Computer-implemented method for intelligent traffic guidance of at least one autonomous vehicle, computer program, storage medium, data carrier signal, system and autonomous vehicle

Abstract

Computer-implemented method for intelligent traffic guidance of at least one autonomous vehicle (6), the method comprising the steps: • Obtaining input data, wherein the input data comprises sensor data, wherein the sensor data comprises traffic scenarios and information on the position, speed and planned route of the autonomous vehicle (6) (V1); • Determining at least one traffic action based on the input data by means of at least one computing unit (7), wherein the computing unit (7) is connected to the autonomous vehicle (6) via a communication interface (9), wherein the at least one traffic action has a direct or indirect influence on other road users (V2); • Initiating at least one of the traffic actions by the computing unit (7) and / or the autonomous vehicle (6), wherein by initiating at least one traffic action a separate pseudo lane is generated for the autonomous vehicle (6) (V3).

Description

The invention relates to a computer-implemented method for intelligent traffic guidance of at least one autonomous vehicle. The invention further relates to a computer program, a storage medium, a data carrier signal, a system and an autonomous vehicle.The following definitions apply to the entire disclosure content.Nowadays, in urban areas, traffic often suffers from very full or even congested roads, especially in times of the rubhour. If there are no separate lanes for the public transit, a public transit vehicle, such as a bus or autonomous vehicle, is concerned with utilizing congested roads, which results in that schedules cannot be followed and thus reduces acceptance and attractiveness of public transit usage. Particularly autonomous vehicles of public short-range traffic are often disadvantageous under such conditions, since these frequently have a deflexive driving behavior and have a greater space requirement, for example for lane changes. To counteract this, one possibility is to create separate lanes for such autonomous vehicles. However, this is often not possible, for example due to space restrictions.Prior art is disclosed in "Transfer Learning versus Multiagent Learning Varying Distributed Decision-Making in Highway Traffic" by Mark Schutera et al., "Distributed traffic light control at uncoupled intersections with real-world topology by deep reinforcement learning" by Mark Schutera et al., "Intelligent Traffic Control System" by Jubair Mohammed Bilal et al., and in US9825483 B2.It was an object of the invention to provide a method and system for intelligent traffic guidance of at least one autonomous vehicle.The subject matters of the independent claim and the subordinate claims each solve this task. Further developments and advantageous embodiments are evident from the dependent claims, the drawings and the description of preferred exemplary embodiments.According to one aspect, the invention provides a computer-implemented method for intelligent traffic guidance of at least one autonomous vehicle, the method comprising the steps of:• obtaining input data, wherein the input data comprises sensor data, wherein the sensor data comprises traffic scenarios and information on the position, speed and planned route of the autonomous vehicle;• Determination of at least one traffic action on the basis of the input data by means of at least one arithmetic unit, wherein the arithmetic unit is connected to the autonomous vehicle via a communication interface, wherein the at least one traffic action has a direct or indirect influence on other road users;• Initiation of at least one of the traffic actions by the computing unit and / or the autonomous vehicle, wherein an own pseudo lane for the autonomous vehicle is generated by the initiation of at least one traffic action.The entire method may be used for one or more autonomous vehicles. In the case that it is used for a plurality of autonomous vehicles, the autonomous vehicle is in each case one of a plurality of individual vehicles.Computer-implemented means that the steps of the method are carried out by a data processing device, for example a computer, a computing unit, a computer network, for example a cloud system, hardware of a control unit, or parts thereof.The computing unit can be designed, for example, as an artificial neural network with a deep Q-learning (DQL) architecture, and an reinforcement learning method (reinforcement learning) can be applied. The artificial neural network can receive various input data and optimize the outputs so that the traffic guidance of an autonomous vehicle is optimized, for example, so that it can move as quickly and safely as possible. The obtaining of input data can take place in real time, so that corresponding traffic actions can be initiated in real time. The traffic guidance can thus be adapted in real time, since the outputs (actions) are derived for each update of the input data (state). However, it is also possible to choose a heuristic approach and thus a rule-based embodiment.Traffic scenarios may include information about other road users, such as number, position, or speed. This data can be acquired as input data, for example, via cameras or other sensors, which can be part of the infrastructure or of the road users. Similarly, traffic scenarios may include information about the infrastructure itself, such as the status of a traffic light circuit or a traffic sign circuit. The information on the position, speed and the planned route of the autonomous vehicle can be detected, for example, by vehicle sensors, such as speed sensors, GPS receivers or other sensors of the autonomous vehicle, and / or provided via map information. The input data can be transmitted to the computing unit, for example, via a radio connection, for example, via V2X. The at least one computing unit can be arranged in particular centrally, for example in a control center or a cloud, in order to create intelligent traffic guidance in particular for a plurality of autonomous vehicles. However, it is also possible for the computing unit to be located in the autonomous vehicle itself. It is likewise possible for there to be a plurality of computing units or for one computing unit to be divided into a plurality of subunits which can be arranged, for example, at least partially at different positions along a route in the infrastructure. The computing unit is connected to the autonomous vehicle via a communication interface in order to be able to receive and process the input data comprising information of the autonomous vehicle.A traffic action can include, for example, switching a variable sign and / or a traffic light, in particular a traffic light, in order to indirectly influence other traffic participants, for example by reducing the speed of a variable sign for other traffic participants or by other traffic participants having to stop and wait at a red-switched traffic light.In the case of networked vehicles, for example, a traffic action can also have a direct influence on other road users by direct commands being sent to other road users, for example in the form of signals. In the case of other autonomous road users, a command can, for example, have a direct influence on the vehicle control. A traffic action with indirect influence on other road users can also comprise, for example, a visual or acoustic instruction to at least one vehicle driver. The instruction can comprise, for example, a request for reducing the travel speed and can be transmitted, for example, via a radio connection to other road users and / or via an infrastructure device, such as, for example, a variable message sign, loudspeaker, screen or others.Other road users are influenced by the at least one traffic action in such a way that the at least one autonomous vehicle is allowed to progress as freely as possible.The method can therefore be understood as a type of intelligent road scraper with respect to other road users, which adapt their driving behavior in such a way that a lane of their own is generated for the autonomous vehicle. The driving lane is a pseudo-driving lane, since this is not implemented structurally, but is generated merely by the adapted driving behavior of other road users.The initiation of at least one traffic action can be effected by the computing unit, for example by sending command signals to variable traffic signs and / or traffic lights and / or other road users, for example. Variable traffic signs can indicate, for example, a speed restriction or else indicate a travel prohibition for a driving lane, in order, for example, to permit the autonomous vehicle to use the only right in the driving lane. The initiation of at least one traffic action can, however, also be effected by the autonomous vehicle itself, in particular if the computing unit is an integral component of the autonomous vehicle. The autonomous vehicle can then initiate one or more traffic actions, for example, via a communication interface.An intelligent system for optimizing the traffic flow is advantageous, in particular for public transportation means such as autonomous buses. With V2X communication implemented, free passage can be allowed for the autonomous bus. For example, by reducing the density of other road users, the probability of critical traffic scenarios caused by other road users, for example, by sudden and / or undisplayed lane changes, etc., could be minimized. These scenarios increase the probability of an accident and the speed of travel of the bus can also be impaired as a result, since an autonomous bus is tied to the traffic laws and must ensure the safety of the passengers at any time, even if the surrounding road users violate the law and are unexpected. This can result in an autonomous bus being braked out under right rules.The method described above and the initiated traffic actions can improve the convenience of autonomous vehicles, in particular autonomous buses, in that traffic flows are optimized and a pseudo-separate lane for the autonomous vehicle is generated. In addition, this increases the safety of autonomous vehicles, for example by reducing the complexity of the traffic, for example by slowing down other road users or reducing the traffic density. The method described generates a pseudo lane for an autonomous vehicle without realizing it structurally.The method can also be used, for example, for correspondingly equipped emergency vehicles.According to a further aspect, the invention provides a computer program for intelligent traffic guidance of at least one autonomous vehicle, the computer program comprising program instructions which cause a computer to execute the steps of a method for intelligent traffic guidance of at least one autonomous vehicle when the computer program is loaded onto or executed by the computer.The instructions of the computer program according to the invention comprise machine instructions, source text or object code written in assembly language, an object-oriented programming language, for example C++, or in a procedural programming language, for example C. The computer program is executed wholly or partly on an integrated circuit or in a remote system comprising a cloud. The integrated circuit can be, for example, a central processing unit (CPU) or a graphics processor (GPU).According to a further aspect, the invention provides a storage medium on which the computer program for intelligent traffic guidance of at least one autonomous vehicle is stored.The storage medium can be, for example, a nonvolatile, permanent memory. A storage medium may be, for example, a USB stick, a hard disk, a memory card, or a CD-ROM. However, the storage medium can also be, for example, a cloud in which the computer program is stored.According to a further aspect, the invention provides a data carrier signal which transmits the computer program for intelligent traffic guidance of at least one autonomous vehicle.The computer program for intelligent traffic guidance of at least one autonomous vehicle can be transmitted via a data carrier signal, for example, from a cloud to a computing unit or to a control device of an autonomous vehicle, for example, via a so-called over-the-air update.According to a further aspect, the invention provides a system for intelligent traffic guidance of at least one autonomous vehicle, the system comprising at least one computing unit for carrying out a computer-implemented method for intelligent traffic guidance of at least one autonomous vehicle, an autonomous vehicle and at least one sensor for generating input data.According to a further aspect, the invention provides an autonomous vehicle, comprising at least one communication interface, wherein the traffic guidance of the autonomous vehicle takes place according to a method for intelligent traffic guidance of at least one autonomous vehicle.The communication interface is required in order to be able to communicate with the computing unit. The communication can be carried out on both sides in order to send input data to the computing unit and also to receive signals from the computing unit. The communication interface can serve to initiate traffic actions, such as, for example, to send commands in the form of signals to infrastructure devices or to other road users. The communication interface can be a radio interface.According to one aspect, the input data comprises map data and / or camera data and / or induction loop data and / or a status of at least one variable message sign and / or a status of at least one traffic light system.Map data may include, for example, information about a road at a particular location in the map, such as number of lanes, maximum speed allowed, information about infrastructure facilities such as road signs and / or traffic lights, or other information. Camera data can likewise comprise, for example, information about the number of lanes, permissible maximum speed and number and / or position of other road users.Induction loop data may contain information on the number of other road users present.The status of a variable message sign may include, for example, information prevailing at the maximum permissible speed. Depending on the variable message sign, the status can also contain, for example, information for permission to drive in individual lanes.The status of a traffic light can contain information on the current circuit state as well as information on the circuit duration and / or on the circuit cycle.According to a further aspect, the at least one traffic action with direct influence on other road users comprises at least one travel command. In the case of other autonomous road users, the travel command can have a direct influence on the longitudinal and / or lateral control of the other road user.According to a further aspect, the at least one traffic action with indirect influence on other road users comprises changing the status of at least one variable sign and / or at least one traffic light system. By changing the status of a variable sign and / or a traffic light system, other traffic participants can be caused to adapt their driving behavior accordingly, for example by reducing the driving speed and / or changing lanes and / or stopping at a traffic light system or others.The invention is explained by way of example with reference to the following exemplary embodiments. The following are shown: FIG. 1 shows a flow diagram of a method for intelligent traffic guidance of at least one autonomous vehicle, FIG. 2 is a schematic illustration of a computer program, FIG. 3 shows a schematic representation of a storage medium, FIG. 4 shows a schematic illustration of a data carrier signal, FIG. 5 shows a schematic illustration of an exemplary embodiment of a system, and FIG. 6 shows a schematic illustration of an exemplary embodiment of an autonomous vehicle.In the figures, like reference numerals designate like or functionally like items. In the figures, only the relevant objects are identified with reference numerals in each case in order to avoid repetitions.The flow diagram of a method for intelligent traffic guidance of at least one autonomous vehicle 6 shown in FIG. 1 comprises a step V 1 of obtaining input data, wherein the input data comprises sensor data, wherein the sensor data comprises traffic scenarios and information about the position, speed and the planned route of the autonomous vehicle 6. The method further comprises a step V 2 of ascertaining at least one traffic action based on the input data by means of at least one arithmetic unit 7, wherein the arithmetic unit 7 is connected to the autonomous vehicle 6 via a communication interface 9, wherein the at least one traffic action has direct or indirect influence on other road users. The method also comprises a step V 3 of initiating at least one of the traffic actions by the computing unit 7 and / or the autonomous vehicle 6, wherein an own pseudo lane for the autonomous vehicle is generated by the initiation of at least one traffic action.The schematic representation of a computer program 1 shown in FIG. 2 shows, by way of example, a computer program 1 for intelligent traffic guidance of at least one autonomous vehicle 6, wherein the computer program 1 according to FIG. 2 is executed by a computer 2.The schematic representation of a storage medium 3 shown in FIG. 3 shows, by way of example, a USB stick as storage medium 3, onto which the computer program 1 has been loaded and which can transmit the computer program 1 to a computer 2.The schematic representation of a data carrier signal 4 shown in FIG. 4 shows, by way of example, a data carrier signal 4 which transmits a computer program 1. According to FIG. 4, the computer program 1 is transmitted, for example, from a cloud by means of the data carrier signal 4.The schematic illustration of an exemplary embodiment of a system 5 shown in FIG. 5 shows, by way of example, a bus as autonomous vehicle 6. according to FIG. 5, the computing unit 7 is arranged, by way of example, within the autonomous vehicle 6. As sensor 8, FIG. 5 shows, by way of example, an induction loop which can detect both the autonomous vehicle 6 and other road users.The schematic illustration of an exemplary embodiment of an autonomous vehicle 6 shown in FIG. 6 shows, by way of example, a bus as autonomous vehicle 6. according to FIG. 6, autonomous vehicle 6 comprises, by way of example, a communication unit 9 which can communicate with computing unit 7 via a radio connection. The computing unit 7 is situated in a cloud, for example, according to FIG. 6. According to FIG. 6, the autonomous vehicle 6 comprises a sensor 8, which is attached, for example, in the rear region of the autonomous vehicle 6. This can be, for example, a camera that can record traffic scenarios of the rearward traffic. The autonomous vehicle 6 may include numerous other sensors 8, for example for localization, speed detection and others.Reference numerals denote reference numeralsV 1-V 3 Method steps 1 Computer program 2 Computer 3 Storage medium 4 Data carrier signal 5 System 6 Autonomous vehicle 7 Computing unit 8 Sensor 9 Communication interfaceReferences included in the specificationThis list of documents cited by the applicant has been produced in an automated manner and is only included for the better information of the reader. The list is not part of the German patent application or utility model application. The DPMA does not take any adhesion for any faults or omissions.Patent Literature citedUS 9825483 B2

[0004]

Claims

Computer-implemented method for intelligent traffic guidance of at least one autonomous vehicle (6), the method comprising the steps of: • obtaining input data, wherein the input data comprise sensor data, wherein the sensor data comprise traffic scenarios and information about the position, speed and the planned route of the autonomous vehicle (6) (V1); • determining at least one traffic action based on the input data by means of at least one computing unit (7), wherein the computing unit (7) is connected to the autonomous vehicle (6) via a communication interface (9), wherein the at least one traffic action has direct or indirect influence on other road users (V2); • Initiation of at least one of the traffic actions by the computing unit (7) and / or the autonomous vehicle (6), wherein an own pseudo lane for the autonomous vehicle (6) is generated (V3) by the initiation of at least one traffic action.Computer-implemented method for intelligent traffic guidance of at least one autonomous vehicle (6) according to Claim 1, wherein the input data comprise map data and / or camera data and / or induction loop data and / or a status of at least one variable sign and / or a status of at least one traffic light system.Computer-implemented method for intelligent traffic guidance of at least one autonomous vehicle (6) according to one of the preceding claims, wherein the at least one traffic action with direct influence on other road users comprises at least one driving command.Computer-implemented method for intelligent traffic guidance of at least one autonomous vehicle (6) according to one of the preceding claims, wherein the at least one traffic action with indirect influence on other road users comprises changing the status of at least one variable traffic sign and / or at least one traffic light system.A computer program (1) for intelligent traffic guidance of at least one autonomous vehicle (6), the computer program (1) comprising program instructions which cause a computer (2) to execute the steps of a method according to claim 1 when the computer program (1) is loaded or executed on the computer (2).Storage medium (3) on which the computer program (1) according to claim 5 is stored.A data carrier signal (4) carrying the computer program (1) according to claim 5.System (5) for intelligent traffic guidance of at least one autonomous vehicle (6), the system (5) comprising at least one arithmetic unit (7) for carrying out a computer-implemented method according to Claim 1, an autonomous vehicle (6) and at least one sensor (8) for generating input data.Autonomous vehicle (6) comprising at least one communication interface (9), wherein the traffic guidance of the autonomous vehicle (6) takes place according to a method according to Claim 1.

Images