Method for operating autonomously controlled vehicle, autonomously controllable motor vehicle and system

Abstract

In planning a coarse trajectory for a road section to be traveled in the future, the invention employs a hierarchical data structure comprising a combination of travel requirements and control command codes, so that it is possible to pre-plan which control command code should be activated / used at which point in time of the trajectory to be traveled.

Description

TECHNICAL FIELD

[0001] The invention relates to a method for operating an autonomously controlled vehicle, an autonomously controllable vehicle and a system comprising a data storage and means for autonomously controlling a vehicle. BACKGROUND

[0002] From CN 110263381 A it is known that for an autonomously driven vehicle a plurality of steps needs to be planned on a trajectory, for example when overtaking, and the planning needs to take into account traffic rules and environmental conditions. For this a logic library can be employed which is built according to a tree principle.

[0003] CN 114261399 A relates to a driving assistance system which plans decisions based on environmental information. There are differences in driving states, for example driving in a lane, changing lanes, etc. The transitions between such states occur directly in the course of driving.

[0004] US 2020 / 0238012 A1 describes a method for managing traffic rules with a data hierarchical architecture. SUMMARY

[0005] It is an object of the invention to make an autonomously controlled vehicle more reliable when moving on a road and to take into account as many requirements as possible, for example traffic rules, judicial precedents, etc.

[0006] This object is achieved by the method for operating an autonomously controlled vehicle according to claim 1, the autonomously controllable motor vehicle according to claim 6 and the system consisting of a data storage and means for autonomously controlling a vehicle according to claim 9.

[0007] Thus, the method for operating an autonomously controlled vehicle according to the invention comprises:

[0008] - providing a hierarchical data structure comprising combinations of driving requirements and control command codes;

[0009] - planning a coarse trajectory for a road section to be driven in the future;

[0010] - determining driving requirements on this coarse trajectory;

[0011] - dividing the trajectory into a plurality of segments, wherein each transition between two segments is assigned a requirement change (and preferably conversely, each requirement change is assigned to a transition between two segments; otherwise, certain driving requirements in the method can be ignored);

[0012] - for each segment, reading a control command code from the hierarchical data structure according to the requirements applicable to this segment;

[0013] - driving through the trajectory in a segmented manner and activating the respective control command code assigned to the segment at this point.

[0014] The invention has two aspects: on the one hand a hierarchical data structure comprising a combination of driving requirements and control command codes (optionally directly usable by the control unit of the vehicle), and on the other hand a pre-planning of the entire trajectory. The distance covered by the pre-planning of the trajectory can preferably be at least 20 meters, further preferably at least 50 meters, still further preferably at least 100 meters, particularly preferably at least 200 meters, very particularly preferably at least 500 meters, and most preferably 1 kilometer. The planning of the rough trajectory can also be determined on the basis of a predetermined length of time, for example 5 seconds in the future, preferably 10 seconds in the future, further preferably 20 seconds in the future, still further preferably 30 seconds in the future, particularly preferably 1 minute in the future, very particularly preferably 2 minutes in the future.

[0015] By pre-planning of the trajectory on the one hand and determination of the requirements and the associated control command codes on the other hand, the autonomously controlled vehicle can achieve particularly fast reactions when a quick response is necessary. For example in the example case of a passing maneuver, information about no passing, speed limit, upcoming bend driving, etc. can be extracted from the navigation map data, so that the corresponding control command codes can be determined in advance by the automated vehicle at this point in time. In this way, it is not necessary to occupy a large amount of computing resources during driving, i.e. in particular during the ongoing passing maneuver, to detect signs and respond quickly. In this respect, the present method is superior to the method described in CN114261399A, which only redefines the state accordingly during driving.

[0016] According to an advantageous embodiment, however, it is also provided accordingly that when driving a segment, it is determined when the requirements change, in order to activate the control command code assigned to the next segment (and in particular, if necessary, also to deactivate the control command code assigned to the previous segment, optionally also only partially).

[0017] In this context, it is also preferably provided that when a change in requirements occurs at a point in time and / or under conditions that do not correspond to the planning, the remaining trajectory is replanned by re-division of the segments. Thus, for example in a passing maneuver, the case can be taken into account that the vehicle to be passed drives faster or slower than expected. Accordingly, the return to the original lane can be delayed or advanced. The conditions that do not correspond to the planning can also determine the point in time or relate to the speed. For example, due to a change in road surface conditions, the autonomously controlled vehicle can not be able to accelerate as quickly and the speed at a certain point in time is lower than the planned speed.

[0018] With regard to the preferred embodiment involving the determination of when a change is required, it can be provided that this is the case when a new traffic sign is detected which introduces or cancels a certain restriction. Here, the sensor arrangement used in the autonomous vehicle for detecting traffic signs can be used to check the map data used at the time of planning and, if necessary, also to determine whether the vehicle is driving as planned.

[0019] The control command code can define an algorithm by which certain parameters which can be measured by measuring devices in the vehicle can be determined. For example, the control command code can specify a minimum speed and / or a maximum speed. The maximum speed can be determined in accordance with a speed limit, and the minimum speed can take into account that the overall traffic flow should be kept smooth and that other road users should not be impeded. Correspondingly, a minimum acceleration and / or a maximum acceleration can also be specified. A limitation on the maximum acceleration can serve, for example, to limit the energy consumption and, if necessary, to facilitate platooning with other vehicles which, for example, follow the current vehicle. A minimum acceleration can be necessary to ensure that there is sufficient time to complete a certain operating process. Correspondingly, the control command code can also specify a minimum rotational speed and / or a maximum rotational speed, wherein, by means of the rotational speed, if necessary in combination with the gear used (in the case of a transmission), the speed can also be determined. Correspondingly, the control command code can also specify a minimum gear and / or a maximum gear. Finally, the control command code can also specify a minimum braking distance and / or a maximum braking distance. A minimum braking distance can be specified in order to prevent a rear vehicle from colliding, and a maximum braking distance can be specified in order to protect other road users in front of the vehicle.

[0020] The autonomously controllable (motor) vehicle according to the application comprises a control unit which can be operated by means of a control command code (i.e. the control command code can be "fed in") and an actuator group which is operated by the control unit. The autonomously controllable vehicle also comprises means for travel planning and a memory in which a hierarchical data structure is provided which comprises a combination of travel requirements and control command codes. The means for travel planning are designed to:

[0021] - plan a coarse trajectory for a route to be traveled in the future;

[0022] - determine travel requirements on the coarse trajectory;

[0023] - divide the trajectory into a plurality of sections, wherein each transition between two sections is assigned a requirement change;

[0024] - for each section, read a control command code from the hierarchical data structure in accordance with the requirements applicable to the section;

[0025] - implement the trajectory in sections and provide the control command code in the control unit.

[0026] Thus, an autonomously controllable motor vehicle carries out the method according to claim 1.

[0027] It is preferred in a motor vehicle that the device for trip planning comprises a self-learning system, which is preferably in the form of a neural network, particularly preferably in the form of a recurrent neural network. Such a system can very well detect complex situations, for example complex situations occurring on the road.

[0028] According to a preferred embodiment, in the hierarchical data structure:

[0029] - in the highest level: a corresponding control command code is assigned for free driving (as there are no restrictions here, at most a general speed limit, for example on a country road);

[0030] - in the next highest level: a corresponding control command code is assigned for a traffic sign with a cancellation of a restriction (placed in the next highest level because after the cancellation of the restriction, in principle free driving can be resumed, so that a return to the highest level is possible);

[0031] - in the third level: a corresponding control command code is assigned for a traffic sign with an introduction of a restriction (for example a location sign, a speed limit, a ban on overtaking, etc., placed in a lower level because such restrictions need to be considered in particular);

[0032] - in the fourth level: a corresponding control command code is assigned for a category of environmental conditions that necessarily leads to a driving restriction (for example, environmental conditions that can be determined according to predetermined criteria, such as "wet road surface", "icy road surface", "poor road surface quality (pits, etc.)", and many more).

[0033] The system according to the invention comprises a data store with a data set comprising a hierarchical data structure defining combinations of driving requirements and associated control command codes for a vehicle control unit, and a device for autonomously controlling a vehicle, which has access to the data set. The above-described content is preferably applicable to the data set.

[0034] The system mentioned is the minimum requirement for carrying out the method according to the invention.

[0035] For application cases or application scenarios that can occur in the method and are not explicitly stated here, it can be provided that an error message is output according to the method and / or a request for inputting user feedback, and / or a default configuration and / or a predetermined initial state is set.

[0036] The application also comprises a control device for a motor vehicle. The control device can comprise a data processing device or a processor device which is set up to execute an embodiment of the method according to the application. To this end, the processor device can have at least one microprocessor and / or at least one microcontroller and / or at least one field programmable gate array (FPGA) and / or at least one digital signal processor (DSP). As a microprocessor, inter alia, a central processing unit (CPU), a graphical processing unit (GPU) or a neural processing unit (NPU) can be used accordingly. Furthermore, the processor device can have a program code which, when executed by the processor device, is set up to execute an embodiment of the method according to the application. The program code can be stored in a data memory of the processor device. The processor circuit of the processor device can have, for example, at least one circuit board and / or at least one system on chip (SoC).

[0037] The application also comprises an improvement of the method according to the application, which has the features as already explained in connection with the improvement of the motor vehicle according to the application. Accordingly, the respective improvement of the method according to the application is not explained here again in detail.

[0038] The motor vehicle according to the application is preferably designed as a car, in particular as a passenger car or a heavy goods vehicle, or as a bus or a motorcycle.

[0039] As a further solution, the application also comprises a computer-readable storage medium which contains commands which, when executed by a computer or a computer cluster, cause the computer or the computer cluster to execute an embodiment of the method according to the application. The storage medium can be designed, for example, at least partially as a non-volatile data memory (for example, a flash memory and / or a solid state drive (SSD)) and / or at least partially as a volatile data memory (for example, a random access memory (RAM)). However, the storage medium can also be operated, for example, as a so-called app store server in the Internet. The computer or the computer cluster can provide a processor unit with at least one microprocessor. The commands can be provided as binary code, as assembly code and / or as source code of a programming language (for example, C).

[0040] The application also comprises combinations of features of the described embodiments. Thus, the application also comprises embodiments which have a combination of features of several of the described embodiments, provided that these embodiments are not stated to be mutually exclusive. BRIEF DESCRIPTION OF DRAWINGS

[0041] Embodiments of the application will be described in more detail below with reference to the drawings, in which:

[0042] Figure 1 Trajectory planning for a passing maneuver is schematically shown; and

[0043] Figure 2 is a flow chart showing the steps of a method according to an embodiment of the application. DETAILED DESCRIPTION

[0044] The embodiments explained below are preferred embodiments of the application. In these embodiments, the described components of the embodiments correspondingly constitute individual features of the application, which should each be considered independently of one another, and which can each also improve the application independently of one another. The disclosure should therefore encompass further combinations than those of the features of the embodiments shown. Furthermore, the described embodiments can also be supplemented by further features of the application already described.

[0045] In the drawings, identical reference numerals identify identical elements.

[0046] Figure 1 The vehicle 1 shown in Fig. 1 comprises a control unit 10, which is coupled to a device 12 for travel planning, wherein actuators 14 can be actuated by the control unit 10 and / or the device 12. A memory 16 is also provided, in which a hierarchical data structure is contained. The hierarchical data structure combines travel requirements with control command codes. The travel requirements are arranged in a hierarchy, so that the control command codes can be read accordingly in relation.

[0047] The device 12 plans a passing maneuver on another vehicle 2 under predetermined conditions. The conditions can for example include that the travel speed of the other vehicle 2 is significantly lower than the speed permitted by the speed limit applicable on the section. Both the vehicle 1 and the vehicle 2 are on a lane 3a, and for the passing maneuver the vehicle 1 must switch to a lane 3b. The device 12 will now plan a trajectory 4: the vehicle 1 has traveled a distance since a start time t Beginn and is at the start of the trajectory 4 at a current time t mom . According to the plan, the vehicle must start to drive off the lane 3a and onto the lane 3b at a time t1. The vehicle reaches the lane 3b at a time t2, must complete the passing maneuver before a time t3, and then returns to the lane 3a in a similar manner until it reaches a destination at a target time t Ziel .

[0048] The method according to the embodiment is based on Figure 2 Starting from step S10, in which a rough trajectory, the trajectory 4 described above, is planned for the section of road to be driven in future, in step S12 the driving requirements on this rough trajectory are determined. For example, on a motorway, it can be specified by means of a variable speed limit (light indication) that the speed limit on the right-hand lane is lower than on the left-hand lane. For example, the speed limit on lane 3a can be set at 80 km / h and the speed limit on lane 3b at 100 km / h. According to step S14, the trajectory is divided into the described sections (between the times t mom , tl, t2, t3, etc.), wherein each transition between two sections is assigned a requirement change. Preferably, each requirement change corresponds to a transition to a new section, i.e. the division of the sections should be as fine as possible in terms of requirements. According to step S16, for each section, the control command code is read from the hierarchical data structure according to the requirements applicable to the section. In this way, the control command code can be planned in advance for the entire trajectory 4 in a section-by-section manner. In step S18, the trajectory is now driven in a section-by-section manner, wherein the control command code assigned to each section is also activated accordingly, i.e. if necessary, written by the device 12 into the control unit 10, so that the control unit 10 accordingly actuates the actuators 14.

[0049] When driving the trajectory, it can optionally be specified that according to A, the division of the sections must be changed by means of the sensor device of the motor vehicle, so that a new division of the sections is carried out according to the re-executed step S14. Alternatively, according to B, it can be returned to step S12, for example to re-determine the driving requirements on the rough trajectory. Thus, when driving through a construction site, a speed limit that was not previously known can be determined by means of the sensor device and reacted to in the present method.

[0050] Overall, the examples show how a method can be provided for implementing traffic rules into an automated driving function.

Claims

1. A method for operating an autonomously controlled vehicle (1), the method comprising: - providing a hierarchical data structure comprising combinations of driving requirements and control command codes; - planning a coarse trajectory (4) for a section of road to be driven in future; - determining driving requirements on the coarse trajectory; - dividing the trajectory (4) into a plurality of segments, wherein each transition between two segments is assigned a requirement change; - for each segment, reading a control command code from the hierarchical data structure according to the requirements applicable to the segment; - driving the trajectory (4) in segments and, in doing so, activating the control command code assigned to the segment.

2. The method of claim 1, wherein, While driving a segment, it is determined when the requirements change, in order to activate the control command code assigned to the next segment.

3. The method of claim 2, wherein, If the requirements change at a location that does not conform to the plan and / or under conditions that do not conform to the plan, the remaining trajectory is replanned by redividing the segments.

4. The method according to claim 2 or 3, characterized in that, When a new traffic sign is detected that introduces or cancels a restriction, the requirements change.

5. The method according to any one of claims 1 to 4, characterized in that, The control command codes specify minimum and / or maximum speed, minimum and / or maximum acceleration, minimum and / or maximum rotational speed, minimum and / or maximum gear, and / or minimum and / or maximum braking distance.

6. An autonomously controllable vehicle (1) comprising: A control unit that can be operated by control command codes, a set of actuators (14) that is operated by the control unit, a device (12) for driving planning, and a memory (16) in which a hierarchical data structure comprising combinations of driving requirements and control command codes is provided; wherein the device (12) for driving planning is designed to: - plan a coarse trajectory (4) for a section of road to be driven in future; - determine driving requirements on the coarse trajectory; - divide the trajectory (4) into a plurality of segments, wherein each transition between two segments is assigned a requirement change; - for each segment, read a control command code from the hierarchical data structure according to the requirements applicable to the segment; - implement driving the trajectory (4) in segments and, in doing so, provide the control command codes to the control unit (10).

7. An autonomously controllable motor vehicle according to claim 6, characterised in that, The device (12) for driving planning comprises a self-learning system, which is preferably in the form of a neural network, particularly preferably in the form of a recurrent neural network.

8. An autonomously controllable motor vehicle according to claim 6 or 7, characterised in that, In the hierarchical data structure: - in the highest level: respective control command codes are assigned for free driving; - in the next highest level: respective control command codes are assigned for traffic signs that cancel restrictions; - in the third level: respective control command codes are assigned for traffic signs that introduce restrictions; - in the fourth level: respective control command codes are assigned for categories of environmental conditions that necessarily lead to driving restrictions.

9. A system comprising: A data memory (16) having a data set containing a hierarchical data structure that defines combinations of driving requirements and associated control command codes for a control unit (10) of a vehicle (1); and a device (12) for autonomously controlling a vehicle, which has access to the data set.

10. The system of claim 9, wherein, In the hierarchical data structure: - in the highest level: the respective control command code is assigned for a free ride; - in the next highest level: the respective control command code is assigned for a traffic sign with cancellation of a restriction; - in the third level: the respective control command code is assigned for a traffic sign with introduction of a restriction; - in the fourth level: the respective control command code is assigned for a category of environmental conditions that inevitably leads to a driving restriction.

Images