Driver assistance system and method for operating a driver assistance system

Abstract

Driver assistance system (101) for a vehicle, comprising: - a detection device (103) for detecting a vehicle environment, - a position determining device (105) for determining a vehicle position relative to the vehicle surroundings, - a database (107, 303) with an ontological data structure in which traffic rules are implemented which the driver must follow in order to participate in road traffic in accordance with the law, and abstractly maps roads, carriageways, lanes, traffic lights, traffic signs, intersection topologies and links them to one another by relations using rules and / or axioms, - a linker (109) for linking the detected vehicle environment and the vehicle position with the ontological data structure to form a linked data structure so that a real traffic situation at an intersection can be assessed with regard to the traffic rules, - wherein a filter for filtering elements of the detected vehicle environment is formed between the detection device and the linker so that only the detected drivable areas are linked to the ontological data structure which can be directly reached by vehicles detected by the detection device (103) in the current situation, - an evaluator (109) for evaluating the linked data structure and - depending on the evaluated linked data structure, an action is carried out by autonomously braking or accelerating and / or steering the vehicle.

Description

[0001] The invention relates to a driver assistance system for a vehicle and a method for operating a driver assistance system of a vehicle. State of the art

[0002] Driver assistance systems for vehicles are well known. Typically, known driver assistance systems include a distance sensor for measuring the distance between the vehicle and other vehicles in or near the vehicle. Depending on the distance between the vehicle and the other vehicles, a warning signal is issued to the driver, for example. It may also be possible for the vehicle to brake autonomously to prevent a potential collision.

[0003] Especially at intersections with multiple intersecting lanes, traffic signs, traffic lights, and multiple vehicles turning in different directions, conventional driver assistance systems are generally not suitable. One reason for this is that conventional driver assistance systems can typically only detect a few vehicles at a time and lack knowledge of the traffic rules that a driver must follow in road traffic. Such an intersection situation is therefore too complex.

[0004] An ontology-based approach to vehicle control is known from the publication "Using Ontology-based Traffic Models for more efficient Decision Making of Autonomous Vehicles" by Ralf Regele, IEEE Fourth International Conference on Autonomie and Autonomous Systems" 2008.

[0005] An ontology-based approach for simulating driver assistance systems is also known from the publication “An Intelligent Driver Assistance System (I-DAS) for Vehicle Safety Modelling using Ontology Approach”, Saravanan Kannan et al., International Journal of UbiComp (IJU), Vol.1, No.3, July 2010.

[0006] DE 10 2005 020 429 A1 discloses a method and device for assisting drivers when crossing intersections. These methods essentially comprise a unit for modeling intersections, a sensor unit for detecting the environment, a computer unit for evaluating and determining a driving movement, and a warning device for alerting the driver of impending hazards. To determine a driving movement, the intersection is first divided into several zones, and a situation-dependent status value for the drivability is determined for each zone based on environmental information. Based on these status values, a driving movement suitable for crossing the intersection is determined before entering the intersection, taking into account a predetermined general direction of movement. The driver assistance system therefore enables the driver to cross the intersection reliably and safely. Disclosure of the invention

[0007] The object underlying the invention can therefore be seen as creating a driver assistance system that can support a driver even in complex traffic situations.

[0008] The object underlying the invention can also be seen in specifying a corresponding method for operating a driver assistance system of a vehicle.

[0009] The object underlying the invention can also be seen in specifying a corresponding storage device for a driver assistance system.

[0010] These objects are achieved by means of the respective subject matter of the independent claims. Advantageous embodiments are the subject matter of the respective dependent subclaims.

[0011] According to the invention, a driver assistance system for a vehicle is provided. The driver assistance system comprises a detection device for detecting a vehicle's surroundings and a position-determining device for determining a vehicle position relative to the vehicle's surroundings. Furthermore, a database is provided, which has an ontological data structure with implemented traffic rules. Furthermore, a linker is provided, which links the detected vehicle's surroundings and the vehicle position with the ontological data structure, so that a linked data structure is formed, in particular an ontologically linked data structure. In particular, the linker applies the implemented axioms and / or rules to the linked data structure. This means, in particular, that the linker is designed to draw a logical conclusion from the linked data structure.Preferably, the linker is further configured to store the linked data structure. The linked data structure is evaluated by an evaluator, for example, to determine a possible collision.

[0012] According to the invention, a method for operating a driver assistance system of a vehicle is further provided. In this method, a vehicle environment and a vehicle position relative to the vehicle environment are detected. Furthermore, the detected vehicle environment and the vehicle position are linked to an ontological data structure in which traffic regulations are implemented, thus forming a linked data structure, in particular an ontologically linked data structure. The linked data structure is subsequently evaluated.

[0013] Furthermore, a storage device is specified in which a database with an ontological data structure is stored, wherein traffic rules are implemented in the ontological data structure.

[0014] The invention therefore particularly encompasses the concept of linking an image of a real or current situation, i.e., the vehicle's surroundings and the vehicle's position relative to it, with an ontological data structure in which traffic rules are implemented. This advantageously creates a linked data structure that maps the real situation within the framework of its ontological data structure. Since traffic rules are also implemented in the ontological data structure, knowledge is advantageously available, for example, to better assess the current situation with regard to permissible and impermissible vehicle maneuvers. For example, the driver assistance system recognizes that the vehicle must yield to another vehicle.Furthermore, the driver assistance system advantageously recognizes which vehicle must give way to which vehicle, particularly in complex traffic situations, for example intersection situations with several vehicles.

[0015] These findings can then be communicated to the driver, for example, particularly visually, preferably by means of a display device comprising, for example, a warning light. In particular, the driver can also be informed acoustically whether or not he has the right of way. A display device can, in particular, comprise a screen or, in English, a display. A haptic notification to the driver can also be provided, for example.

[0016] Traffic regulations, as used herein, refer in particular to rules that drivers must follow in order to participate in road traffic in compliance with the law. In Germany, for example, traffic regulations include, in particular, the German Road Traffic Act (StVO).

[0017] An ontological data structure within the meaning of the present invention comprises, in particular, a data structure in which the data is classified or categorized and linked by relations to one another in advance or through the use of rules and / or axioms. In this respect, it is preferably a heavyweight ontology, i.e., an ontology with extensive use of axioms and / or rules. The ontological data structure abstractly maps, in particular, elements of road traffic such as roads, carriageways, lanes, traffic lights, traffic signs, and intersection topologies and links them, preferably through relations using rules and / or axioms. Axioms and / or rules can, for example, be right-of-way rules. Axioms are, in particular, statements that are always true, thus advantageously providing knowledge.Rules, on the other hand, are not always true, but are only executed at a specific point in time. The individual elements or data are preferably structured or classified hierarchically, thus forming a classification hierarchy. In particular, "traffic sign" can be a generic term for a "give-way" traffic sign and / or a "give-way" traffic sign.

[0018] According to the invention, a moving object is detected by the detection device and classified, for example, as a vehicle if certain rules and / or axioms are met. Such a rule or axiom can be as follows, for example: If the object is assigned to a lane, then it is a vehicle. Generally speaking, rules and / or axioms are additionally stored in the database as background knowledge, which are applied, in particular, by the linker to the detected data in order to form new relations and / or further classify the data. This classification can, for example, be a new classification or according to its hierarchy.

[0019] Linking in the sense of the present invention means in particular linking data corresponding to the detected vehicle environment and / or the relative vehicle position with data of the ontological data structure.

[0020] Detecting a vehicle's surroundings within the meaning of the present invention particularly comprises detecting a direction in which a detected additional vehicle is traveling or intends to travel. For example, a flashing light of the additional vehicle can be monitored, with flashing generally indicating the direction in which the additional vehicle intends to travel. Detecting can particularly also comprise detecting a single element or multiple elements or all elements of the vehicle's surroundings, which are then classified as corresponding objects such as a vehicle, lane, or traffic sign, and are further preferably provided with attributes such as position and / or speed.

[0021] According to one embodiment, the current traffic situation can be stored so that it can be used for later assessments or evaluations of current traffic situations. Potential future traffic situations can also be calculated and, in particular, stored.

[0022] A traffic situation comprises in particular a road topology such as an intersection topology, preferably one or more vehicles, for example signaling systems, in particular traffic signs.

[0023] According to the invention, the linker is configured to extract elements of road traffic and, for example, other vehicles, traffic signs, traffic lights and / or roadways with, for example, one or more lanes from the detected vehicle environment and to link them to the ontological data structure, in particular to the elements of the data structure.

[0024] According to the invention, a filter for filtering elements of the detected vehicle surroundings is formed between the detection device and the linker. This advantageously enables only certain elements of the current situation to be linked to the ontological data structure, which consumes less computing power. For example, while the detection device detects all drivable areas of the surroundings, the filter then filters out, for example, those drivable areas that cannot be directly reached by, for example, detected vehicles in the current situation. This means, in particular, that only the detected drivable areas that can be directly reached by detected vehicles in the current situation are linked to the ontological data structure.

[0025] According to one embodiment, the implemented traffic rules and / or the underlying axioms and / or rules are formulated in such a way that only those elements of the traffic situation that are assigned to detected vehicles are considered. In particular, priority rules can, for example, only consider lane / lane pairs to which vehicles are assigned, so that priority relations are only considered between vehicle pairs and not for all lane / lane pairs.

[0026] In a further embodiment, the detection device comprises a GPS sensor and / or a radar sensor and / or a camera, for example a video camera or a stereo video camera, and / or a lidar sensor and / or ultrasonic sensors and / or a so-called photonic mixer device (PMD) sensor, in particular an infrared sensor for measuring a distance, or so-called time of flight (TOF) sensors. Detection within the meaning of the invention particularly comprises sensory detection. Navigation data from a navigation system can preferably also be used to detect the vehicle's surroundings. In particular, sensory detection also means that further sensor data from external sources such as other vehicles and / or signaling systems are detected or used to define or determine the vehicle's surroundings.In particular, data related to the vehicle's surroundings is recorded, which can change over time. In particular, vehicle sensors such as a rain sensor and / or a temperature sensor can also be used to detect the vehicle's surroundings. Preferably, an angle between two lanes is also recorded.

[0027] According to another embodiment, it can be provided that an action is carried out depending on the evaluated linked data structure. In particular, it can be provided that the vehicle is braked or accelerated and / or steered autonomously. Preferably, a visual and / or acoustic and / or haptic warning signal is issued. In particular, a belt tensioning system can be activated. For example, an airbag system can also be activated. Preferably, further driver assistance systems are therefore activated so that they can then carry out collision severity-reducing measures in the event of a collision, for example. A driver assistance system within the meaning of the present invention can, for example, be an active or a passive system. In particular, it is a comfort system or a safety system.

[0028] In a further embodiment, the detected vehicle surroundings and / or the relative vehicle position are checked for consistency or validity with respect to the ontological data structure. Preferably, a checking device is provided for this purpose, which can be integrated, for example, in the linker. This advantageously makes it possible to detect an error in the detection or position determination. For example, according to the detection device, a vehicle should be in a lane that can only be driven in the opposite direction. However, this linking of a vehicle with a lane that can be driven in the opposite direction is defined as not possible or as incorrect according to the ontological data structure. Detection or position determination can thus, for example, be repeated, or an error message can be generated by the sensors, or a corresponding action can be carried out by the driver assistance systems.

[0029] In another embodiment, the evaluator is integrated into the combiner. Preferably, a processor can be provided in which the evaluator and / or the combiner and / or the filter are integrated. Preferably, a processor can be provided in which the evaluator and / or the combiner and / or the filter are integrated in software. The processor is thus configured to calculate and execute the corresponding combinations, evaluations, and filter actions.

[0030] The invention will be explained in more detail below using preferred embodiments with reference to figures. Fig. 1 a driver assistance system, Fig. 2 a flowchart of a method for operating a driver assistance system, Fig:3 a storage device Fig. 4 an intersection situation and Fig. 5 another intersection situation.

[0031] In the following, the same reference symbols are used for the same features.

[0032] Fig. 1 shows a driver assistance system 101 for a vehicle (not shown). The driver assistance system 101 comprises a detection device 103 for detecting the vehicle's surroundings and a position determination device 105 for determining a vehicle position relative to the vehicle's surroundings. Preferably, the vehicle speed is also determined. Furthermore, a database 107 is formed, which has an ontological data structure with implemented traffic rules.

[0033] The driver assistance system 101 further comprises a linker 109 configured to link data corresponding to the detected vehicle surroundings and the relative vehicle position with the ontological data structure to obtain a linked data structure. This linked data structure is evaluated by an evaluator 111.

[0034] The driver assistance system 101 is thus advantageously capable of detecting a current or real vehicle situation, in particular by sensory means, and mapping the data corresponding to this vehicle situation to an ontological data structure. Since the ontological data structure implements traffic rules, it is also advantageously possible for the driver assistance system 101 to provide a driver, for example, with instructions or suggestions for a driving maneuver, even in complex intersection situations with multiple traffic signs and / or traffic lights.

[0035] In an embodiment not shown, it may be provided that the position-determining device 105 and the detection device 103 are formed integrally as a single sensor. This means, in particular, that such a sensor has both functionalities.

[0036] For example, it may also be provided that user inputs are recorded by means of an input device not shown, wherein the user inputs may include, for example, requests to the driver assistance system 101 as to which vehicle has right of way over which.

[0037] Furthermore, it is preferably provided that the evaluator also submits queries to the database via, for example, the linker.

[0038] In an exemplary embodiment not shown, it can preferably be provided that a filter is formed between the detection device 103 and the linker 109, which filter filters certain elements from the data according to the detected vehicle environment. Only the filtered data are then linked to the ontological data structure. Filtering has the particular advantage that only data relevant to a specific vehicle situation is used for the linking. For example, in traffic regulations, elements of a vehicle environment such as roadways and / or lanes that have no further connection to vehicles are not relevant for an assessment of a possible collision. Thus, less data needs to be processed, which significantly reduces computational effort and enables quasi-real-time processing.

[0039] Fig. 2 shows a flowchart of a method for operating a driver assistance system of a vehicle. In a step 201, the vehicle's surroundings are detected, in particular by sensors. In a step 203, a vehicle position relative to the detected vehicle's surroundings is detected or determined. The corresponding data of the detected vehicle's surroundings and the relative vehicle position are then linked in a step 205 with an ontological data structure comprising implemented traffic rules, so that a linked data structure is formed. In a step 207, this linked data structure is evaluated, for example, to provide a driver with instructions or suggestions for a specific driving maneuver, particularly in an intersection situation.

[0040] Fig. 3 shows a storage device 301 in which a database 303 is stored. The database 303 comprises an ontological data structure with implemented axioms and / or rules, in particular traffic rules. These axioms and / or rules (in particular traffic rules) stored or implemented in the database are furthermore executed, in particular, by the linker 109, with the results being stored, in particular.

[0041] Fig. Figure 4 shows an intersection 401 with five intersecting lanes, 403, 405, 407, 409, and 411. In the intersection area, traffic signs associated with lanes 403, 405, 407, 409, and 411 are symbolically indicated. Reference sign 413 denotes a priority road traffic sign. This means, in particular, that traffic sign 413 designates the corresponding lane as a priority road. Reference sign 415 denotes a give-way traffic sign. This means, in particular, that traffic sign 415 indicates to a driver that they must yield the right of way to vehicles approaching from other lanes with priority, or to vehicles approaching from the right from lanes with equal priority. Furthermore, traffic sign 417 is also indicated, indicating a speed limit.

[0042] Vehicles on lanes 403, 405, 407, 409, and 411 are marked with a circle bearing the reference number 419. A number from 1 to 6 is drawn in the circle 419 to better distinguish each individual vehicle. The arrows with the reference number 421 indicate an example of the route taken by the corresponding vehicles 419. The arrows with the reference number 423 indicate the lanes located on lanes 403, 405, 407, 409, and 411 and the permitted route the vehicles are permitted to take.

[0043] Although not explicitly shown here, intersection 401 may have traffic lights in addition to or instead of traffic signs 413, 415, and 417. The intersection topology and traffic situation shown here, particularly the number and arrangement of vehicles, are intended only as examples. For example, intersection topologies with more or fewer than five lanes may also be provided.

[0044] The driver assistance system 101 includes the Fig. 4 and, due to the link between the ontological data structure and the implemented traffic rules, is able to make statements about, for example, which vehicle must yield to which vehicle. Depending on this, a warning or accident-avoiding action can then be executed, for example, the vehicle can be braked autonomously if the driver ignores a corresponding warning signal.

[0045] Fig. 5 shows another intersection 501 with four intersecting lanes 503, 505, 507, and 509. On lane 505, there is a vehicle 511 moving away from intersection 501. On lane 507, there is another vehicle 513 approaching intersection 501. On lane 509, there is another vehicle 515, also approaching intersection 501.

[0046] A linking of these individual classified elements through relations among each other is possible in Fig. 5 are symbolically represented by arrows or double arrows with corresponding reference symbols, which are explained below. These links are represented in particular in the ontological data structure.

[0047] The link by a relation with the reference number 517 links the two lanes 503 and 505 and includes the information that lane 503 is located to the right of lane 505, viewed in the direction of intersection 501. Similarly, the link 519 links the two lanes 503 and 505 and includes the information that lane 505 is located to the left of lane 503, viewed in the direction of intersection 501. The relations with the reference numbers 517 and 519 exist in pairs between all adjacent lanes 503, 505, 507, and 509.

[0048] Furthermore, in particular, traffic sign 415 is linked to lane 509 or intersection 501, which is indicated by an arrow with reference number 525. Link 525 thus includes, in particular, the information that traffic sign 415 belongs to lane 509 and is relevant to intersection 501.

[0049] Furthermore, a link 527 is provided between the vehicle 513 and the roadway 507, which in particular includes the information that the vehicle 513 is located on the roadway 507. A further link 529 between the vehicle 513 and the intersection 501 includes the information that the vehicle 513 is approaching the intersection 501.

[0050] The link with the reference number 533 indicates that the corresponding vehicle 511 is moving away from the intersection 501.

[0051] Yet another link 531 between the roadway 503 and the intersection 501 includes in particular the information that the roadway 503 is connected to the intersection 503.

[0052] The traffic signs 413 and 415 are ontologically structured in such a way that a generic term “traffic sign” is formed in the data structure, whereby a hierarchy is then provided into a sub-term “give way traffic sign” and a further sub-term “priority road traffic sign”, in which the traffic signs 413 and 415 are classified or assigned accordingly.

[0053] The link or relation 521 between vehicles 513 and 515 includes the information that vehicle 513 has the right of way over vehicle 515. The link 523 between vehicles 515 and 513 includes the information that vehicle 515 must yield the right of way to vehicle 513. This information is generated in particular by the linker and the ontological data structure and preferably stored or saved in the ontological data structure.

[0054] The examples and traffic situations, as well as intersection topologies explained above, are intended to be understood as examples only. For example, there may be more or fewer lanes than those shown. In particular, the number of vehicles may be entirely different. The invention is therefore applicable in all variants of any complexity.

[0055] In summary, the invention particularly encompasses the idea of ​​providing information, in particular traffic regulations, to a driver assistance system by means of a linker, so that a real traffic situation, for example at an intersection, can be assessed with regard to the traffic regulations, but in particular also with regard to who is driving where, what the traffic signal phase is, i.e., in particular, current, situational information, for example to provide instructions or suggestions for legally compliant driving maneuvers or to steer and / or brake the vehicle autonomously in the event of danger. Because traffic regulations, in particular, are implemented in an ontological data structure, the recorded traffic situation can be processed by the ontological data structure, taking the traffic regulations into account, without the need for additional external algorithms.In this case, an ontology is in particular a formalization or serves to describe a knowledge base.

Claims

[1] Driver assistance system (101) for a vehicle, comprising: - a detection device (103) for detecting a vehicle environment, - a position determining device (105) for determining a vehicle position relative to the vehicle surroundings, - a database (107, 303) with an ontological data structure in which traffic rules are implemented which the driver must follow in order to participate in road traffic in accordance with the law, and abstractly maps roads, carriageways, lanes, traffic lights, traffic signs, intersection topologies and links them to one another by relations using rules and / or axioms, - a linker (109) for linking the detected vehicle environment and the vehicle position with the ontological data structure to form a linked data structure so that a real traffic situation at an intersection can be assessed with regard to the traffic rules, - wherein a filter for filtering elements of the detected vehicle environment is formed between the detection device and the linker so that only the detected drivable areas are linked to the ontological data structure which can be directly reached by vehicles detected by the detection device (103) in the current situation, - an evaluator (109) for evaluating the linked data structure and - depending on the evaluated linked data structure, an action is carried out by autonomously braking or accelerating and / or steering the vehicle. [2] Driver assistance system (101) according to claim 1, wherein a checking device is formed for checking the detected vehicle surroundings and / or the relative vehicle position for consistency with the ontological data structure. [3] Driver assistance system (101) according to one of the preceding claims, wherein the detection device (103) comprises a GPS sensor, a radar sensor, a stereo video camera, a lidar sensor, an ultrasonic sensor, a PMD sensor and / or a camera. [4] Method for operating a driver assistance system (101) of a vehicle, comprising the following steps: - Detection of a vehicle environment (201), - detecting a vehicle position relative to the vehicle environment (203) and - Linking the recorded vehicle environment and the vehicle position with an ontological data structure in which traffic rules are implemented which the driver must follow in order to participate in road traffic in accordance with the law, and abstractly mapping roads, carriageways, lanes, traffic lights, traffic signs, intersection topologies and linking them together through relations by means of rules and / or axioms in order to form a linked data structure (205) so that a real traffic situation at an intersection can be assessed with regard to the traffic rules, wherein elements from the detected vehicle environment are filtered and the filtered detected vehicle environment is linked to the ontological data structure, so that only the detected drivable areas are linked to the ontological data structure that can be directly reached by vehicles detected by the detection device (103) in the current situation, - Evaluating the linked data structure (207), - Performing an action depending on the evaluated linked data structure by autonomously braking or accelerating and / or steering the vehicle. [5] Method according to claim 4, wherein the detected vehicle surroundings and / or the relative vehicle position are checked for consistency with respect to the ontological data structure. [6] Storage device (301) for a driver assistance system (101), comprising a database (107, 303) comprising an ontological data structure in which traffic rules are implemented.

Images