A method and system for converting traffic rules into machine language

Abstract

The application relates to a traffic rule conversion method and system, belonging to the technical field of traffic rule conversion, which comprises the following steps: obtaining a target natural language traffic rule; determining the fuzzy relationship of traffic participants in the natural language traffic rule according to the scene information of the natural language traffic rule; formalizing each main component in the fuzzy relationship of the traffic participants in the natural language traffic rule based on a logic language; and combining each formalized main component into a complete machine language traffic rule according to logic operators and time sequence operators. The combination of the fuzzy time relationship and the fuzzy position relationship and the application of the logic operators and the time sequence operators realize the complete conversion of the natural language traffic rule into the machine language traffic rule, the machine can intuitively understand, and the method can be applied to most traffic rules, simple scenes and complex scenes, and has good applicability.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of traffic rule language conversion, and particularly relates to a traffic rule to machine language conversion method and system. BACKGROUND

[0002] With the development of society, more and more machines in modern life are developing towards automation and intelligence, and cars for mobile travel are no exception. Intelligent cars are gradually entering people's daily life. In recent years, ADAS (Advanced Driving Assistant System) plays a very important role in intelligent cars. It uses various sensors installed on the car to sense the environment around the car at any time during driving, collects data, identifies, detects and tracks stationary and moving objects, and combines navigation map data to perform system operation and analysis, so as to pre-alert the driver of possible dangers and effectively increase the comfort and safety of car driving. It can be said that true automatic driving is the result of the development of ADAS to the extreme.

[0003] The functions to be realized by automatic driving mainly include perception and prediction, decision planning and vehicle control, etc. The perception module is the "eyes" of the automatic driving car, which is used to perceive the environment and obstacles, etc. The prediction module will predict the trajectory of the obstacle in the subsequent certain time according to the perceived and tracked obstacle information. The decision planning module is the "brain" of the automatic driving car, which generally realizes three functions, including path planning, behavior decision and motion planning, etc. At present, the design and testing of the decision planning module mainly focus on driving safety and part of simple traffic rules, such as no collision with obstacles, compliance with traffic lights and road signs, no speeding, etc., which far from cover enough traffic rules. And the first problem of the automatic driving system found in the closed test field is "traffic rule compliance is not sound", so whether the automatic driving car complies with the traffic rules is extremely important in the planning module design stage or the automatic driving vehicle or control module testing stage.

[0004] However, the current traffic rules are described by natural language and are oriented to human drivers. The natural language traffic rules often contain some fuzzy semantics, and the related technologies for checking compliance with traffic rules are mostly based on human judgment, which is difficult for machines to understand directly. For example, "motor vehicles entering the highway from the ramp shall enter the lane without hindering the normal driving of motor vehicles already on the highway", and it is difficult for machines to understand what "without hindering the normal driving of other vehicles" means. Therefore, how to convert the existing natural language described traffic rules into machine language traffic rules that machines can understand has become a technical problem to be solved in the prior art. SUMMARY

[0005] The application provides a traffic rule to machine language conversion method and system to solve the technical problem that machines are difficult to understand natural language ambiguous traffic rules in the prior art.

[0006] The technical scheme provided by the application is as follows:

[0007] In one aspect, a traffic rule to machine language conversion method comprises the following steps:

[0008] Obtaining a target natural language traffic rule; the target natural language traffic rule comprises scene information;

[0009] Determining the ambiguous relationship of traffic participants in the natural language traffic rule according to the scene information of the natural language traffic rule; the ambiguous relationship of traffic participants comprises an ambiguous time relationship between actions of traffic participants and an ambiguous location relationship between traffic participants;

[0010] Formalizing each main component in the ambiguous relationship of traffic participants in the natural language traffic rule based on a logic language;

[0011] Combining the formalized main components into a complete machine language traffic rule according to logic operators and time sequence operators.

[0012] Optionally, the step of obtaining a target natural language traffic rule comprises the following steps:

[0013] Obtaining a natural language traffic rule;

[0014] Determining a target natural language traffic rule in the natural language traffic rule based on a screening rule for screening motor vehicle states and behaviors.

[0015] Optionally, the step of determining the ambiguous relationship of traffic participants in the natural language traffic rule according to the scene information of the natural language traffic rule comprises the following steps:

[0016] Searching for all verbs in the target natural language traffic rule and analyzing whether there is an ambiguous time sequence relationship between verbs and verbs;

[0017] If there is an ambiguous time sequence relationship between verbs and verbs, determining the verbs with ambiguous time sequence relationship and corresponding adjectives as an ambiguous time relationship field;

[0018] Determining the time interval between actions in the ambiguous time relationship field according to the scene information in the natural language traffic rule;

[0019] Determining the ambiguous time relationship between actions of traffic participants according to the ambiguous time relationship field and the time interval.

[0020] Optionally, the time interval is the sum of a scene required time interval and a time allowance.

[0021] In the case that the preset emergency word is contained in the scene information, the scene requirement time interval is sub-second level, and the time margin is ten millisecond level.

[0022] In the case that the preset emergency word is not contained in the scene information, the scene requirement time interval is second level, and the time margin is sub-second level.

[0023] Optionally, the determining the ambiguous relationship of the traffic participants in the natural language traffic regulation according to the scene information of the natural language traffic regulation comprises:

[0024] searching all position relationship related nouns and corresponding modifiers in the target natural language traffic regulation, and analyzing whether any of the nouns and the corresponding modifiers are ambiguous position relationship;

[0025] if any of the nouns and the corresponding modifiers are ambiguous position relationship, determining the nouns and the corresponding modifiers as the ambiguous position relationship field;

[0026] determining the distance interval of the position relationship in the ambiguous position relationship field according to the scene information in the natural language traffic regulation;

[0027] determining the ambiguous position relationship between the traffic participants according to the ambiguous position relationship field and the distance interval.

[0028] Optionally, the distance interval is the sum of the scene requirement distance interval and the distance margin.

[0029] The distance interval comprises longitudinal safety distance, lateral safety distance and comprehensive safety distance.

[0030] In the case that the distance interval is longitudinal safety distance, the scene requirement distance interval is at least one of time interval, absolute distance, TTC distance or STD distance.

[0031] In the case that the distance interval is lateral safety distance, the scene requirement distance interval is at least one of time interval, absolute distance, TTC distance or STD distance.

[0032] In the case that the distance interval is comprehensive safety distance, the distance interval is decomposed into two directions of lateral and longitudinal, and the scene requirement distance interval is at least one of time interval, absolute distance, TTC distance or STD distance.

[0033] Optionally, the logical operator comprises at least one of negation, conjunction, disjunction, deduction and equivalence.

[0034] The timing operator comprises at least one of next time, until, always, eventually, always in the time period and in the time period.

[0035] In another aspect, a traffic rule to machine language conversion system comprises an acquisition module, a fuzzy relationship determination module, a formalization module and a synthesis module.

[0036] The acquisition module is configured to acquire a target natural language traffic rule.

[0037] The fuzzy relationship determination module is configured to determine a fuzzy relationship of a traffic participant in the natural language traffic rule according to scene information of the natural language traffic rule.

[0038] The formalization module is configured to formalize each main component in the fuzzy relationship of the traffic participant in the natural language traffic rule based on a logic language.

[0039] The synthesis module is configured to combine the formalized main components into a complete machine language traffic rule according to a logic operator and a timing operator.

[0040] The present application has the following advantages:

[0041] The traffic rule to machine language conversion method and system provided by the present application comprises the following steps: acquiring a target natural language traffic rule; determining a fuzzy relationship of a traffic participant in the natural language traffic rule according to scene information of the natural language traffic rule; formalizing each main component in the fuzzy relationship of the traffic participant in the natural language traffic rule based on a logic language; and combining the formalized main components into a complete machine language traffic rule according to a logic operator and a timing operator. BRIEF DESCRIPTION OF DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0043] Figure 1 A flowchart of a traffic rule to machine language conversion method provided by the present application is shown in the figure.

[0044] Figure 2 A specific target natural language rule formalization example schematic diagram is provided for the embodiment of the present application.

[0045] Figure 3 A structure schematic diagram of a traffic rule to machine language conversion system is provided for the embodiment of the present application. DETAILED DESCRIPTION

[0046] In order to make the objectives, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be described in detail below. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of the present application.

[0047] The decision planning module of automatic driving is the "brain" of the automatic driving car, which generally implements three functions, including path planning, behavior decision and motion planning. Among them, path planning refers to planning a path from the starting point to the destination according to the starting point and the destination, combined with the road information in the map; behavior decision will make specific behavior decisions such as overtaking, lane changing, etc. according to the results of path planning and current environment and obstacle information; motion planning will plan a trajectory point that satisfies certain constraints according to the results of behavior decision, such as trajectory point when overtaking.

[0048] Automatic driving vehicles need to be tested in various ways, which are currently mainly concentrated in closed or open road testing at the whole vehicle level or compliance and performance testing at the component level. The decision planning module as the "on-board brain" is very important for automatic driving cars, and whether the output of the decision planning module is correct directly affects the safety of the automatic driving car, so the decision planning module needs to follow some safety, legality and other criteria whether in the design stage or in the testing stage. The current design and testing of the decision planning module are mainly concentrated in driving safety and part of simple traffic rule legality, such as not colliding with obstacles, following traffic lights and road signs, not speeding, etc., which far from cover enough traffic rules; and the first problem of the automatic driving system found in the closed test field is "traffic rule compliance is not sound", so whether in the planning module design stage or in the automatic driving whole vehicle or control module testing stage, the test of whether the automatic driving car complies with the traffic rules is a very important link.

[0049] However, the current traffic rules are described by natural language and are oriented to human drivers, and the natural language traffic rules often contain some ambiguous semantics, and the checking of compliance with the traffic rules in the related art is also based on human judgment, and machines are difficult to directly understand. For example, "the motor vehicle driving from the ramp into the highway shall drive into the lane without hindering the normal driving of the motor vehicle in the highway", and the machine is difficult to understand what is "without hindering the normal driving of the other vehicle". Therefore, how to formulate a machine language traffic rule convenient for machine understanding based on the existing natural language description of the traffic rule has become a technical problem to be solved in the prior art.

[0050] Based on this, the embodiment of the present application provides a traffic rule to machine language conversion method.

[0051] Figure 1 A flowchart of a traffic rule to machine language conversion method provided by the embodiment of the present application is shown in Figure 1 The method provided by the embodiment of the present application can include the following steps:

[0052] S11, obtaining a target natural language traffic rule; the target natural language traffic rule includes scene information.

[0053] In a specific implementation process, the traffic rule to machine language conversion method provided by the present application can be applied in any natural language to machine language conversion system, especially the traffic rule to machine language conversion method provided by the present application can be applied in a traffic regulation digitization system; the traffic rule to machine language conversion method provided by the embodiment of the present application can also be applied in an automatic driving simulation system or an automatic driving vehicle system to realize formalization (digitization) of the traffic regulations required to be complied with by the automatic driving decision planning module, and the present application is not limited in particular.

[0054] For example, the target natural language traffic rule can be obtained first, that is, the traffic rule described by natural language is obtained.

[0055] In some embodiments, optionally, the target natural language traffic rule is obtained by: obtaining a natural language traffic rule; determining the target natural language traffic rule in the natural language traffic rule based on a screening rule screening the state and behavior of the motor vehicle.

[0056] For example, the natural language traffic regulation can be obtained by traffic regulations, and the clauses that need to be formalized can be screened out one by one. In the screening process, the natural language traffic regulation screened out can be the target natural language traffic regulation according to the principle of restricting the state and behavior of the motor vehicle. The target natural language traffic regulation is not limited here, and is described by an embodiment, for example, the target natural language traffic regulation can be: "When a motor vehicle breaks down on the road and needs to be parked to troubleshoot, the driver should immediately turn on the hazard warning flasher". The scene information can also be "When a motor vehicle breaks down on the road and needs to be parked to troubleshoot, the driver should immediately turn on the hazard warning flasher", or part of the information, which is not limited here.

[0057] S12, determining the fuzzy relationship of the traffic participants in the natural language traffic regulation according to the scene information of the natural language traffic regulation.

[0058] In this embodiment, the fuzzy relationship of the traffic participants includes a fuzzy time relationship between the actions of the traffic participants and a fuzzy position relationship between the traffic participants.

[0059] When the fuzzy relationship of the traffic participants is the fuzzy time relationship between the actions of the traffic participants, in some embodiments, the fuzzy relationship of the traffic participants in the natural language traffic regulation can be determined according to the scene information of the natural language traffic regulation, including: searching for all verbs in the target natural language traffic regulation, analyzing whether there is an ambiguous time sequence relationship between the verbs and the verbs; if there is an ambiguous time sequence relationship between the verbs and the verbs, determining the verbs and the corresponding modifiers with the ambiguous time sequence relationship as the fuzzy time relationship field; determining the time interval between the actions in the fuzzy time relationship field according to the scene information in the natural language traffic regulation; and determining the fuzzy time relationship between the actions of the traffic participants according to the fuzzy time relationship field and the time interval.

[0060] For example, all verbs in the target natural language traffic regulation can be searched according to existing search techniques, and whether there is an ambiguous time sequence relationship between the verbs and the verbs can be analyzed to form a candidate fuzzy time relationship field set. For example, the ambiguous time sequence relationship can be no specific time interval.

[0061] In the embodiment, the fuzzy time relationship field can be determined by marking. When there is an ambiguous time sequence between verbs, the verbs and their related modifiers that have a relationship are marked as the fuzzy time relationship field. For example, "when a motor vehicle breaks down on the road and needs to be parked to eliminate the fault, the driver should immediately turn on the hazard warning flasher", wherein "occurs" and "turn on" are verbs, and the actions "motor vehicle breaks down on the road" and "turn on the hazard warning flasher" have a time sequence and no specific time interval, so "breaks down" and "should immediately turn on" are marked as candidate fuzzy time relationship fields; otherwise, no marking is performed.

[0062] In some embodiments, optionally, the time interval is the sum of the scene requirement time interval and the time margin; when the preset emergency word is included in the scene information, the scene requirement time interval is sub-second, and the time margin is ten milliseconds; when the preset emergency word is not included in the scene information, the scene requirement time interval is second, and the time margin is sub-second.

[0063] For example, the time interval Δt between actions in the fuzzy time relationship field can be determined according to different scenes, Δt = Δt s + Δt margin ; wherein, is the scene requirement time interval, and Δt margin is the time margin.

[0064] In the embodiment, the preset emergency word can be "immediately", "soon", etc., which is not specifically limited here and can be set according to specific requirements. In the embodiment, "immediately" is taken as an example to illustrate the technical solution. For example, for a scene in which the time interval requirement strict word such as "immediately" appears, the scene requirement time interval Δt is sub-second, and the time margin Δt margin is ten milliseconds; for a scene in which the time interval requirement loose word such as "immediately" does not appear, the scene requirement time interval Δt is second, and the time margin Δt margin is sub-second.

[0065] Based on the fuzzy time relationship field and the time interval, the fuzzy time relationship between the actions of the traffic participants can be determined. For example, the fuzzy time relationship between the actions of the traffic participants in the traffic participant action of "when a motor vehicle breaks down on the road and needs to be parked for troubleshooting, the driver should immediately turn on the hazard warning flasher" is that the fuzzy time relationship field is "breaks down" and "should immediately turn on", and the time interval is 0.5 seconds. The calculation method of 0.5 seconds is explained as follows: the scene information can be "when a motor vehicle breaks down on the road and needs to be parked for troubleshooting, the driver should immediately turn on the hazard warning flasher", which contains "immediately", so the scene requirement time interval can be sub-second, the time margin is millisecond, and the sum is 0.5 seconds. Here, 0.5 seconds is only an example and is not limited.

[0066] When the fuzzy relationship of the traffic participants is the fuzzy position relationship between the traffic participants, in some embodiments, the fuzzy relationship of the traffic participants in the natural language traffic regulation is determined according to the scene information of the natural language traffic regulation, including: searching all position relationship related nouns and corresponding modifiers in the target natural language traffic regulation, analyzing whether any noun and corresponding modifier is an ambiguous position relationship; if any noun and corresponding modifier is an uncertain position relationship, determining that the noun and the corresponding modifier are the fuzzy position relationship field; determining the distance interval of the position relationship in the fuzzy position relationship field according to the scene information in the natural language traffic regulation; and determining the fuzzy position relationship between the traffic participants according to the fuzzy position relationship field and the distance interval.

[0067] For example, all position relationship related nouns and their modifiers in the target natural language traffic regulation can be searched according to existing search techniques, and whether they are an ambiguous position relationship is analyzed to form a candidate fuzzy position relationship field set. For example, the ambiguous position relationship can be a position relationship without a specific value or mathematical model.

[0068] In this embodiment, the fuzzy position relationship field can be determined by marking. When the position relationship related noun and its modifier describe an ambiguous position relationship, the position related noun and its related modifier are marked as the fuzzy position relationship field. For example, "the following vehicle should overtake from the left side of the front vehicle after confirming that there is sufficient safety distance", wherein "sufficient safety distance" is a position relationship related noun and its modifier, but there is no specific position relationship to explain what sufficient safety distance is, and the sufficient safety distance is an ambiguous position relationship, so it is marked as a candidate fuzzy position relationship field; otherwise, it is not marked.

[0069] In some embodiments, optionally, the distance interval is a sum of a scene-required distance interval and a distance margin; the distance interval comprises a longitudinal safety distance and / or a lateral safety distance; when the distance interval is the longitudinal safety distance, the scene-required distance interval is at least one of a time-to-collision (TTC) distance, an absolute distance, a time-to-closest (TTC) distance, or a standard time-to-collision (STD) distance; when the distance interval is the lateral safety distance, the scene-required distance interval is at least one of a time-to-collision (TTC) distance, an absolute distance, a time-to-closest (TTC) distance, or a standard time-to-collision (STD) distance; when the distance interval is a comprehensive safety distance, the distance interval is decomposed into a lateral direction and a longitudinal direction, and the scene-required distance interval is at least one of a time-to-collision (TTC) distance, an absolute distance, a time-to-closest (TTC) distance, or a standard time-to-collision (STD) distance.

[0070] wherein the TTC distance is a time-to-collision (TTC) distance algorithm; the TTC distance and the STD distance are both prior art, and are not described herein.

[0071] For example, the distance interval of the candidate ambiguous position relationship is determined according to different scenes

[0072] △d = △d s + △d margin wherein, according to different scenes, is a scene-required distance interval, and △d margin is a distance margin.

[0073] In this embodiment, for the case of the longitudinal safety distance: The model used can be a time-to-collision (TTC) distance, an absolute distance, a time-to-closest (TTC) distance, a standard time-to-collision (STD) distance, or a joint model of various models.

[0074] For the case of the lateral safety distance: The model used can be a time-to-collision (TTC) distance, an absolute distance, a time-to-closest (TTC) distance, a standard time-to-collision (STD) distance, or a joint model of various models.

[0075] For the case where it is not specified whether it is a longitudinal safety distance or a lateral safety distance: both the longitudinal safety distance and the lateral safety distance are considered, can be decomposed into a lateral direction and a longitudinal direction, and the corresponding and △d margin are also decomposed into a lateral direction and a longitudinal direction. The model used can be a time-to-collision (TTC) distance, an absolute distance, a time-to-closest (TTC) distance, a standard time-to-collision (STD) distance, or a joint model of various models.

[0076] For a close-range scene △d margin is centimeter-level, and for a long-range scene △d margin is decimeter-level.

[0077] Based on the fuzzy position relationship field and the distance interval, the fuzzy position relationship between the traffic participants can be determined. For example, the fuzzy position relationship between the traffic participants of "the rear vehicle should overtake the front vehicle from the left side of the front vehicle after confirming that there is sufficient safety distance" is that the fuzzy position relationship field is "sufficient safety distance" and the distance interval is 10 meters. The 10 meters is calculated according to the specific scene distance interval and distance margin, and the calculation process is simple addition, subtraction, multiplication and division operation, which is not listed here.

[0078] S13, based on the logic language, formalizing each main component in the fuzzy relationship of the traffic participants in the natural language traffic rules.

[0079] After obtaining the specific fuzzy relationship of the traffic participants, the natural language traffic rules after clarification can be obtained, and each main component in the natural language traffic rules after clarification is formalized according to the logic language. The logic language can be linear temporal logic (LTL).

[0080] Figure 2 A specific target natural language traffic rule formalization example diagram is provided for the embodiments of the present application.

[0081] For example, refer to Figure 2 The target natural language traffic rule is "when a vehicle on a road breaks down and needs to be parked for troubleshooting, the hazard warning flasher should be turned on immediately and the minimum risk braking should be performed", and the natural language traffic rule after clarification is "when a vehicle breaks down and needs to be parked for troubleshooting, the hazard warning flasher should be turned on within 0.5 seconds and parked within 2 seconds". The formalization of the natural language traffic rule after clarification is performed by taking onRoad, stBreakDown, stSignalEmergency and stStop as the formalized propositions of each main part in the natural language traffic rule after clarification. It should be noted that this is only an enumeration of formalization, and is not limited.

[0082] S14, combining each main component after formalization into complete machine language traffic rules according to logic operators and temporal operators.

[0083] In some embodiments, the logic operators include at least one of negation, conjunction, disjunction, implication and equivalence, and the temporal operators include at least one of next time, until, always, finally, always within the time period and within the time period.

[0084] For example, Table 1 is a summary table of logic operators provided by the embodiments of the present application, and Table 2 is a summary table of temporal operators provided by the embodiments of the present application.

[0085] Table 1 Summary table of logic operators

[0086]

[0087] Table 2 Summary of temporal operators

[0088]

[0089] According to the logical operators and the temporal operators, the formalized main components are combined to form a complete machine language traffic regulation.

[0090] For example, referring to Figure 2 , the formalized main components are combined according to the logical operators and the temporal operators, which can be .

[0091] Referring to Figure 2 , after obtaining the natural language traffic regulation in the first step, the fuzzy time relationship is determined in the manner provided in the embodiments, and then the main texts of the natural language traffic regulation are formalized using temporal logic (such as LTL or MTL), and finally all the formalized components are combined into a complete formalized traffic regulation, i.e., a machine language traffic regulation, using the logical operators (Table 1) and the temporal operators (Table 2).

[0092] In some embodiments, when the formalized traffic regulation is applied subsequently, the propositions onRoad, stBreakDown, stSignalEmergency, and stStop can output true or false based on input data (such as vehicle trajectory information, environmental perception information, etc.), and in combination with the logical conjunction words, the true or false of the entire formalized traffic regulation proposition is obtained, so as to determine whether the autonomous vehicle violates the regulation.

[0093] In the above-described embodiments, the subject of the conversion of the traffic regulation into the machine language can be a traffic regulation formalization personnel, or a machine, such as a computer specially used for formalizing natural language traffic regulations or a “brain” of an autonomous vehicle. Only the above steps of obtaining the traffic regulation, searching, and determining the fuzzy time relationship and the position relationship by a person are replaced by machine operation.

[0094] In the present embodiment, the conversion method of the traffic regulation into the machine language provided in the present application can be used online or offline, which is not specifically limited in the present embodiment. For example, the conversion of the existing traffic regulation into the machine language is the formalization of the traffic regulation offline; for the autonomous vehicle driving on the road, if the updated natural language traffic regulation is transmitted to the vehicle-mounted “brain” through a wireless signal or the regulation is recorded in the form of words on a sign, and the autonomous vehicle recognizes the regulation described in the words through a vehicle-mounted sensor such as a camera, and the vehicle-mounted brain performs the above formalization operation on the traffic regulation, which is the real-time formalization of the traffic regulation online.

[0095] The traffic rule to machine language conversion method provided by the embodiment of the present application comprises the following steps: obtaining a target natural language traffic rule; determining a fuzzy relationship of traffic participants in the natural language traffic rule according to scene information of the natural language traffic rule; formalizing each main component in the fuzzy relationship of the traffic participants in the natural language traffic rule based on a logic language; and combining each formalized main component into a complete machine language traffic rule according to a logic operator and a timing operator. The complete conversion of the natural language traffic rule to the machine language traffic rule is realized by the combined application of the fuzzy time relationship and the fuzzy position relationship pre-logic operator and the timing operator, so that the machine can intuitively understand, can be applied to most traffic rules, can be applied to simple scenes and complex scenes, and has good applicability.

[0096] Based on the overall inventive concept, the embodiment of the present application further provides a traffic rule to machine language conversion system.

[0097] Figure 3 The structure diagram of the traffic rule to machine language conversion system provided by the embodiment of the present application is shown in Figure 3 The system provided by the embodiment of the present application can comprise an acquisition module 31, a fuzzy relationship determination module 32, a formalization module 33 and a combination module 34.

[0098] The acquisition module 31 is configured to acquire a target natural language traffic rule; the target natural language traffic rule comprises scene information.

[0099] The fuzzy relationship determination module 32 is configured to determine a fuzzy relationship of traffic participants in the natural language traffic rule according to the scene information of the natural language traffic rule; the fuzzy relationship of the traffic participants comprises a fuzzy time relationship between actions of the traffic participants and a fuzzy position relationship between the traffic participants.

[0100] The formalization module 33 is configured to formalize each main component in the fuzzy relationship of the traffic participants in the natural language traffic rule based on a logic language.

[0101] The combination module 34 is configured to combine each formalized main component into a complete machine language traffic rule according to a logic operator and a timing operator.

[0102] In some embodiments, the acquisition module 31 is configured to acquire a natural language traffic rule; and the target natural language traffic rule is determined in the natural language traffic rule based on a screening rule for screening the state and behavior of a motor vehicle.

[0103] In some embodiments, the fuzzy relationship determination module 32 is configured to search all verbs in the target natural language traffic rule and analyze whether there is an ambiguous time sequence relationship between the verbs.

[0104] If there is an ambiguous time sequence between the verbs, the verb with the ambiguous time sequence and the corresponding modifier are determined as the ambiguous time relationship field;

[0105] According to the scene information in the natural language traffic rules, the time interval between the actions in the ambiguous time relationship field is determined.

[0106] According to the ambiguous time relationship field and the time interval, the ambiguous time relationship between the actions of the traffic participants is determined.

[0107] In some embodiments, the optional ambiguous relationship determining module 32 is configured to search for all position relationship related nouns and corresponding modifiers in the target natural language traffic rules, and analyze whether any noun and corresponding modifier is an ambiguous position relationship.

[0108] If any noun and corresponding modifier is an ambiguous position relationship, the noun and corresponding modifier are determined as the ambiguous position relationship field.

[0109] According to the scene information in the natural language traffic rules, the distance interval of the position relationship in the ambiguous position relationship field is determined.

[0110] According to the ambiguous position relationship field and the distance interval, the ambiguous position relationship between the traffic participants is determined.

[0111] As to the system in the above embodiments, the specific manner in which each module performs the operation has been described in detail in the embodiments related to the method, and will not be described in detail here.

[0112] The traffic rule to machine language conversion system provided by the embodiments of the present application converts the natural language traffic rules to machine language traffic rules by obtaining the target natural language traffic rules, determining the ambiguous relationship of the traffic participants in the natural language traffic rules according to the scene information in the natural language traffic rules, formalizing each main component in the ambiguous relationship of the traffic participants in the natural language traffic rules based on the logical language, and combining the formalized main components into complete machine language traffic rules according to the logical operator and the time sequence operator. The complete conversion of the natural language traffic rules to machine language traffic rules is realized by the combined application of the fuzzy time relationship and the fuzzy position relationship pre-logical operator and time sequence operator, so that the machine can intuitively understand, can be applied to most traffic rules, can be applied to simple scenes and complex scenes, and has good applicability.

[0113] The above merely provides a specific implementation of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0114] It can be understood that the same or similar parts in the above-mentioned embodiments can be mutually referred to, and the content not described in detail in some embodiments can be referred to the same or similar content in other embodiments.

[0115] It should be noted that, in the description of the present application, the terms "first", "second" and the like are used only for descriptive purposes, and cannot be construed as indicating or implying relative importance. In addition, in the description of the present application, unless otherwise specified, the meaning of "a plurality of" is at least two.

[0116] Any process or method descriptions in flow charts or described herein otherwise can be understood as representing code modules, segments, or portions of code that include one or more executable instructions for performing specific logic functions or steps in the process, and the various embodiments of the present application include additional implementations in which the functions described with reference to a given code module, segment, or portion of code are combined with functions described with reference to one or more other code modules, segments, or portions of code. The computer-readable media can include a computer-readable storage medium that is non-transitory.

[0117] It should be understood that each part of the present application can be realized by hardware, software, firmware or their combination. In the above-mentioned embodiments, a plurality of steps or methods can be realized by software or firmware stored in the memory and executed by the appropriate instruction execution system. For example, if realized by hardware, and as in another embodiment, it can be realized by any one or their combination of the following technologies known in the art: discrete logic circuit with logic gate circuit for implementing logic function on data signal, application specific integrated circuit with appropriate combination of logic gate circuit, programmable gate array (PGA), field programmable gate array (FPGA) and the like.

[0118] Those skilled in the art of the present technology can understand that all or part of the steps carried out by the above-mentioned embodiment method can be completed by program instructions to the relevant hardware, and the program can be stored in a computer-readable storage medium, which includes one or a combination of steps of the method embodiments when executed.

[0119] In addition, each functional unit in each embodiment of the present application can be integrated in one processing module, or each unit can be physically present separately, or two or more units can be integrated in one module. The above-mentioned integrated module can be realized in the form of hardware or in the form of software functional module. The integrated module, if realized in the form of software functional module and sold or used as an independent product, can also be stored in a computer-readable storage medium.

[0120] The above-mentioned storage medium can be read-only memory, magnetic disk or optical disk, etc.

[0121] In the description of the specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the specification, the illustrative description of the above terms does not necessarily mean the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0122] Although the embodiments of the present application have been shown and described above, it is understood that the above-described embodiments are exemplary, and cannot be construed as limiting the present application, and those skilled in the art can make changes, modifications, replacements and variations to the above-described embodiments within the scope of the present application.

Claims

1. A method of converting traffic rules into machine language, characterized by, The method comprises the following steps: acquiring a target natural language traffic regulation; the target natural language traffic regulation comprises scene information; determining the ambiguous relationship of traffic participants in the natural language traffic regulation according to the scene information of the natural language traffic regulation; the ambiguous relationship of traffic participants comprises an ambiguous time relationship between actions of traffic participants and an ambiguous position relationship between traffic participants; wherein, the step of determining the ambiguous relationship of traffic participants in the natural language traffic regulation according to the scene information of the natural language traffic regulation comprises: searching all verbs in the target natural language traffic regulation, and analyzing whether there is an ambiguous time sequence relationship between verbs; if there is an ambiguous time sequence relationship between verbs, determining the verbs and corresponding modifiers with the ambiguous time sequence relationship as a fuzzy time relationship field; determining the time interval between actions in the fuzzy time relationship field according to the scene information in the natural language traffic regulation; determining the ambiguous time relationship between actions of traffic participants according to the fuzzy time relationship field and the time interval; and, searching all position relationship related nouns and corresponding modifiers in the target natural language traffic regulation, and analyzing whether any of the nouns and the corresponding modifiers is an ambiguous position relationship; if any of the nouns and the corresponding modifiers is an ambiguous position relationship, determining the nouns and the corresponding modifiers as a fuzzy position relationship field; determining the distance interval of the position relationship in the fuzzy position relationship field according to the scene information in the natural language traffic regulation; determining the ambiguous position relationship between traffic participants according to the fuzzy position relationship field and the distance interval; formalizing each main component in the ambiguous relationship of traffic participants in the natural language traffic regulation based on a logical language; wherein, the logical language is linear temporal logic; combining the formalized main components into a complete machine language traffic regulation according to logical operators and temporal operators.

2. The method of claim 1, wherein, The method of acquiring a target natural language traffic regulation comprises: acquiring a natural language traffic regulation; determining a target natural language traffic regulation in the natural language traffic regulation based on a screening rule for screening the state and behavior of motor vehicles.

3. The method of claim 1, wherein, The time interval is the sum of a scene required time interval and a time allowance; when the scene information contains a preset emergency word, the scene required time interval is sub-second level, and the time allowance is ten milliseconds level; when the scene information does not contain a preset emergency word, the scene required time interval is second level, and the time allowance is sub-second level.

4. The method of claim 1, wherein, The distance interval is the sum of a scene required distance interval and a distance allowance; the distance interval comprises longitudinal safety distance, lateral safety distance and comprehensive safety distance; when the distance interval is longitudinal safety distance, the scene required distance interval is at least one of time distance, absolute distance, TTC distance or STD distance; when the distance interval is lateral safety distance, the scene required distance interval is at least one of time distance, absolute distance, TTC distance or STD distance. When the distance interval is a comprehensive safety distance, the distance interval is decomposed into two directions of lateral and longitudinal, and the scenario requires the distance interval to be at least one of time distance, absolute distance, TTC distance or STD distance.

5. The method of claim 1, wherein, The logic operator includes at least one of negation, conjunction, disjunction, implication and equivalence; The timing operator includes at least one of next time, until, always, finally, always within the time period and within the time period.

6. A traffic rule to machine language conversion system characterized by, Comprise: The acquisition module, the fuzzy relationship determination module, the formalization module and the synthesis module; The acquisition module is used for acquiring the target natural language intersection regulation; The target natural language intersection regulation includes scene information; The fuzzy relationship determination module is used for determining the fuzzy relationship of the traffic participants in the natural language intersection regulation according to the scene information of the natural language intersection regulation; the fuzzy relationship of the traffic participants includes a fuzzy time relationship between the actions of the traffic participants and a fuzzy position relationship between the traffic participants; specifically used for: searching all verbs in the target natural language intersection regulation, and analyzing whether there is an ambiguous time sequence relationship between the verbs and the verbs; if there is an ambiguous time sequence relationship between the verbs and the verbs, the verbs and the corresponding modifiers with the ambiguous time sequence relationship are determined as a fuzzy time relationship field; determining the time interval between the actions in the fuzzy time relationship field according to the scene information in the natural language intersection regulation; determining the fuzzy time relationship between the actions of the traffic participants according to the fuzzy time relationship field and the time interval; And, Searching all position relationship related nouns and corresponding modifiers in the target natural language intersection regulation, and analyzing whether any of the nouns and the corresponding modifiers is an ambiguous position relationship; if any of the nouns and the corresponding modifiers is an ambiguous position relationship, the nouns and the corresponding modifiers are determined as a fuzzy position relationship field; determining the distance interval of the position relationship in the fuzzy position relationship field according to the scene information in the natural language intersection regulation; determining the fuzzy position relationship between the traffic participants according to the fuzzy position relationship field and the distance interval; The formalization module is used for formalizing each main component in the fuzzy relationship of the traffic participants in the natural language intersection regulation based on a logic language; wherein the logic language is linear temporal logic; The synthesis module is used for combining the formalized main components into a complete machine language intersection regulation according to logic operators and timing operators.

Images