CN113486628A - Method and system for converting traffic rules into machine language - Google Patents
Method and system for converting traffic rules into machine language Download PDFInfo
- Publication number
- CN113486628A CN113486628A CN202110763241.1A CN202110763241A CN113486628A CN 113486628 A CN113486628 A CN 113486628A CN 202110763241 A CN202110763241 A CN 202110763241A CN 113486628 A CN113486628 A CN 113486628A
- Authority
- CN
- China
- Prior art keywords
- natural language
- fuzzy
- distance
- time
- relation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 39
- 239000003607 modifier Substances 0.000 claims description 26
- 238000012216 screening Methods 0.000 claims description 10
- 230000015572 biosynthetic process Effects 0.000 claims description 6
- 238000003786 synthesis reaction Methods 0.000 claims description 6
- 230000002123 temporal effect Effects 0.000 claims description 5
- 238000006243 chemical reaction Methods 0.000 abstract description 8
- 238000012360 testing method Methods 0.000 description 11
- 230000006399 behavior Effects 0.000 description 9
- 210000004556 brain Anatomy 0.000 description 6
- 230000006870 function Effects 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000013461 design Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000033001 locomotion Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000006467 substitution reaction Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 230000008447 perception Effects 0.000 description 2
- 238000013024 troubleshooting Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011056 performance test Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/12—Use of codes for handling textual entities
- G06F40/151—Transformation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Artificial Intelligence (AREA)
- Tourism & Hospitality (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention relates to a method and a system for converting traffic rules into machine language, belonging to the technical field of cross rule language conversion, wherein the method comprises the steps of obtaining a target natural language cross rule; determining a fuzzy relation of traffic participants in the natural language traffic rule according to scene information of the natural language traffic rule; based on the logic language, formalizing all main components in the fuzzy relation of the traffic participants in the natural language intersection; and combining the formalized main components into a complete machine language intersection according to the logic operator and the time sequence operator. By the combined application of the fuzzy time relation, the fuzzy position relation pre-logic operator and the time sequence operator, the complete conversion from the natural language intersection to the machine language intersection is realized, so that the machine can be intuitively understood, most intersection rules can be applied, simple scenes and complex scenes can be applied, and the method has good applicability.
Description
Technical Field
The invention belongs to the technical field of traffic rule language conversion, and particularly relates to a method and a system for converting traffic rules into machine language.
Background
With the development of society, more and more machines in modern life develop towards automation, intellectuality, and the car that the removal was used for the trip is no exception, and intelligent car is gradually getting into people's daily life. In recent years, ADAS (Advanced Driving assistance System) plays an important role in smart vehicles, and various sensors mounted on vehicles are used to sense the surrounding environment at any time during the Driving process of the vehicle, collect data, identify, detect and track stationary and moving objects, and combine with navigator map data to perform systematic calculation and analysis, so that drivers can detect possible dangers in advance, and the comfort and safety of vehicle Driving are effectively improved. It can be said that true autopilot is an extremely product of the development of ADAS.
The functions to be realized by automatic driving mainly comprise perception prediction, decision planning, vehicle control and the like. The sensing module is the 'eyes' of the automatic driving automobile and is used for sensing the environment, obstacles and the like; the prediction module predicts the track of the obstacle in a certain follow-up time according to the sensed and tracked obstacle information; the decision planning module is the brain of the automatic driving automobile, and generally realizes three functions including path planning, behavior decision, motion planning and the like. At present, the design and test of a decision planning module mainly focus on driving safety and partial simple traffic regulation legitimacy, such as no collision with obstacles, following traffic lights and road signs, no overspeed and the like, and far from enough traffic regulations are not covered; and the first main problem of the automatic driving system found in the closed test field is 'intersection compliance is not sound', so whether in the design stage of a planning module or in the test stage of an automatic driving vehicle or a regulation module, the test of whether the automatic driving vehicle complies with the intersection is an extremely important ring.
However, the current traffic regulations are described by natural language and are oriented to human drivers, natural language intersection rules often include fuzzy semantics, inspection for compliance with the intersection rules in the related art is mostly based on human judgment, and machines are difficult to directly understand. For example, "a vehicle enters a highway from a ramp and should enter the lane without preventing the vehicle already in the highway from normally traveling," the machine has difficulty in understanding what is called "without preventing other vehicles from normally traveling". Therefore, how to convert the existing intersection rules described by natural language into machine language intersection rules understood by machines becomes a technical problem to be solved urgently in the prior art.
Disclosure of Invention
The invention provides a method and a system for converting traffic rules into machine language, which aim to solve the technical problem that a machine in the prior art cannot understand a natural language fuzzy intersection rule easily.
The technical scheme provided by the invention is as follows:
in one aspect, a method for converting traffic rules into machine language includes:
acquiring a target natural language intersecting rule; the target natural language intersecting rule comprises scene information;
determining a fuzzy relation of traffic participants in the natural language traffic rule according to scene information of the natural language traffic rule; the fuzzy relation of the traffic participants comprises a fuzzy time relation between the actions of the traffic participants and a fuzzy position relation between the traffic participants;
formalizing, based on a logical language, principal components in the fuzzy relationship of the traffic participants in the natural language intersection;
and combining the formalized main components into a complete machine language intersection according to the logic operator and the time sequence operator.
Optionally, the obtaining the target natural language intersection comprises:
acquiring a natural language cross rule;
and determining a target natural language intersection in the natural language intersection based on a screening rule for screening and restricting the state and the behavior of the motor vehicle.
Optionally, the determining the fuzzy relationship of the traffic participants in the natural language intersection according to the scene information of the natural language intersection includes:
searching all verbs in the target natural language cross rule, and analyzing whether an ambiguous time precedence relationship exists between the verbs;
if an ambiguous time sequence relation exists between the verb and the verb, determining that the verb with the ambiguous time sequence relation and the corresponding modifier are fuzzy time relation fields;
determining the time interval between actions in the fuzzy time relation field according to scene information in the natural language intersection rule;
and determining the fuzzy time relation between the actions of the traffic participants according to the fuzzy time relation field and the time interval.
Optionally, the time interval is a sum of a scene required time interval and a time margin;
when the scene information contains preset emergency words, the scene requires a time interval of a sub-second level, and the time allowance is a ten-millisecond level;
when the scene information does not contain the preset emergency words, the scene requires time intervals of a second level, and the time allowance is of a sub-second level.
Optionally, the determining the fuzzy relationship of the traffic participants in the natural language intersection according to the scene information of the natural language intersection includes:
searching all nouns and corresponding modifiers related to the position relationship in the target natural language intersection, and analyzing whether any noun and corresponding modifier are in an ambiguous position relationship or not;
if any noun and the corresponding modifier are in uncertain position relation, determining the noun and the corresponding modifier as a fuzzy position relation field;
determining the distance interval of the position relation in the fuzzy position relation field according to the scene information in the natural language intersection rule;
and determining the fuzzy position relation between the traffic participants according to the fuzzy position relation field and the distance interval.
Optionally, the distance interval is a sum of a distance interval required by the scene and a distance margin;
the distance interval comprises a longitudinal safety distance, a transverse safety distance and a comprehensive safety distance;
when the distance interval is a longitudinal safe distance, the scene requires the distance interval to be at least one of a time distance, an absolute distance, a TTC distance or an STD distance;
when the distance interval is a lateral safe distance, the scene requires the distance interval to be at least one of a time distance, an absolute distance, a TTC distance, or a STD distance;
when the distance interval is an integrated safe distance, the distance interval is divided into two directions of a transverse direction and a longitudinal direction, and the scene requirement distance interval is at least one of a time distance, an absolute distance, a TTC distance or an STD distance.
Optionally, the logical operator includes: at least one of negation, conjunctive, disjunctive, derivational, and equivalents;
the temporal operator includes: at least one of next time, until, always, finally, always occurring within the time period, and occurring within the time period.
In yet another aspect, a system for converting traffic rules to machine language, comprising: the device comprises an acquisition module, a fuzzy relation determination module, a formalization module and a synthesis module;
the acquisition module is used for acquiring the target natural language intersecting rule; the target natural language intersecting rule comprises scene information;
the fuzzy relation determining module is used for determining the fuzzy relation of the traffic participants in the natural language traffic regulations according to the scene information of the natural language traffic regulations; the fuzzy relation of the traffic participants comprises a fuzzy time relation between the actions of the traffic participants and a fuzzy position relation between the traffic participants;
the formalization module is used for formalizing each main component in the fuzzy relation of the traffic participants in the natural language intersection rule based on a logic language;
and the synthesis module is used for combining the formalized main components into a complete machine language intersection according to the logic operator and the time sequence operator.
The invention has the beneficial effects that:
the method and the system for converting the traffic rules into the machine language provided by the embodiment of the invention acquire the target natural language traffic rules; determining a fuzzy relation of traffic participants in the natural language traffic rule according to scene information of the natural language traffic rule; based on the logic language, formalizing all main components in the fuzzy relation of the traffic participants in the natural language intersection; and combining the formalized main components into a complete machine language intersection according to the logic operator and the time sequence operator. By the combined application of the fuzzy time relation, the fuzzy position relation pre-logic operator and the time sequence operator, the complete conversion from the natural language intersection to the machine language intersection is realized, so that the machine can be intuitively understood, most intersection rules can be applied, simple scenes and complex scenes can be applied, and the method has good applicability.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flow chart of a method for converting a traffic rule into a machine language according to an embodiment of the present invention;
FIG. 2 is a diagram illustrating an example of cross-formatting of a specific target natural language according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a system for converting traffic rules into machine language according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be described in detail below. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be derived by a person skilled in the art from the examples given herein without any inventive step, are within the scope of the present invention.
The decision planning module for automatic driving is the brain of an automatic driving automobile, and generally realizes three functions including path planning, behavior decision, motion planning and the like. The path planning refers to planning a path from an initial place to a destination according to the initial place and the destination and by combining information such as roads in a map; the behavior decision can make specific behavior decisions such as overtaking, lane changing and the like according to the result of the path planning and the information of the current environment, obstacles and the like; the motion planning can be used for planning a track point meeting certain constraint conditions, such as the track point in overtaking cars, according to the result of the behavior decision.
Autonomous vehicles are required to pass a variety of tests, mainly focusing on closed or open road tests at the vehicle level or compliance and performance tests at the component level. The decision planning module as a vehicle-mounted brain is very important for the automatic driving automobile, and the correctness of the output of the decision planning module directly influences the safety of the automatic driving automobile, so that the decision planning module needs to follow some safety, legality and other criteria no matter in a design stage or a test stage. At present, the design and test of a decision planning module mainly focus on driving safety and partial simple traffic regulation legitimacy, such as no collision with barriers, following of traffic lights and road signs, no overspeed and the like, and far no enough traffic regulations are covered; and the first main problem of the automatic driving system found in the closed test field is 'intersection compliance is not sound', so whether in the design stage of a planning module or in the test stage of an automatic driving whole vehicle or a regulation module, the test of whether the automatic driving vehicle complies with the intersection is an extremely important ring.
However, the current traffic regulations are described by natural language and are oriented to human drivers, natural language intersection rules often include fuzzy semantics, inspection for compliance with the intersection rules in the related art is mostly based on human judgment, and machines are difficult to directly understand. For example, "a vehicle enters a highway from a ramp and should enter the lane without preventing the vehicle already in the highway from normally traveling," the machine has difficulty in understanding what is called "without preventing other vehicles from normally traveling". Therefore, how to formulate a set of machine language interaction rules convenient for machine understanding based on the existing interaction rules described by natural language becomes a technical problem to be solved urgently in the prior art.
Based on the above, the embodiment of the invention provides a method for converting traffic rules into machine language.
Fig. 1 is a schematic flow chart of a method for converting a traffic rule into a machine language according to an embodiment of the present invention, and referring to fig. 1, the method according to the embodiment of the present invention may include the following steps:
s11, acquiring a target natural language intersecting rule; the target natural language intersection includes scene information.
In a specific implementation process, the method for converting the traffic rule into the machine language provided by the application can be applied to any system for converting the natural language into the machine language, and particularly can be applied to a system for converting the traffic rule into the machine language provided by the application; the method for converting the traffic rule into the machine language provided by the embodiment of the invention can be applied to an automatic driving simulation system or an automatic driving automobile system, so that the traffic rule to be complied with by an automatic driving decision planning module can be formalized (digitalized), and the invention is not particularly limited.
For example, a target natural language traffic rule, i.e., a traffic rule described in a natural language, may be obtained first.
In some embodiments, optionally, obtaining the target natural language intersection comprises: acquiring a natural language cross rule; and determining a target natural language intersection in the natural language intersection based on a screening rule for screening and restricting the state and the behavior of the motor vehicle.
For example, the natural language traffic rules can be acquired and the items to be formalized can be screened out by combing one by one according to the traffic rules. In the screening process, screening can be carried out according to the principle of restricting the state and the behavior of the motor vehicle, and the screened natural language traffic regulations are target natural language traffic regulations. The target natural language intersection is not limited herein, but is described in an embodiment, for example, the target natural language intersection may be: "when the motor vehicle breaks down on the road and needs to stop to remove the fault, the driver should turn on the hazard warning flash lamp immediately". The scene information may also be "the vehicle has a fault on the road, and when the vehicle needs to be stopped to remove the fault, the driver should immediately turn on the hazard warning flash lamp", or may also be part of the information, and is not specifically limited herein.
And S12, determining the fuzzy relation of the traffic participants in the natural language traffic regulations according to the scene information of the natural language traffic regulations.
In this embodiment, the fuzzy relation of the traffic participants includes a fuzzy time relation between the actions of the traffic participants and a fuzzy position relation between the traffic participants.
When the fuzzy relationship of the traffic participant is a fuzzy time relationship between the motions of the traffic participant, in some embodiments, optionally, determining the fuzzy relationship of the traffic participant in the natural language intersection according to the scene information of the natural language intersection includes: searching all verbs in the target natural language cross rule, and analyzing whether an ambiguous time precedence relationship exists between the verbs; if an ambiguous time sequence relation exists between the verb and the verb, determining that the verb with the ambiguous time sequence relation and the corresponding modifier are fuzzy time relation fields; determining the time interval between actions in the fuzzy time relation field according to scene information in the natural language intersection; and determining the fuzzy time relation between the actions of the traffic participants according to the fuzzy time relation field and the time interval.
For example, according to the existing search technology, all verbs are searched in the target natural language cross rule, and whether an ambiguous time-sequence relationship exists between the verbs or not is analyzed to form a candidate fuzzy time relationship field set. For example, an ambiguous chronological relationship may be that there is no specific time interval.
In this embodiment, the fuzzy time relationship field may be determined in a tag manner. And when the indefinite time precedence relationship exists between the verbs, marking the verb with the occurrence relationship and the related modifier thereof as a fuzzy time relationship field. For example, "when a motor vehicle has a fault on a road and needs to be stopped to remove the fault, a driver should immediately turn on a hazard warning flash lamp", wherein, "occurrence" and "turn on" are used as verbs, and the actions described in the verbs include that the motor vehicle has a fault on the road and the hazard warning flash lamp is turned on, and no specific time interval is clear, so that the "fault occurrence" and the "should be immediately turned on" are marked as candidate fuzzy time relation fields; otherwise, it is not marked.
In some embodiments, optionally, the time interval is a sum of the scene requirement time interval and a time margin; when the scene information contains preset emergency words, the scene requires a time interval of a sub-second level and a time margin of a ten-millisecond level; when the scene information does not contain the preset emergency words, the scene requires time intervals of a second level and the time allowance is of a sub-second level.
For example, the time interval between actions in the fuzzy time relationship field may be determined according to different scenes
Wherein, Δ tｓA time interval is required for a scene,
is the time margin.
In this embodiment, the preset emergency words may be: the present invention is not limited to the above embodiments, and the embodiments are described in detail with reference to the following examples. For example, for a scenario where a time interval such as "immediate" requires strict words, the scenario requires a time interval Δ tsIs of sub-second grade,
On the order of ten milliseconds; for scenes where such time intervals do not appear "immediate" require looser words, the scene requires a time interval Δ tsIs in the second class,
On the sub-second scale.
Based on the fuzzy time relationship field and the time interval, a fuzzy time relationship between the traffic participant actions may be determined. For example, the fuzzy time relationship between the actions of the traffic participants that the driver should turn on the hazard warning flash immediately when the motor vehicle breaks down on the road and needs to stop to remove the fault is as follows: the fuzzy time relationship fields are "failed" and "should be turned on immediately" with a time interval of 0.5 seconds. The calculation method of 0.5 second is explained below: the scene information may be "when the motor vehicle has a fault on the road and needs to be stopped to remove the fault, the driver should turn on the hazard warning flash immediately", where "immediate" is included, and the scene requirement time interval may be in the sub-second level, the time margin is in the millisecond level, and the sum is 0.5 seconds, where 0.5 seconds is merely an example and is not a limitation.
When the fuzzy relationship of the traffic participant is a fuzzy position relationship between the traffic participants, in some embodiments, optionally, determining the fuzzy relationship of the traffic participant in the natural language intersection according to the scene information of the natural language intersection includes: searching all nouns and corresponding modifiers related to the position relationship in the target natural language intersection, and analyzing whether any noun and corresponding modifier are in an ambiguous position relationship; if any noun and the corresponding modifier are in uncertain position relation, determining the noun and the corresponding modifier as a fuzzy position relation field; determining the distance interval of the position relation in the fuzzy position relation field according to scene information in the natural language intersection rule; and determining the fuzzy position relation among the traffic participants according to the fuzzy position relation field and the distance interval.
For example, according to the existing search technology, all nouns and modifiers related to the positional relationship are searched in the target natural language cross-reference, and whether the nouns and modifiers are ambiguous positional relationships or not is analyzed to form a set of candidate ambiguous positional relationship fields. For example, the ambiguous positional relationship may be a numerical value or a mathematical model having no specific positional relationship.
In this embodiment, the fuzzy positional relationship field may be determined by means of a flag. When the noun and the modifier thereof related to the position relation describe the ambiguous position relation, the noun and the modifier thereof related to the position relation are marked as the ambiguous position relation field. For example, "the rear vehicle should overtake the left side of the front vehicle after confirming that there is a sufficient safe distance," wherein the "sufficient safe distance" is a noun related to the positional relationship and a modifier thereof, but there is no specific positional relationship to explain what is the sufficient safe distance which is an ambiguous positional relationship, and therefore, it is marked as a candidate ambiguous positional relationship field; otherwise, it is not marked.
In some embodiments, optionally, the distance interval is a sum of the scene requirement distance interval and a distance margin; the distance interval comprises a longitudinal safety distance and/or a transverse safety distance; when the distance interval is a longitudinal safe distance, the scene requires the distance interval to be at least one of a time distance, an absolute distance, a TTC distance or an STD distance; when the distance interval is a transverse safe distance, the scene requires the distance interval to be at least one of a time distance, an absolute distance, a TTC distance or an STD distance; when the distance interval is the comprehensive safe distance, the distance interval is divided into the transverse direction and the longitudinal direction, and the scene requires the distance interval to be at least one of the time distance, the absolute distance, the TTC distance or the STD distance.
Wherein, the TTC distance is a TTC (time To precision) distance measurement algorithm; the TTC distance and the STD distance are both in the prior art, and are not described herein.
For example, the distance interval of the candidate fuzzy position relation is determined according to different scenes
Wherein, depending on the scene, Δ dsA distance separation is required for the scene,
is the distance margin.
In this embodiment, for the longitudinal safe distance case: Δ dsThe adopted model can be time distance, absolute distance, TTC distance, STD distance or a combined model of various models, and the like.
For the lateral safe distance case: Δ dsThe adopted model can be time distance, absolute distance, TTC distance, STD distance or a combined model of various models, and the like.
For the case where neither lateral nor longitudinal safety distance is specified: the safe distance in both directions is considered, and the delta d can be decomposed into two directions of transverse direction and longitudinal direction, and the corresponding delta dsAnd
the same applies to both the transverse and longitudinal directions. Δ dsThe adopted model can be time distance, absolute distance, TTC distance, STD distance or a combined model of various models, and the like.
For close range scenes
In centimeter level, for long-distance scene
Is in decimeter level.
Based on the ambiguous positional relationship field and the distance interval, ambiguous positional relationships between the traffic participants may be determined. For example, the ambiguous positional relationship between traffic participants for "a rear car should overrun from the left side of a front car after confirming that there is a sufficient safe distance" is: the ambiguous location relationship field is "sufficient safe distance" with a distance interval of 10 meters. Wherein, 10 meters is calculated according to the distance interval and the distance margin of the specific scene, and the calculation process is simple addition, subtraction, multiplication, division and operation, which is not described herein.
And S13, based on the logic language, formalizing each main component in the fuzzy relation of the traffic participants in the natural language intersection.
After the fuzzy relation of the specific traffic participants is obtained, the clear natural language intersecting standard can be obtained, and main components in the clear natural language intersecting standard are formalized according to the logic language. The Logic language may be Linear Temporal Logic (LTL).
Fig. 2 is a schematic diagram of an example of cross-formatting a specific target natural language according to an embodiment of the present invention.
For example, referring to fig. 2, the target natural language cross rule is "when the vehicle has a fault on the road and needs to be stopped for troubleshooting, the hazard warning flash lamp should be turned on immediately to execute the minimum risk braking", and the clear natural language cross rule is "when the vehicle has a fault and needs to be stopped for troubleshooting, the hazard warning flash lamp should be turned on within 0.5 second and the vehicle should be stopped within 2 seconds". Formalizing the clear natural language cross rule by using onRoad, stBreakDown, stSignalEmergency and stStop as propositions after formalizing each main part in the clear natural language cross rule. It should be noted that the present invention is merely illustrative in form and not restrictive.
And S14, combining the formalized main components into a complete machine language cross-standard according to the logic operator and the time sequence operator.
In some embodiments, optionally, the logical operator, comprises: at least one of negation, conjunctive, disjunctive, derivational, and equivalents; the time sequence operator comprises: at least one of next time, until, always, finally, always occurring within the time period, and occurring within the time period.
For example, table 1 is a logic operator summary table provided in the embodiment of the present invention, and table 2 is a time sequence operator summary table provided in the embodiment of the present invention.
TABLE 1 summary of logical operators
TABLE 2 summary of time-series operators
| Time sequence operator | Means of |
| X | The next moment (next) |
| U | Up to (until) |
| G | Always (always, global) |
| F | Finally (eventualy, future) |
| G(t1，t2) | At a time from t1～t2Always in |
| F(t1，t2) | At a time from t1～t2Can happen internally |
And combining the formalized main components according to the logic operator and the time sequence operator to form the complete machine language cross-standard.
For example, referring to FIG. 2, the formalized principal components may be combined into onRoad Λ stBreakDown → F using a logical operator and a temporal operator[0，0.5]stSlgnalEmergency∧F[0，2]stStop。
Referring to fig. 2, after the natural language cross-standard is obtained in the first step, the fuzzy time relationship is determined in the manner provided by the embodiment, then each main composition of the natural language cross-standard is formalized by using the time sequence logic (such as LTL or MTL), and finally all the formalized compositions are combined into a complete formalized cross-standard, i.e. the machine language cross-standard, by using the logic operator (table 1) and the time sequence operator (table 2).
In some embodiments, when the formal intersection is subsequently applied, the propositions onRoad, stbreekdown, stsignalemployee, stStop may output true and false based on input data (e.g., vehicle trajectory information, environmental perception information, etc.), and then combine with the logical connection words to obtain true and false of the proposition of the overall formal intersection, thereby determining whether the auto-driven vehicle violates rules.
In the above-described embodiment, the subject of the conversion of the traffic rules into the machine language may be a traffic formalized person, or may be a machine, such as a computer dedicated to natural language traffic formalization or directly the "brain" of an autonomous vehicle. Only a series of steps of obtaining the cross rule, searching, clarifying the fuzzy time relation and the position relation and the like operated by the human are replaced by machine operation.
In this embodiment, the method for converting the traffic rule into the machine language provided by the present application may be used online, or may be used offline, and this embodiment is not particularly limited. For example, the conversion from the traffic rule to the machine language is performed on the existing traffic rule, namely, the traffic rule is in an off-line traffic rule formalization; for an automatically-driven automobile running on the road, if the updated natural language traffic regulations are transmitted to a vehicle-mounted brain through wireless signals or the rules are recorded on a sign in a text form, the rules described by the characters are recognized by the automatically-driven automobile through a vehicle-mounted sensor such as a camera, and the vehicle-mounted brain performs the formalization operation on the traffic regulations, namely online real-time traffic regulations formalization.
The method for converting the traffic rule into the machine language provided by the embodiment of the invention comprises the steps of obtaining a target natural language intersection rule; determining a fuzzy relation of traffic participants in the natural language traffic rule according to scene information of the natural language traffic rule; based on the logic language, formalizing all main components in the fuzzy relation of the traffic participants in the natural language intersection; and combining the formalized main components into a complete machine language intersection according to the logic operator and the time sequence operator. By the combined application of the fuzzy time relation, the fuzzy position relation pre-logic operator and the time sequence operator, the complete conversion from the natural language intersection to the machine language intersection is realized, so that the machine can be intuitively understood, most intersection rules can be applied, simple scenes and complex scenes can be applied, and the method has good applicability.
Based on a general inventive concept, the embodiment of the invention also provides a system for converting the traffic rules into the machine language.
Fig. 3 is a schematic structural diagram of a system for converting a traffic rule into a machine language according to an embodiment of the present invention, and referring to fig. 3, the system according to an embodiment of the present invention may include: an acquisition module 31, a fuzzy relation determination module 32, a formalization module 33 and a synthesis module 34.
The acquiring module 31 is configured to acquire a target natural language intersecting standard; the target natural language intersecting rule comprises scene information;
the fuzzy relation determining module 32 is configured to determine a fuzzy relation between traffic participants in the natural language traffic regulations according to the scene information of the natural language traffic regulations; the fuzzy relation of the traffic participants comprises a fuzzy time relation between the actions of the traffic participants and a fuzzy position relation between the traffic participants;
a formalization module 33, configured to formalize, based on the logic language, each main component in the fuzzy relationship of the traffic participants in the natural language intersection;
and the synthesis module 34 is used for combining the formalized main components into a complete machine language cross-standard according to the logic operator and the time sequence operator.
In some embodiments, optionally, the obtaining module 31 is configured to obtain a natural language intersection; and determining a target natural language intersection in the natural language intersection based on a screening rule for screening and restricting the state and the behavior of the motor vehicle.
In some embodiments, optionally, the fuzzy relation determining module 32 is configured to search all verbs in the target natural language cross rule, and analyze whether there is an ambiguous temporal precedence relation between the verbs;
if an ambiguous time sequence relation exists between the verb and the verb, determining that the verb with the ambiguous time sequence relation and the corresponding modifier are fuzzy time relation fields;
determining the time interval between actions in the fuzzy time relation field according to scene information in the natural language intersection;
and determining the fuzzy time relation between the actions of the traffic participants according to the fuzzy time relation field and the time interval.
In some embodiments, optionally, the fuzzy relation determining module 32 is configured to search all nouns and corresponding modifiers related to the position relation in the target natural language intersection, and analyze whether any noun and corresponding modifier are in an ambiguous position relation;
if any noun and the corresponding modifier are in uncertain position relation, determining the noun and the corresponding modifier as a fuzzy position relation field;
determining the distance interval of the position relation in the fuzzy position relation field according to scene information in the natural language intersection rule;
and determining the fuzzy position relation among the traffic participants according to the fuzzy position relation field and the distance interval.
With regard to the system in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.
The system for converting the traffic rules into the machine language provided by the embodiment of the invention obtains the target natural language traffic rules; determining a fuzzy relation of traffic participants in the natural language traffic rule according to scene information of the natural language traffic rule; based on the logic language, formalizing all main components in the fuzzy relation of the traffic participants in the natural language intersection; and combining the formalized main components into a complete machine language intersection according to the logic operator and the time sequence operator. By the combined application of the fuzzy time relation, the fuzzy position relation pre-logic operator and the time sequence operator, the complete conversion from the natural language intersection to the machine language intersection is realized, so that the machine can be intuitively understood, most intersection rules can be applied, simple scenes and complex scenes can be applied, and the method has good applicability.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.
It is understood that the same or similar parts in the above embodiments may be mutually referred to, and the same or similar parts in other embodiments may be referred to for the content which is not described in detail in some embodiments.
It should be noted that the terms "first," "second," and the like in the description of the present invention are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. Further, in the description of the present invention, the meaning of "a plurality" means at least two unless otherwise specified.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and alternate implementations are included within the scope of the preferred embodiment of the present invention in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present invention.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present invention may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.
Claims (8)
1. A method for converting traffic rules into machine language, comprising:
acquiring a target natural language intersecting rule; the target natural language intersecting rule comprises scene information;
determining a fuzzy relation of traffic participants in the natural language traffic rule according to scene information of the natural language traffic rule; the fuzzy relation of the traffic participants comprises a fuzzy time relation between the actions of the traffic participants and a fuzzy position relation between the traffic participants;
formalizing, based on a logical language, principal components in the fuzzy relationship of the traffic participants in the natural language intersection;
and combining the formalized main components into a complete machine language intersection according to the logic operator and the time sequence operator.
2. The method of claim 1, wherein obtaining the target natural language intersection comprises:
acquiring a natural language cross rule;
and determining a target natural language intersection in the natural language intersection based on a screening rule for screening and restricting the state and the behavior of the motor vehicle.
3. The method of claim 1, wherein determining fuzzy relationships of traffic participants in the natural language intersection according to the scene information of the natural language intersection comprises:
searching all verbs in the target natural language cross rule, and analyzing whether an ambiguous time precedence relationship exists between the verbs;
if an ambiguous time sequence relation exists between the verb and the verb, determining that the verb with the ambiguous time sequence relation and the corresponding modifier are fuzzy time relation fields;
determining the time interval between actions in the fuzzy time relation field according to scene information in the natural language intersection rule;
and determining the fuzzy time relation between the actions of the traffic participants according to the fuzzy time relation field and the time interval.
4. The method of claim 3, wherein the time interval is a sum of a scene requirement time interval and a time margin;
when the scene information contains preset emergency words, the scene requires a time interval of a sub-second level, and the time allowance is a ten-millisecond level;
when the scene information does not contain the preset emergency words, the scene requires time intervals of a second level, and the time allowance is of a sub-second level.
5. The method of claim 1, wherein determining fuzzy relationships of traffic participants in the natural language intersection according to the scene information of the natural language intersection comprises:
searching all nouns and corresponding modifiers related to the position relationship in the target natural language intersection, and analyzing whether any noun and corresponding modifier are in an ambiguous position relationship or not;
if any noun and the corresponding modifier are in uncertain position relation, determining the noun and the corresponding modifier as a fuzzy position relation field;
determining the distance interval of the position relation in the fuzzy position relation field according to the scene information in the natural language intersection rule;
and determining the fuzzy position relation between the traffic participants according to the fuzzy position relation field and the distance interval.
6. The method of claim 5, wherein the distance interval is a sum of a scene requirement distance interval and a distance margin;
the distance interval comprises a longitudinal safety distance, a transverse safety distance and a comprehensive safety distance;
when the distance interval is a longitudinal safe distance, the scene requires the distance interval to be at least one of a time distance, an absolute distance, a TTC distance or an STD distance;
when the distance interval is a lateral safe distance, the scene requires the distance interval to be at least one of a time distance, an absolute distance, a TTC distance, or a STD distance;
when the distance interval is an integrated safe distance, the distance interval is divided into two directions of a transverse direction and a longitudinal direction, and the scene requirement distance interval is at least one of a time distance, an absolute distance, a TTC distance or an STD distance.
7. The method of claim 1, wherein the logical operator comprises: at least one of negation, conjunctive, disjunctive, derivational, and equivalents;
the temporal operator includes: at least one of next time, until, always, finally, always occurring within the time period, and occurring within the time period.
8. A system for converting traffic rules into machine language, comprising: the device comprises an acquisition module, a fuzzy relation determination module, a formalization module and a synthesis module;
the acquisition module is used for acquiring the target natural language intersecting rule; the target natural language intersecting rule comprises scene information;
the fuzzy relation determining module is used for determining the fuzzy relation of the traffic participants in the natural language traffic regulations according to the scene information of the natural language traffic regulations; the fuzzy relation of the traffic participants comprises a fuzzy time relation between the actions of the traffic participants and a fuzzy position relation between the traffic participants;
the formalization module is used for formalizing each main component in the fuzzy relation of the traffic participants in the natural language intersection rule based on a logic language;
and the synthesis module is used for combining the formalized main components into a complete machine language intersection according to the logic operator and the time sequence operator.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110763241.1A CN113486628A (en) | 2021-07-06 | 2021-07-06 | Method and system for converting traffic rules into machine language |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110763241.1A CN113486628A (en) | 2021-07-06 | 2021-07-06 | Method and system for converting traffic rules into machine language |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113486628A true CN113486628A (en) | 2021-10-08 |
Family
ID=77940704
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110763241.1A Pending CN113486628A (en) | 2021-07-06 | 2021-07-06 | Method and system for converting traffic rules into machine language |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113486628A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP4332824A1 (en) * | 2022-09-02 | 2024-03-06 | Continental Automotive Technologies GmbH | System and method for translating natural language traffic rules into formal logic for autonomous moving vehicles |
-
2021
- 2021-07-06 CN CN202110763241.1A patent/CN113486628A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP4332824A1 (en) * | 2022-09-02 | 2024-03-06 | Continental Automotive Technologies GmbH | System and method for translating natural language traffic rules into formal logic for autonomous moving vehicles |
| WO2024046787A1 (en) * | 2022-09-02 | 2024-03-07 | Continental Automotive Technologies GmbH | System and method for translating natural language traffic rules into formal logic for autonomous moving vehicles |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112781887B (en) | Method, device and system for testing vehicle performance | |
| US20170132117A1 (en) | Method and device for generating test cases for autonomous vehicles | |
| CN111382768A (en) | Multi-sensor data fusion method and device | |
| JP7119365B2 (en) | Driving behavior data generator, driving behavior database | |
| CN104859662A (en) | Fault handling in an autonomous vehicle | |
| Alozi et al. | Evaluating the safety of autonomous vehicle–pedestrian interactions: An extreme value theory approach | |
| Elgharbawy et al. | Adaptive functional testing for autonomous trucks | |
| Wang et al. | Overcoming challenges of validation automated driving and identification of critical scenarios | |
| CN113895456A (en) | Intersection driving method and device for automatic driving vehicle, vehicle and medium | |
| Nakamura et al. | Defining reasonably foreseeable parameter ranges using real-world traffic data for scenario-based safety assessment of automated vehicles | |
| Zlocki et al. | Logical scenarios parameterization for automated vehicle safety assessment: Comparison of deceleration and cut-in scenarios from Japanese and German highways | |
| Philipp et al. | Systematization of relevant road users for the evaluation of autonomous vehicle perception | |
| CN113486628A (en) | Method and system for converting traffic rules into machine language | |
| CN114475656A (en) | Travel track prediction method, travel track prediction device, electronic device, and storage medium | |
| CN114647954A (en) | Simulation scene generation method and device, computer equipment and storage medium | |
| Rösch et al. | Space, Time, and Interaction: A Taxonomy of Corner Cases in Trajectory Datasets for Automated Driving | |
| Akagi | Ontology based collection and analysis of traffic event data for developing intelligent vehicles | |
| CN113538909A (en) | Traffic incident prediction method and device for automatic driving vehicle | |
| Bäumler et al. | Report on validation of the stochastic traffic simulation (Part B) | |
| CN115169477A (en) | Method, device and equipment for evaluating safety sense of assistant driving system and storage medium | |
| CN115112138A (en) | Trajectory planning information generation method and device, electronic equipment and storage medium | |
| CN115269371A (en) | Platform for path planning system development of an autonomous driving system | |
| CN115071747A (en) | Driving assistance apparatus, system, and method for vehicle | |
| Yu | Method for quantitative evaluation of traffic complexity on the highway | |
| Obuhuma et al. | A software agent for vehicle driver modeling |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |