Method, apparatus, computer program and storage medium for improving security at traversing facility

Abstract

A method for improving safety at a crossing facility comprises the steps of: receiving an environment signal representative of an environment comprising the crossing facility of a motor vehicle traveling towards and desiring to cross the crossing facility; specifying a boundary condition and / or an operating command for a cross-facility maneuver to be planned for the motor vehicle on the basis of the environment signal; planning a cross-facility maneuver for the motor vehicle on the basis of the environment signal, the planning comprising ascertaining a target trajectory for the motor vehicle that satisfies the specification on the basis of the specification, and if a target trajectory for the motor vehicle that satisfies the specification cannot be ascertained, the planning comprising determining a target trajectory for the motor vehicle that satisfies the specification, using the continued relaxed boundary conditions and / or the relaxed operating instructions for a long time in order to determine a spare target trajectory that satisfies the relaxed boundary conditions and / or operating instructions; and outputting the corresponding solved target trajectory. The invention relates to a device, a computer program and a machine-readable storage medium.

Description

Technical Field

[0001] The invention relates to a method, a device, a computer program and a machine-readable storage medium for increasing safety at a crossing facility, in particular a pedestrian crossing. Background Art

[0002] Publication US2020 / 0283014 A1 discloses a method for controlling an autonomous vehicle.

[0003] Publication US 2022 / 0185299 A1 discloses a system for operating an autonomous vehicle.

[0004] Publication US 2022 / 0236737 A1 discloses a system for determining the trajectory of an autonomous vehicle.

[0005] Patent document US Pat. No. 11,392,120 B2 discloses a method for determining a trajectory of a vehicle.

[0006] At crossing facilities, such as pedestrian crossings, the situation for an automated vehicle can change unexpectedly. For example, a pedestrian can unexpectedly walk out of a building or a vehicle, or occlusions and perception errors can lead to pedestrians being recognized too late. As a result, it can happen that the automated vehicle enters a situation from which it can no longer find a collision-free trajectory while meeting all boundary conditions. This can happen, for example, when a vehicle is approaching a crossing facility and a pedestrian suddenly enters the crossing facility so urgently that it is no longer possible to stop in front of the crossing facility in compliance with the regulations by means of the necessary braking path. Summary of the invention

[0007] The object on which the present invention is based can be seen as providing a concept for increasing safety at crossing facilities.

[0008] This object is achieved by means of a method, a device, a computer program and a machine-readable storage medium according to the invention for increasing safety at crossing facilities, in particular pedestrian crossings. Advantageous embodiments of the invention are the subject matter of the respective preferred embodiments.

[0009] According to a first aspect, a method for improving safety at a crossing facility, in particular a pedestrian crossing, is provided, the method comprising the following steps:

[0010] receiving an environment signal representing an environment including the crossing facility of a motor vehicle traveling toward the crossing facility and intending to cross the crossing facility,

[0011] Based on the environmental signal, a crossing facility maneuver to be planned for the motor vehicle is performed. to predetermine at least one boundary condition and / or at least one operating instruction,

[0012] planning a crossing facility maneuver for the motor vehicle based on the environmental signal,

[0013] The planning includes determining a target trajectory for the motor vehicle that satisfies the at least one predetermined boundary condition and / or the at least one operating command based on the at least one predetermined boundary condition and / or the at least one operating command,

[0014] If no target trajectory for the motor vehicle can be determined that satisfies at least one predetermined boundary condition and / or at least one operating command, the planning includes using at least one relaxed boundary condition and / or at least one relaxed operating command in order to determine an alternative target trajectory that satisfies the relaxed at least one boundary condition and / or at least one operating command,

[0015] Output the corresponding target trajectory.

[0016] According to a second aspect, an apparatus is provided, which is configured to carry out all the steps of the method according to the first aspect.

[0017] According to a third aspect, there is provided a computer program comprising instructions which, when executed by a computer, for example by an apparatus according to the second aspect, cause the computer to carry out the method according to the first aspect.

[0018] According to a fourth aspect, there is provided a machine-readable storage medium having stored thereon a computer program according to the third aspect.

[0019] The present invention is based on and includes the following recognition, that is, the above-mentioned task is solved in the following manner: for the planning of the maneuvering action of crossing the facility, at least one boundary condition and / or at least one operating instruction are pre-given, wherein, if the target trajectory for the motor vehicle that satisfies the at least one boundary condition and / or at least one operating instruction cannot be obtained, at least one relaxed boundary condition and / or at least one relaxed operating instruction is used to obtain a spare target trajectory that satisfies the at least one relaxed boundary condition and / or at least one relaxed operating instruction. This therefore means that if the original boundary condition or the original operating instruction cannot be met or complied with, at least one spare level (Rückfallebene) is used to obtain the target trajectory for the motor vehicle. Therefore, when there is no longer a solution for these original boundary conditions or operating instructions under the original boundary conditions or operating instructions, the target trajectory can still be obtained by setting at least one spare level. Therefore, in particular, the rigid requirements for the target trajectory are gradually relaxed so that the violation of the boundary conditions or operating instructions is as little as possible.

[0020] In particular, the following technical advantage is achieved in that a concept is provided which allows a motor vehicle to automatically withdraw from a critical situation occurring at a crossing by means of a targeted softening or relaxation of specified boundary conditions or operating instructions.

[0021] In particular, the following technical advantage is thereby further achieved: the safety at the crossing facility can be effectively increased.

[0022] If a boundary condition is used in the plural, it should always be followed by the singular and vice versa. If an operating instruction is used in the plural, it should always be followed by the singular and vice versa.

[0023] The expression "at least one" means "one or more".

[0024] In the sense of this specification, a crossing facility is, for example, one of the following crossing facilities: a pedestrian crossing, also known as a zebra crossing, a pedestrian lane Traffic island.

[0025] In one embodiment of the method, it is provided that, if a backup target trajectory for the motor vehicle that satisfies at least one relaxed boundary condition and / or at least one operating instruction cannot be determined, the planning includes using at least one further relaxed boundary condition and / or at least one further relaxed operating instruction for as long as necessary until a backup target trajectory that satisfies at least one corresponding further relaxed boundary condition and / or at least one operating instruction is determined, or until a backup target trajectory that satisfies at least one corresponding further relaxed boundary condition and / or at least one operating instruction is determined, at least within a predetermined range.

[0026] This achieves the following technical advantage, for example: the target trajectory can still be found even when no solution exists while observing the boundary conditions or operating instructions. The rigid requirements for the target trajectory are thus gradually relaxed so that the boundary conditions or operating instructions are violated as little as possible. At the same time, the downgrade or relaxation is only performed when no solution can be found for the strict boundary conditions.

[0027] In one specific embodiment of the method, it is provided that a motor vehicle state signal describing a state of the motor vehicle is received, wherein the predefinition and the planning are carried out based on the motor vehicle state signal.

[0028] This achieves, for example, the technical advantage that the specification and the planning can be carried out efficiently.

[0029] The motor vehicle state is described, for example, by the attitude and / or speed and / or acceleration of the motor vehicle.

[0030] In one specific embodiment of the method, it is provided that the specification is carried out on the basis of a predetermined rule set which comprises one or more rules for the behavior of a motor vehicle at a crossing.

[0031] This achieves, for example, the technical advantage that the specification can be carried out efficiently.

[0032] A rule is, for example, that if a crossing is free, the motor vehicle should cross it.

[0033] Rules are, for example, that motor vehicles should stop in time before crossing facilities used by pedestrians, generally living things, especially animals. Before crossing facilities means that the virtual stop line or virtual stop point before which the motor vehicle should stop is located before the crossing facilities in the direction of travel of the motor vehicle.

[0034] For example, the rule is: if the motor vehicle cannot stop in time before a crossing used by pedestrians, the motor vehicle should stop in time before the pedestrians on the crossing. On the crossing before the pedestrians means that the virtual stop line or virtual stop point before which the motor vehicle should stop is located on the crossing before the pedestrians in the direction of travel of the motor vehicle.

[0035] The rule is, for example, that if the motor vehicle is unable to stop in time ahead of the pedestrian at a crossing used by the pedestrian, the motor vehicle must perform an evasive maneuver in order to avoid the pedestrian.

[0036] In one embodiment of the method, it is provided that, based on the environmental signal, it is determined whether the crossing facility is free, wherein, if so, at least one boundary condition and / or at least one operating instruction predetermines that the travel of the motor vehicle is restricted to its current lane, wherein, if not, at least one boundary condition and / or at least one operating instruction predetermines that the motor vehicle stops before a virtual stop line that is located before the crossing facility in the travel direction of the motor vehicle.

[0037] This achieves the following technical advantages, for example: when the crossing is idle, the motor vehicle can efficiently pass through the crossing, or when the crossing is not idle, the motor vehicle can still stop before the crossing.

[0038] In one embodiment of the method, it is provided that at least one boundary condition that is (further) relaxed and / or at least one operating instruction that is (further) relaxed respectively predetermines one or more of the following predetermined values: stopping before a virtual stop line located on the crossing facility; stopping before a virtual stop line located on the crossing facility and before a traffic participant, in particular a pedestrian, who wishes to cross the crossing facility; driving around an obstacle, in particular a pedestrian, without collision; braking at a maximum predetermined maximum deceleration.

[0039] In the sense of this description, driving around means driving around without collision, ie driving around obstacles on the outside. The term "avoiding" can be used as a synonym for driving around.

[0040] This achieves, for example, the technical advantage that a useful plurality of fallback levels can be predefined or used.

[0041] In one embodiment of the method, it is provided that after the feedback "it is not possible to determine a backup target trajectory for the motor vehicle that satisfies at least one predetermined boundary condition and / or at least one operating instruction or the relaxed at least one boundary condition and / or at least one operating instruction", at least one relaxed boundary condition and / or at least one operating instruction are calculated (continued) sequentially, or wherein at least one relaxed boundary condition and / or at least one operating instruction are calculated (continued) in parallel, so that for the planning of a maneuver across the facility, the (continued) relaxed at least one boundary condition and / or at least one operating instruction is already available.

[0042] This achieves the technical advantage that, in the case of a sequential calculation, the backup level is calculated only if this is necessary in the specific situation. Existing hardware resources can thus be used efficiently.

[0043] This achieves the following technical advantage, for example: In the case of parallel computing, the reserve level can already be used for planning of a crossing maneuver. Thus, the planning of a crossing maneuver can be performed efficiently and quickly, so that a correspondingly fast response can be made to the specific existing traffic situation.

[0044] In one specific embodiment of the method, it is provided that the method is a computer-implemented method.

[0045] For example, the device according to the second aspect is configured in terms of programming to execute a computer program.

[0046] The embodiments described in connection with the device apply analogously to the method and vice versa, which means that device features are analogously derived from corresponding method features and vice versa.

[0047] When using the term "boundary conditions" it should always be used in conjunction with "operating instructions" and vice versa.

[0048] For example, the motor vehicle is an at least partially automated motor vehicle. This means that the motor vehicle can be guided at least partially automatically, ie the longitudinal guidance and / or the transverse guidance of the motor vehicle can be controlled at least partially automatically.

[0049] The expression "at least partially automatically guided" includes one or more of the following situations: assisted guided, partially automatically guided, highly automatically guided, fully automatically guided. Therefore, the expression "at least partially automatically guided" includes one or more of the following expressions: assisted, partially automatically guided, highly automatically guided, fully automatically guided.

[0050] Assisted guidance means that the driver of the motor vehicle continuously carries out either the transverse guidance or the longitudinal guidance of the motor vehicle. The respective other driving tasks are automatically performed (ie the longitudinal guidance or the transverse guidance of the motor vehicle is controlled). Thus, when the motor vehicle is assisted in guiding, either the transverse guidance or the longitudinal guidance is automatically controlled.

[0051] Partially automated guidance means that the longitudinal and lateral guidance of the motor vehicle is automatically controlled in certain situations (e.g. driving on a motorway, driving in a parking lot, overtaking an object, driving in a lane defined by lane markings) and / or for certain time periods. The driver of the motor vehicle does not have to manually control the longitudinal and lateral guidance of the motor vehicle himself. However, the driver must constantly monitor the automated control of the longitudinal and lateral guidance so that he can intervene manually if necessary. The driver must be ready to take over the full control of the motor vehicle at any time.

[0052] Highly automated guidance means that the longitudinal guidance and lateral guidance of the motor vehicle are automatically controlled for a specific period of time under specific conditions (for example: driving on a highway, driving in a parking lot, overtaking an object, driving in a lane determined by lane markings). The driver of the motor vehicle does not have to manually control the longitudinal guidance and lateral guidance of the motor vehicle himself. The driver does not have to continuously monitor the automated control of the longitudinal guidance and lateral guidance so that manual intervention can be made when necessary. When necessary, a takeover request is automatically output to the driver in order to take over the control of the longitudinal guidance and lateral guidance, especially with sufficient time margin. Therefore, the driver must potentially be able to take over the control of the longitudinal guidance and lateral guidance. The limits of the automated control of the lateral guidance and longitudinal guidance are automatically identified. In the case of highly automated guidance, it is not possible to automatically reach a state with minimal risk under various initial conditions.

[0053] Fully automated guidance means that the longitudinal guidance and transverse guidance of the motor vehicle are automatically controlled in certain situations (e.g., driving on a highway, driving in a parking lot, overtaking an object, driving in a lane determined by lane markings). The driver of the motor vehicle does not have to manually control the longitudinal guidance and transverse guidance of the motor vehicle. The driver does not have to monitor the automated control of the longitudinal guidance and transverse guidance so that manual intervention is possible when necessary. Before the automated control of the transverse guidance and longitudinal guidance ends, a request is automatically made to the driver to take over the driving task (control of the transverse guidance and longitudinal guidance of the motor vehicle), especially with sufficient time margin. If the driver does not take over the driving task, the risk is automatically returned to the state with the lowest risk. The limits of the automated control of the transverse guidance and longitudinal guidance are automatically identified. The automated return to the system state with the lowest risk can be achieved in all situations. The limits of the automated control of the longitudinal guidance and transverse guidance are automatically identified. The automated return to the system state with the lowest risk can be achieved in any situation.

[0054] The device comprises, for example, an analysis module and a planning module 807. The analysis module is implemented, for example, as a hardware module and / or as a software module. The planning module is implemented, for example, as a hardware module and / or as a software module.

[0055] For example, the two modules 805 and 807 are independent modules, that is, they are not related to each other.

[0056] For example, the device receives an environment signal representing an environment including the crossing facility of a motor vehicle traveling toward the crossing facility and wishing to cross the crossing facility. The environment signal comprises, for example, an environment model of the motor vehicle.

[0057] The environmental signal may be used, for example, by an analysis module and / or, for example, by a planning module.

[0058] For example, the analysis module analyzes the environmental signals, ie, analyzes the environment, and based on the analysis specifies, for example, one or more operating instructions and / or one or more boundary conditions to the planning module.

[0059] For example, the planning module plans a crossing facility maneuver for the motor vehicle based on the at least one predetermined operating instruction and / or at least one boundary condition and based on the environmental signal. If the planning module can no longer find a solution for the at least one predetermined operating instruction and / or at least one boundary condition, 805 can, for example, determine a relaxed boundary condition or a relaxed operating instruction and predetermine it to the planning module.

[0060] As a result of the planning by the planning module, a target trajectory is ascertained and output, based on which the motor vehicle can be guided at least partially automatically.

[0061] The embodiments and implementations described here can be combined with one another in any desired manner, even if this is not explicitly described. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] The present invention will be described in more detail below based on preferred embodiments. It is shown here:

[0063] Figure 1 : Block diagram of the method according to the first aspect,

[0064] Figure 2 : According to the device of the second aspect,

[0065] Figure 3 : According to the machine-readable storage medium of the fourth aspect,

[0066] Figure 4-7 :Different backup levels,

[0067] Figure 8 : a block diagram of a device according to the second aspect, and

[0068] Fig. 9 : Multiple backup levels.

[0069] In the following, the same reference numerals are used for the same features. DETAILED DESCRIPTION

[0070] Figure 1 A flow chart of a method for increasing safety at a crossing facility, in particular a pedestrian crossing, is shown, the method comprising the following steps:

[0071] receiving 101 an environment signal representing an environment including a crossing facility of a motor vehicle travelling towards the crossing facility and wishing to cross the crossing facility,

[0072] predefining 103 at least one boundary condition and / or at least one operating instruction for a crossing facility maneuver to be planned for the motor vehicle based on the environmental signal,

[0073] Based on the environmental signal, a crossing facility maneuver for the motor vehicle is planned 105 ,

[0074] Planning 105 includes determining 107 a target trajectory for the motor vehicle that satisfies the at least one predetermined boundary condition and / or the at least one operating command based on the at least one predetermined boundary condition and / or the at least one operating command,

[0075] If no target trajectory for the motor vehicle that satisfies the predetermined at least one boundary condition and / or at least one operating instruction can be determined, planning 105 includes using 109 at least one relaxed boundary condition and / or at least one relaxed operating instruction in order to determine an alternative target trajectory that satisfies the relaxed at least one boundary condition and / or at least one operating instruction,

[0076] Output 111 corresponds to the desired target trajectory.

[0077] For example, the method includes at least partially automated control of the transverse guidance and / or longitudinal guidance of the motor vehicle based on the output target trajectory.

[0078] Figure 2 A device 201 is shown which is configured to carry out all steps of the method according to the first aspect.

[0079] Figure 3 A machine-readable storage medium 301 is shown, on which a computer program 303 is stored. The computer program 303 comprises instructions which, when the computer program 303 is executed by a computer, cause the computer to carry out the method according to the first aspect.

[0080] Figure 4 A pedestrian crossing 401 is shown as an example of a crossing facility. A motor vehicle 403 is traveling toward the crossing facility, that is, the pedestrian crossing 401. Figure 4 , a pedestrian crossing 401, also known as a zebra crossing, is free. Therefore, a motor vehicle 403 can pass the pedestrian crossing 401. Reference numeral 405 points to a correspondingly determined target trajectory, based on which the motor vehicle 403 can pass the pedestrian crossing 401 at least partially automatically.

[0081] Figure 5 A pedestrian crossing 401 is shown, wherein a pedestrian 501 wishes to cross the pedestrian crossing 401. In this case, a boundary condition is that the motor vehicle 403 stops before a virtual stop line 503 located before the pedestrian crossing 401. In addition, an "×" is shown with reference mark 505, which is located on the virtual stop line 503 and symbolically identifies the following point: before this point, the motor vehicle 403 should stop.

[0082] If this boundary condition, ie stopping before the virtual stop line 503 which is still located before the pedestrian crossing 401, can be met, then according to Figure 6 It is provided that a relaxed boundary condition is used, which prescribes that the motor vehicle 403 stops before a virtual stop line 503 located on the pedestrian crossing 401 but still before the pedestrian 501, according to Figure 6The representation shown in , shows that pedestrian 501 has entered crosswalk 401 during this period.

[0083] If this boundary condition can no longer be satisfied, a further relaxed boundary condition is predefined: the motor vehicle 403 should drive around the pedestrian 501, that is, avoid the pedestrian 501. Figure 7 Target trajectory 405 is shown in FIG. 5 , which travels around an obstacle, currently a pedestrian 501 .

[0084] Figure 8 A block diagram 801 of an apparatus 803 according to the second aspect is shown.

[0085] The device 803 includes an analysis module 805 and a planning module 807. The analysis module 805 is implemented, for example, as a hardware module and / or as a software module. The planning module 807 is implemented, for example, as a hardware module and / or as a software module.

[0086] For example, the two modules 805 and 807 are independent modules, that is, they are not related to each other.

[0087] An environment signal 809 is provided to the device 803. This means that the device 803 receives an environment signal 809 representing the environment of a motor vehicle traveling towards the crossing and wishing to cross the crossing, including the crossing. The environment signal 809 comprises, for example, an environment model of the motor vehicle.

[0088] The ambient signal 809 is available both to the analysis module 805 and to the planning module 807 .

[0089] The analysis module analyzes the environmental signal 809, i.e. analyzes the environment, and based on the analysis, for example, predetermines one or more operating instructions and / or one or more boundary conditions to the planning module 807. The result of the analysis, i.e., the one or more operating instructions and / or one or more boundary conditions, is symbolically identified by a square with reference numeral 811.

[0090] The planning module 807 plans a crossing facility maneuver for the motor vehicle based on the at least one predetermined operating instruction and / or at least one boundary condition and based on the environmental signal. If the planning module 807 can no longer find a solution for the at least one predetermined operating instruction and / or at least one boundary condition, the analysis module 805 can, for example, determine a relaxed boundary condition or a relaxed operating instruction and prescribe it to the planning module 807.

[0091] As a result of the planning by planning module 807 , a target trajectory 813 is ascertained and output, based on which the motor vehicle can be guided at least partially automatically.

[0092] For example, the planning module 807 can reply feedback to the analysis module 805: it is not possible to obtain a spare target trajectory for the motor vehicle that satisfies at least one predetermined boundary condition and / or at least one operating instruction or at least one relaxed boundary condition and / or at least one operating instruction. In this case, the analysis module can, for example, further relax the at least one predetermined boundary condition and / or at least one operating instruction or at least one relaxed boundary condition and / or at least one operating instruction, and output the further relaxed at least one boundary condition and / or at least one operating instruction to the planning module 807. If the planning module 807 is also unable to obtain a spare target trajectory that satisfies at least one (further) relaxed boundary condition and / or at least one operating instruction for at least one (further) relaxed boundary condition and / or at least one operating instruction, the planning module 807 can, for example, again reply corresponding feedback to the analysis module 805, so that the analysis module can re-obtain at least one boundary condition and / or at least one operating instruction that continues to be relaxed and output it to the planning module 807.

[0093] Thus, for example, a feedback loop can be implemented, according to which, if the planning module 807 is unable to obtain a backup target trajectory for the motor vehicle that satisfies at least one predetermined boundary condition and / or at least one operating instruction or at least one relaxed boundary condition and / or at least one operating instruction, the planning module 807 always returns feedback to the analysis module 805, so that the analysis module 805 executes the step of obtaining at least one further relaxed boundary condition and / or at least one operating instruction, and executes outputting at least one further relaxed boundary condition and / or at least one operating instruction to the planning module 807, wherein the feedback loop runs in a loop in this way until the planning module 807 is able to obtain a backup target trajectory for the motor vehicle that satisfies at least one predetermined boundary condition and / or at least one operating instruction or at least one relaxed boundary condition and / or at least one operating instruction.

[0094] This optional feedback loop is symbolically indicated by a dashed arrow with reference numeral 815 .

[0095] Fig. 9 In the accompanying drawings, Figures 5 to 7 The traffic conditions shown in the top are shown in Figure 5 The traffic conditions shown in the figure are shown in the figure. Figure 7 The traffic conditions shown in the middle are shown in Figure 6 Traffic conditions shown in .

[0096] according to Figure 5The traffic situation at the top, that is, the traffic situation shown at the top, is assigned two boundary conditions, namely, the first boundary condition 901 and the second boundary condition 903. The traffic situation in the middle, that is, Figure 6 The traffic situation shown in FIG. 1 is assigned a third boundary condition 905 and a fourth boundary condition 907. The traffic situation below, i.e. Figure 7 A fifth boundary condition 909 is assigned to the traffic situation shown in FIG.

[0097] Boundary conditions 901 to 909 are further relaxed boundary conditions. For example, boundary conditions 901 to 909 include different braking behaviors, which can be symbolically represented by a diagram including abscissa 913 and ordinate 915. Three different braking levels are drawn on abscissa 913, respectively labeled I, II and III. Here, braking level II is greater than braking level I, and braking level III is greater than braking level II. Vertical axis 915 represents deceleration.

[0098] Depending on boundary conditions 901 to 909 , the corresponding marking for one of the braking levels I to III is shown in bold in order to make it clear that this braking level is predefined for this boundary condition.

[0099] Braking level indicator: According to the boundary conditions, the motor vehicle should brake at what deceleration level.

[0100] In summary, the approach described here is based in particular on a rule-set-based analysis module which, depending on the current situation, specifies boundary conditions for a facility-crossing maneuver to be planned, in particular a zebra crossing maneuver, to a subordinate planning module.

[0101] The following describes the solution using a zebra crossing as an example, and the solution can also be applied to other crossing facilities.

[0102] The analysis module does not rely on the feedback of the results of the lower-level planning layer, because it is not designed to generate driving strategies / options based on drivability, but on a proven set of rules. The drivability is then evaluated at a lower level, especially by the planning module. For example, if a time stabilization of the decision is necessary, feedback can be provided. According to the following scheme, the analysis module softens the boundary conditions or operating instructions to be observed when creating candidate solutions:

[0103] 1. Normal situation: Either the zebra crossing is empty, or the automated vehicle can cross the zebra crossing in compliance with the rules without danger, or the automated vehicle can stop at the stop line before the zebra crossing.

[0104] 2. Accident avoidance 1: The zebra crossing is occupied and the automated vehicle can no longer stop close to the zebra crossing. In this case, the stop line is set on the zebra crossing but before the pedestrian closest to the automated vehicle along the carriageway.

[0105] 3. Accident avoidance 2: The zebra crossing is occupied and the automated vehicle can no longer brake to a stop ahead of the nearest pedestrian. In this case, the stop line is completely eliminated and the object is predefined as an obstacle that the vehicle must drive around, i.e., the vehicle must avoid it. In addition, the vehicle must be predefined to brake.

[0106] The advantage of this solution is in particular that, when there is no longer a solution under compliance with all boundary conditions or operating instructions, the automated vehicle configured in this way can still find the target trajectory through the pre-given backup level. Here, the rigid requirements for the target trajectory are gradually relaxed so that the violations of boundary conditions or operating instructions are as few as possible. At the same time, downgrading is only carried out when solutions can no longer be found for stricter boundary conditions. In the approval of automated vehicles, rule-compliant and risk-avoiding handling of pedestrians, especially at zebra crossings, will be part of the approval argument and must be shown accordingly. The described solution can be used as a clear model in such approvals in order to disclose the mechanisms used and the resulting behaviors to the relevant authorities.

[0107] Legal regulations, laws and results from HARA (“Hazard analysis and risk assessment”; German: Gefahren-und Risikoanalyse (GuR)) define situation-dependent maximum deceleration limits. The described approach enables this maximum deceleration limit to be selected in a suitable manner.

[0108] For example, digital maps, dynamic occupancy grid maps and motor vehicle states (such as posture, speed, acceleration) are used. For example, the analysis module determines whether the zebra crossing is free based on the occupancy grid map. To this end, the analysis module uses, for example, the capture area input in the digital map. If the zebra crossing is free, the analysis module, for example, limits the travel of the automated motor vehicle to the lane only, so that the subordinate planning module plans: go forward through the zebra crossing. If another motor vehicle is traveling before the automated motor vehicle, the automated motor vehicle follows it. On the contrary, if a pedestrian is found in the capture area of ​​the zebra crossing (i.e., an occupied cell in the occupancy grid map), the analysis module, for example, forces the motor vehicle to stop at the stop line before the zebra crossing, which is also taken from the digital map.

[0109] If the subordinate planning module can no longer find a solution for stopping at a stop line under the predefined boundary conditions, the analysis module determines, by projecting the occupancy cells onto the lane and subtracting a safety margin in the longitudinal direction, a new stop line, for example, at which the automated vehicle must have stopped. In the case of multiple pedestrians in the capture area of ​​the zebra crossing, the stop line closest to the automated vehicle is selected. One embodiment provides that a rapid stop is additionally produced by adapting the optimization decision criterion in the subordinate planning module.

[0110] If the automated vehicle can no longer continue to comply with the stop line because the subordinate planning module can no longer find a solution for this, the stop line is completely cancelled. Instead, pedestrians are marked as obstacles that the automated vehicle must drive around by a circle surrounding the occupied cell, i.e., the vehicle must avoid the obstacle. For this purpose, the optimization criterion can also be adapted again in such a way that additional braking is achieved. However, if avoidance cannot be achieved at the same time (e.g., sliding lane) under the desired braking deceleration, the optimization criterion can be adapted to the situation in the sense of multi-objective optimization. However, this matching is always carried out, for example, so that the solution that best matches the situation from a technical point of view is provided, and, for example, a life-to-life confrontation is never carried out. In one embodiment, pedestrians can be represented more accurately based on predictions through polygons rather than through circles.

[0111] Alternatively or additionally, an object representation of the surroundings can be used instead of the occupancy representation. Furthermore, a fusion of the occupancy grid map with the object representation can be provided.

[0112] In one embodiment, the standard behavior and its backup strategy are not calculated sequentially, but in parallel. Due to the situation, there is little time available for reaction. A fixed predetermined time is available for the calculation planning module of the corresponding behavior. If the calculation cannot be completed within the predetermined time window, the corresponding behavior is classified as invalid. After the time is over or when all behaviors have been calculated, the behavior with the highest ranking is selected from the remaining valid behaviors.

[0113] In one embodiment, it is determined at an early stage which solutions are hopeless and therefore invalid based on an estimate. This evaluation can be performed with little computational effort. For example, an estimate of the braking path can be such an estimate.

[0114] In another embodiment, the maximum deceleration boundary is selected according to the conditions from laws, regulations, and results from HARA (see Fig. 9). Therefore, the braking deceleration can be divided into at least two categories. By assigning the behavior modes (normal situation, accident avoidance 1, accident avoidance 2) to the corresponding class of braking deceleration, a linearly ordered combination diversity is formed.

Claims

1. A method for improving safety at a crossing facility, in particular a pedestrian crossing (401), the method comprising the following steps: receiving (101) an environment signal (809) representing an environment of a motor vehicle (403) travelling towards a crossing facility and wishing to cross the crossing facility, including the crossing facility, Based on the environmental signal (809), at least one boundary condition and / or at least one operating instruction is predefined (103) for a planned crossing facility maneuver for the motor vehicle (403), planning (105) a facility-crossing maneuver for the motor vehicle (403) based on the environmental signal (809), The planning (105) comprises determining (107) a target trajectory (405, 813) for the motor vehicle (403) that satisfies the at least one predetermined boundary condition and / or the at least one operating instruction based on the at least one predetermined boundary condition and / or the at least one operating instruction, wherein, if a target trajectory (405, 813) for the motor vehicle (403) that satisfies at least one predetermined boundary condition and / or at least one operating instruction cannot be determined, the planning (105) includes using (109) at least one relaxed boundary condition and / or at least one relaxed operating instruction in order to determine an alternative target trajectory (405, 813) that satisfies the at least one relaxed boundary condition and / or at least one operating instruction, The output (111) corresponds to the obtained target trajectory (405, 813).

2. The method according to claim 1, wherein: If a backup target trajectory for the motor vehicle (403) that satisfies the at least one relaxed boundary condition and / or at least one operating instruction cannot be determined, the planning (105) includes using at least one further relaxed boundary condition and / or at least one further relaxed operating instruction for a long time until a backup target trajectory that satisfies the corresponding further relaxed at least one boundary condition and / or at least one operating instruction is determined, or until a backup target trajectory that satisfies the corresponding further relaxed at least one boundary condition and / or at least one operating instruction is determined at least within a predetermined range.

3. The method according to claim 1 or 2, wherein: A motor vehicle state signal describing a motor vehicle state is received, wherein the specification (103) and the planning (105) are performed based on the motor vehicle state signal.

4. A method according to any one of the preceding claims, wherein: The specification (103) is performed based on a predetermined rule set, which includes one or more rules for the behavior of a motor vehicle (403) at a crossing facility.

5. A method according to any one of the preceding claims, wherein: Based on the environmental signal (809), it is determined whether the crossing facility is idle, wherein, if yes, at least one boundary condition and / or at least one operating instruction is predetermined: the travel of the motor vehicle (403) is restricted to its current lane, wherein, if no, at least one boundary condition and / or at least one operating instruction is predetermined: the motor vehicle (403) stops before the following virtual stop line (503): the virtual stop line (503) is located before the crossing facility in the travel direction of the motor vehicle (403).

6. The method according to claim 5, wherein: At least one (continued) relaxed boundary condition and / or at least one (continued) relaxed operating instruction respectively predetermine one or more of the following predetermined conditions: stopping before a virtual stop line (503) located on the crossing facility; stopping before a virtual stop line (503) located on the crossing facility and before a traffic participant, especially a pedestrian (501), who wishes to cross the crossing facility; driving around an obstacle, especially a pedestrian (501) without collision; braking at a maximum predetermined maximum deceleration.

7. A method according to any one of the preceding claims, wherein: After the feedback that "it is not possible to determine a backup target trajectory for the motor vehicle (403) that satisfies at least one predetermined boundary condition and / or at least one operating instruction or at least one relaxed boundary condition and / or at least one operating instruction", the (further) relaxed at least one boundary condition and / or at least one operating instruction are calculated sequentially, or the (further) relaxed at least one boundary condition and / or at least one operating instruction are calculated in parallel, so that the (further) relaxed at least one boundary condition and / or at least one operating instruction are already available for the planning (105) of the maneuvering action of crossing the facility.

8. A device (201, 803) arranged to perform all the steps of the method according to any one of the preceding claims.

9. A computer program (303), comprising instructions which, when the computer program (303) is executed by a computer, cause the computer to carry out the method according to any one of claims 1 to 7.

10. A machine-readable storage medium (301) on which a computer program (303) according to claim 9 is stored.

Images