DE102020003667A1 - Method for operating an automated vehicle - Google Patents

Abstract

The invention relates to a method for operating an automatically driving vehicle (1), with future routes of other vehicles in the vicinity of the vehicle (1) being taken into account when determining a trajectory of the vehicle (1). According to the invention it is provided that - buses (3) and trucks in the vicinity of the vehicle (1) are detected by sensors, - positions of vehicle wheels (3.1) of the buses (3) and trucks are determined relative to a vehicle geometry of the buses (3) and trucks - based on the position of the vehicle wheels (3.1) relative to the vehicle geometry, at least one towing curve (S1 to S4) relevant to the vehicle (1) of the buses (3) and trucks is determined and - if the at least one determined towing curve is likely to overlap (S1 to S4) with a lane (F1) of the vehicle (1), this overlap is taken into account when planning the trajectory of the vehicle (1).

Description

The invention relates to a method for operating an automatically driving vehicle, with future routes of further vehicles located in the vicinity of the vehicle being taken into account when determining a trajectory of the vehicle.

From the DE 10 2019 001 956 A1 a method for operating an automated, in particular highly automated or autonomously driving vehicle is known. A vehicle environment is recorded and a trajectory plan is used to predict an area of the vehicle environment that the vehicle is likely to sweep during a journey. In addition, edge areas adjoining a lane or lane, which can be covered by body overhangs of the vehicle and / or protruding load of the vehicle, the maximum height of which is less than a minimum height of a free space below the corresponding body overhang and / or the corresponding load, are determined. In the case of an automated drive of the vehicle, the trajectory planning is carried out in such a way that the vehicle wheels do not leave the lane and the body overhangs and / or the load, if necessary, sweep over the border areas that can be passed over.

The invention is based on the object of specifying a method for operating an automated vehicle.

The object is achieved according to the invention by a method which has the features specified in claim 1.

Advantageous embodiments of the invention are the subject of the subclaims.

A method for operating an automatically driving vehicle provides that, when determining a trajectory of the vehicle, future routes of further vehicles located in the vicinity of the vehicle are taken into account. According to the invention, buses and trucks in the vicinity of the vehicle are sensed and positions of vehicle wheels of the buses and trucks relative to a vehicle geometry of the buses and trucks are determined. Based on the position of the vehicle wheels relative to the vehicle geometry, at least one towing curve relevant for the vehicle is determined for the buses and trucks, and if the at least one determined towing curve is expected to overlap with a lane of the vehicle, this overlap is taken into account when planning the trajectory of the vehicle.

By using the method, a comparatively complex driving space stress, in particular of buses and trucks, is determined and taken into account when planning the trajectory of the vehicle, whereby an improved and safe defensive-cooperative driving strategy of the vehicle can be implemented.

Embodiments of the invention are explained in more detail below with reference to drawings.

• 1 schematisch ein automatisiert auf eine Kreuzung zufahrendes Fahrzeug und ein vergrößertes Bild einer fahrzeugseitigen Kamera,
• 2 schematisch einen auf eine Fahrbahn des Fahrzeuges abbiegenden Bus zu einem ersten Zeitpunkt,
• 3 schematisch den auf die Fahrbahn des Fahrzeuges abbiegenden Bus zu einem zweiten Zeitpunkt,
• 4 schematisch den auf die Fahrbahn des Fahrzeuges abbiegenden Bus zu einem dritten Zeitpunkt,
• 5 schematisch den auf die Fahrbahn des Fahrzeuges abbiegenden Bus zu einem vierten Zeitpunkt,
• 6 schematisch einzelne Schleppkurven des Busses beim Abbiegen auf die Fahrbahn des Fahrzeuges,
• 7 schematisch eine Recheneinheit des Fahrzeuges mit einzelnen Funktionseinheiten und
• 8 schematisch Fremdfahrzeug-Prädiktionsmodul.

Show:

• 1 schematically a vehicle automatically approaching an intersection and an enlarged image from a camera on the vehicle,
• 2 schematically a bus turning onto a lane of the vehicle at a first point in time,
• 3 schematically the bus turning onto the lane of the vehicle at a second point in time,
• 4th schematically the bus turning onto the lane of the vehicle at a third point in time,
• 5 schematically the bus turning onto the lane of the vehicle at a fourth point in time,
• 6th schematic individual towing curves of the bus when turning onto the vehicle's lane,
• 7th schematically a computing unit of the vehicle with individual functional units and
• 8th Schematic foreign vehicle prediction module.

Corresponding parts are provided with the same reference symbols in all figures.

1 shows an automated on an intersection K approaching vehicle 1 and an enlarged image of a vehicle-side detection sensor system 2 , one being at the intersection K standing bus 3 intends to hit a roadway F. of the vehicle 1 turn.

Automated vehicles will be used in the future 1 to be found increasingly in road traffic. Determining your own free driving space is necessary for planning a trajectory for the vehicle 1 essential. It is comparatively important to predictively assess other road users in such a way that their foreseeable driving space requirements on a roadway F. is correctly assessed.

A vehicle typically has a certain wheelbase A, the wheelbase A being a distance between two axles of the vehicle 1 designated. In addition, the vehicle has 1 a certain length I, with a vehicle body in relation to the vehicle wheels 1.1 protrudes on the front and rear. Such a front-side body section is referred to below as a protruding front-side area B1 and a rear-side body section is referred to below as a protruding rear-side area B2.

Drives the vehicle 1 around a curve, sweep over corner areas of the protruding areas B1, B2 roadway areas, these roadway areas as drag curves S1 to S4 which are exemplified in 6th are shown.

The towing curves S1 to S4 are themselves when the vehicle is traveling straight ahead 1 very similar, but when cornering, the drag curves show S1 to S2 different dimensions.

The larger the protruding areas B1, B2, the more different are the drag curves S1 to S4 when cornering.
Especially with a bus 3 the dimensions in relation to length I and wheelbase A can be considerably different, so that the protruding areas B1, B2 are comparatively large or comparatively long.

About such drag curves S1 to S4 when creating a trajectory of the vehicle 1 a procedure described below is provided.

As described above, the vehicle has 1 the detection sensors 2 on, which can be designed as a lidar-based sensor, as a stereo camera and / or as a mono camera and the one with a vehicle-mounted computing unit 4th connected is.

Based on the signals recorded by the detection sensors 2 becomes an area in front of the vehicle 1 measured, with semantic data of an environment of the vehicle 1 , ie an infrastructure, further vehicles, ie third-party vehicles, pedestrians, traffic lights, traffic signs and other objects are determined, which are necessary for a collision-free automated driving of the vehicle 1 are relevant.

With respect to third-party vehicles, these are found in passenger cars, trucks, motorcycles, bicycles and buses 3 distinguished.

Will be a bus 3 or, alternatively, a truck that is used to plan a trajectory of the vehicle 1 is relevant, for example in the 1 to 5 shown are the positions of the vehicle wheels 3.1 , in particular relative to a vehicle geometry determined.

The exemplary embodiments described below relate to a bus 3 , whereby the method can be used analogously for trucks.

Based on the position of the vehicle wheels 3.1 of the bus 3 relative to the vehicle geometry are those for the vehicle 1 relevant drag curves S1 to S4 determined.

Will the drag curves S1 to S4 then into an environment modeling of the vehicle 1 included, the determined drag curves can overlap S1 to S4 with one lane F1 of the vehicle 1 to be determined. It can therefore be determined which area of the lane F1 of the vehicle 1 from the protruding areas B1, B2 of the bus 3 be slipped on.

Such information is used when planning the trajectory of the vehicle 1 taken into account so that it can be adjusted in a defensive manner.

According to the in 1 In the present embodiment, the vehicle is running 1 on the intersection K too, with the bus standing there 3 if necessary, start immediately and onto the road F. of the vehicle 1 turns. The vehicle 1 drives defensively, whereby it is therefore necessary to predict how a separate driving area by swiveling the bus out 3 restricted, ie restricted, is.

The 2 to 5 point to the road F. of the vehicle 1 turning bus 3 each at a further point in time t1 to t4.

The bus travels at a first point in time t1 3 straight into an area of the intersection K one, with the front vehicle wheels 3.1 are turned in at a second time t2 and the bus 3 in sections with its front protruding area B1 on the lane F1 of the vehicle 1 is located. The bus is at a third point in time t3 3 almost completely on the road F. bent, with the front protruding area B1 and a front vehicle wheel 3.1 in the lane F1 of the vehicle 1 are located. The bus is at a fourth point in time t4 3 turned and drives on an opposite lane F2 of the vehicle 1 .

The bus 3 must, as in the 2 to 4th is shown with its rear right vehicle wheel 3.1 Driving around the curve, this and a maximum steering angle of the bus being decisive for a route of the bus 3 contribute.

In 6th are the drag curves S1 to S4 from corner areas of the protruding areas B1, B2 of the bus 3 shown.

A most sweeping first drag curve S1 is to be assigned to a left corner area of the protruding area B1 at the front, whereby this first drag curve S1 far into the lane F1 of the vehicle 1 extends so that it is slipped over or overlapped.

A second drag curve S2 is to be assigned to a right corner area of the protruding area B1 at the front, with this second drag curve S2 up to a lane marking M. between the lane F1 and the opposite lane F2 extends.

7th shows the computing unit as an example and in a greatly simplified manner 4th of the vehicle 1 , the signaling with the detection sensors 2 connected is.

Is the detection sensors 2 designed as a three-dimensional recording is a quality of an analysis and calculation of the drag curves S1 to S4 comparatively high, with the analysis and determination of the overlap of the drag curves S1 to S4 with the lane F1 of the vehicle 1 can be done after a data fusion.

In the arithmetic unit 4th image processing takes place BV by means of the detection sensors 2 detected signals, with the processed signals, ie with fusion-relevant data, a fusion Fu with map data KD , recorded signals from acceleration sensor 5 , Rotation rate sensors 6th and other suitable sensors.

The image processing BV is carried out in particular in an image processing unit, not shown in more detail, which comprises a stereo module, a semantics module and further modules. The semantics module can be based on at least one neural network that was trained with vehicle / vehicle wheel information. Alternatively or additionally, the vehicle 1 a dedicated wheel detector.

Furthermore, the Fusion Fu is made with recorded signals from an ultrasound-based sensor 7th , a radar-based sensor 8th and a lidar-based sensor 9 carried out, one by means of a satellite-based position determination unit 10 determined current position of the vehicle 1 and information from a central computer unit 11 be taken into account in the merger Fu.

The merged data is sent to an in 8th Other vehicle prediction module shown in more detail 12 and a behavior planning module 13 fed, which a trajectory planning module 14th includes.

The by means of the trajectory planning module 14th created trajectories for the further automated driving of the vehicle 1 taking into account the towing curves of the bus 3 become an actuator 15th of the vehicle 1 fed.
By means of the actuators 15th a steering, a driving stability device, a brake, a sound signal generating device, a turn signal, etc. are controlled.

In 8th are the foreign vehicle prediction module 12 and a process sequence carried out by means of this is shown.

The foreign vehicle prediction module 12 vehicles, ie also detected buses, are detected using the Fusion Fu 3 and trucks, and a course of the lane F1 of the vehicle 1 fed.

In the foreign vehicle prediction module 12 a process sequence is stored, which for all detected other vehicles, but in particular for comparatively strong swinging other vehicles such. B. trucks, articulated buses, and buses 3 , is performed. The following describes the process flow with regard to a detected bus 3 described. In the case of decisions to be made during the course of the procedure, a yes with a y and a no with an n are indicated.

In a first method step V1 it is determined whether the vehicle wheels 3.1 of the bus 3 were segmented. Were the vehicle wheels 3.1 not segmented, the trajectory planning module is used in a second method step V2 14th informs accordingly which one uses a standard model for third-party vehicles, with no separate calculation based on drag curves S1 to S4 is possible.

Were the vehicle wheels 3.1 of the bus 3 , that is of the other vehicle, are segmented, the drag curves are in a third method step V3 S1 to S4 of the bus 3 calculated, with the dimensions of the protruding areas B1, B2 of the bus 3 can be used.

If the further vehicle is a truck that includes so-called lifting axles, then these are recognized as such, these being related to the determination of the towing curves S1 to S4 remain unconsidered.

In a fourth method step V4 takes place on the basis of the drag curves S1 to S4 a calculation of a through the protruding areas B1, B2 covered area of the lane F1 of the vehicle 1 , ie an overlap of the drag curves S1 to S4 with the lane F1 .
If such an overlap is determined in a fifth method step V5, in a sixth method step V6 an area, in particular an area, of this overlap is determined and sent to the trajectory planning module 14th transmitted, so that this overlap when determining the trajectory of the vehicle 1 is taken into account.

If it is determined in the fifth method step V5 that there is no overlap of the drag curves S1 to S4 of the bus 3 with the lane F1 of the vehicle 1 is present, the method sequence is carried out for a detected further next vehicle.

List of reference symbols

1

vehicle

1.1

Vehicle wheel

2

Detection sensors

3

bus

3.1

Vehicle wheel

4th

Arithmetic unit

5

Accelerometer

6th

Rotation rate sensor

7th

ultrasound-based sensor

8th

radar-based sensor

9

lidar-based sensor

10

Position determination unit

11

central processing unit

12

Prediction module

13

Behavior planning module

14th

Trajectory planning module

15th

Actuators

A.

wheelbase

B1, B2

protruding area

BV

Image processing

F.

roadway

F1

lane

F2

Oncoming lane

Fu

fusion

K

crossing

KD

Card data

I.

length

M.

Lane marking

S1 to S4

Trailing curve

t1 to t4

time

V1 to V6

Process step

QUOTES INCLUDED IN THE DESCRIPTION

This list of the documents listed by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent literature cited

• DE 102019001956 A1 [0002]

Claims (1)

Method for operating an automatically driving vehicle (1), with future routes of other vehicles in the vicinity of the vehicle (1) being taken into account when determining a trajectory of the vehicle (1), characterized in that - buses (3) and trucks in the surroundings of the vehicle (1) are detected by sensors, - positions of vehicle wheels (3.1) of buses (3) and trucks are determined relative to a vehicle geometry of buses (3) and trucks, - based on the position of the vehicle wheels (3.1) relative to the vehicle geometry at least one for the vehicle (1) relevant tow curve (S1 to S4) of the buses (3) and trucks is determined and - if the at least one determined tow curve (S1 to S4) is likely to overlap with a lane (F1) of the Vehicle (1) this overlap is taken into account when planning the trajectory of the vehicle (1).

Images