, wherein the position of the sensors on the vehicles is defined from the positions of the vehicles, and the areas visible around the ego vehicle can be determined using sensor parameters such as focal length or depth of field. To determine the respective actually monitored area of ​​the ego vehicle, originally visible areas around the ego vehicle that are obscured by objects must be taken into account.

[0016] In further additional or alternative embodiments, monitored areas of the surroundings of the ego vehicle are determined using sensors of the ego vehicle and aggregated into the representation of the overall area monitored by the other vehicle. The sensors present in the ego vehicle are advantageously used to aggregate the areas monitored by these sensors together with the monitored areas of the other vehicles into the most complete representation possible of the monitored overall area. For example, the monitored overall area of ​​the other vehicles does not cover an area near the mirrors of the ego vehicle. Sensors installed in the mirrors of the ego vehicle can detect this missing area and supplement this area missing from the overall area in the aggregation.

[0017] The vehicle according to the invention is configured to carry out the method described above. The vehicle has means for receiving data from other networked vehicles concerning monitored areas of the vehicle's surroundings.

[0018] The vehicle has further means for aggregating the monitored areas in the environment and for determining a coverage level of a monitored overall area. If the monitored overall area meets a predetermined minimum coverage value, the further means output a signal to enable automated driving. The ego vehicle previously mentioned for the method is referred to here for the sake of simplicity as a vehicle. The mentioned means are designed as a control unit for calculating the overall area and the coverage level. The control unit is a hardware device with a processor, preferably embodied as an integrated circuit, which performs computing operations with the aid of data and / or programs stored, for example, in a ROM (Read Only Memory), RAM (Random Access Memory) and / or FlashROM memory.The output release data are temporarily stored in a memory of the control unit as required and output from the processor via an internal bus system, preferably to a control unit for automated driving.

[0019] The present invention is explained in more detail below using exemplary embodiments with reference to the accompanying drawings, in which the same or similar parts are designated by the same reference numerals.

[0020] It shows: Fig. 1 a schematically depicted traffic situation with an ego vehicle and other vehicles and objects, Fig. 2 the vehicles and objects from Fig. 1 in another traffic situation and Fig. 3 Flowchart of the method according to the invention.

[0021] Figure 1 shows an ego vehicle 1 connected to other vehicles 3a and 3b, hereinafter referred to as 3a,b, via the radio network 5, also known as Car2X, indicated by a symbol. Further objects depicted are a non-networked vehicle 7, a two-wheeler 9, and a pedestrian 11 at a pedestrian crossing 12. All road users 1, 3, 7, 9, 11 are moving on or next to a two-lane road 13.

[0022] Vehicles 3a, b are equipped with sensors for environmental monitoring (not shown in detail). The detection angle of the sensors of vehicle 3a in the rear area is represented by lines 15a, and the detection angle of the sensors of vehicle 3b by lines 17a. The diagonally hatched area 18 between vehicles 3a, b is monitored as an entire area by the sensors of vehicles 3a, 3b. Indicated by the dashed lines 15b and 17b, the overall monitoring area 18 is limited by the contour of the host vehicle, so that unmonitored areas 21, shown without hatching and referred to below as dead zones, are formed next to the host vehicle. The maximum distance 23 of the dead zone from the host vehicle is indicated by arrow 23.This distance specifies a minimum value which means that for automated driving of the ego vehicle, the monitored total area must reach at least a distance of 23 from the ego vehicle.

[0023] The ego vehicle 1 is equipped with a system for automated driving, which includes a control unit for controlling actuators such as the steering, brakes, or signal transmitters. The control unit processes information about objects 3a, 3b, 7, 9, and 11 in the vehicle's surroundings as input signals. The ego vehicle 1 itself has no or insufficient sensors to capture all the information in the surroundings required for automated driving.

[0024] The ego vehicle 1 receives information required for automated driving from the other vehicles 3a, 3b via the radio network 5. This information includes dimensions, positions, speeds, acceleration values, trajectories, orientation, and / or signal outputs of the other vehicles 3a, 3b themselves as well as about the objects 7, 9, 11, which in turn are detected by the sensors of the other vehicles 3a, 3b.

[0025] The procedure for activating the automated driving mode of the ego vehicle 1 runs according to Fig. 3 in the following steps. In principle, the ego vehicle 1 is in permanent contact with the surrounding vehicles 3a, 3b via the network 5.

[0026] In a first step 100, the ego vehicle receives the area in its surroundings monitored by the other vehicles 3a, 3b. Vehicle 3a detects the area in front of, and vehicle 3b detects the area behind, the ego vehicle 1. The areas are determined from geometric data of the sensors, such as the detection angle or detection range, and are converted to the position of the ego vehicle 1 based on the relative positions of the vehicles. In this process, coverage within the detection range of the sensors of the other vehicles 3a, 3b caused by the geometry of the ego vehicle 1 is taken into account.

[0027] In a further step 110, the ego vehicle 1 aggregates a monitored total area 18 and the above-mentioned dead zones 21 from the areas monitored by the other vehicles 3a,b.

[0028] If the monitored total area 18 now has a predetermined minimum value, the ego vehicle 1 reports this information back to the other vehicles 3a,b in step 120, requesting further information on non-networked objects.

[0029] For example, in the present case, the minimum value specifies that the total monitoring area must approach the ego vehicle 1 up to a predetermined distance 23 of, for example, 0.3 m, and thus the dead zones 21 are correspondingly small. In addition, it can be defined, for example, that the total monitoring area extends at least to the dashed line designated by reference numeral 25, which defines a distance to the contour of the ego vehicle 1. In the example of Fig. 1, the total area 18 has the minimum value, whereupon the ego vehicle 1 receives the data required for automated driving operation from the other vehicles 3a,b in step 130. As shown, the minimum value ensures that all objects 7, 9, 11 in the relevant environment of the ego vehicle 1 are recognizable by the other vehicles 3a,b and usable by the ego vehicle 1 for automated driving operation. A control unit responsible for automated driving can activate automated driving operation in step 140 using received information about the vehicles 3a,b and the non-networked objects 7, 9, 11, provided that required conditions such as distances between the vehicles and objects, driving speed, etc., are met.

[0030] In the Fig. 2 is another traffic scenario with the same road users as in Fig. 1, but shown in a different arrangement. Because the vehicle 3a is closer to the ego vehicle 1 than in Fig. 1 is moving, the dead zone 21 is enlarged due to the reduced detection range, as a result of which the monitored total area does not reach the minimum value. The monitoring area approaching the ego vehicle 1, indicated by the arrow 25, no longer reaches the distance 23 specified as at least a minimum value, and the monitored total area 18 therefore does not reach a specified minimum value for covering the vehicle's surroundings. In this case, this has the consequence that although the objects 7, 11 located in the vicinity of the ego vehicle 1 are within the detection range, the vehicle 9 is at least partially located in the dead zone. Automated driving of the ego vehicle 1 would therefore with a high probability cause a collision with the invisible vehicle 9.

[0031] According to the method described above, due to the total monitored area 18 being too small or the dead zones 21 being too large, the ego vehicle 1 would not request any further information from the other vehicles 3a, 3b in order to limit the data volume and would not activate automated driving operation.

[0032] In a variant without taking the data volume into account, the additional vehicles 3a, 3b can, of course, continuously transmit information from monitored areas with the information about the additional objects 7, 9, 11 to the ego vehicle 1. The ego vehicle 1 then has all the information required to decide whether to activate automated driving mode at the same time. In a first step, it is determined whether the total monitored area 18 reaches the minimum value; in a second step, it is checked whether the surrounding objects and environmental conditions permit automated driving mode. If both conditions are met, the ego vehicle 1 will activate automated driving mode, which is implemented using the sensor data for environmental monitoring received from the additional vehicles 3a, b.

Claims

[1] Method for operating an ego vehicle (1) which is designed to receive information for automated or at least partially automated operation from other vehicles (3a,b) in the surrounding area, where monitored areas of the environment of the ego vehicle (1) are determined by means of sensors of the other vehicles (3a,b) located in the environment of the ego vehicle and the areas are aggregated to form a representation of a monitored overall area (18), wherein as a condition for activating the at least partially automated operation by means of the information received from the other vehicles (3a,b), it must be fulfilled that the monitored total area (18) covers a predetermined minimum value (23) of the vehicle's surroundings, characterized by , that as soon as the total area (18) covers a predetermined minimum value of the vehicle environment, the ego vehicle (1) receives data of all detected objects (7,9,11) in the environment of the ego vehicle (1) from the other vehicles (3a,b). [2] Method according to claim 1, characterized by that the ego vehicle (1) receives the areas monitored by the other vehicles (3a,b) for aggregating the total area (18). [3] Method according to one of the preceding claims, characterized by that the information received by the ego vehicle (1) from the other vehicles (3a,b) includes at least one of the following: -positions, speeds, acceleration values, trajectories, orientation, or signal outputs of the other vehicles (3a,b), -Positions, speeds, acceleration values, trajectories, orientation, or signal outputs of the objects detected by the other vehicles in the vicinity of the ego vehicle (1). [4] Method according to one of the preceding claims, characterized by that the sensors of the further vehicle (3a,b) for detecting the areas and for detecting the information required for automated driving comprise lidar sensors, ultrasonic sensors, cameras and / or radar sensors. [5] Method according to one of the preceding claims, characterized by that for at least partially automated driving operation, at least one of the following conditions must be met: -driving speed of the ego vehicle (1) and / or the other vehicles (3a,b) are within a predetermined range, -Distance of the ego vehicle (1) to other vehicles or road users is above a specified value, -permissible weather conditions exist or -Driving on a route approved for at least partially automated driving operations. [6] Method according to one of the preceding claims, characterized bythat the positions of the other vehicles (3a,b) and the ego vehicle (1) are determined from GPS data, wherein the areas of the environment of the ego vehicle (1) detected by the other vehicles (3a,b) are determined on the basis of the positions and intrinsic parameters of the sensors of the other vehicles (3a,b). [7] Method according to one of the preceding claims, characterized by that monitored areas of the environment of the ego vehicle (1) are determined by means of sensors of the ego vehicle (1) and aggregated with the representation of the total area (18) monitored by the other vehicles (3a,b). [8] Vehicle (1) equipped to carry out the method according to one of the preceding claims with Means for receiving data from other networked vehicles (3a,b) concerning monitored areas of the surroundings of the vehicle (1), further means for aggregating the monitored areas into a monitored overall area (18) in the surroundings of the vehicle (1), provided that the monitored total area (18) meets a predetermined minimum coverage value, the further means output a signal to enable automated driving, characterized by , that as soon as the total area (18) covers a predetermined minimum value of the vehicle surroundings, the vehicle (1) receives data of all detected objects (7,9,11) in the surroundings of the ego vehicle (1) from the other vehicles (3a,b).

Images