CN112329208A - Method for realizing modeling and verification of multi-clock-constrained collaborative unmanned system - Google Patents

Method for realizing modeling and verification of multi-clock-constrained collaborative unmanned system Download PDF

Info

Publication number
CN112329208A
CN112329208A CN202011107719.7A CN202011107719A CN112329208A CN 112329208 A CN112329208 A CN 112329208A CN 202011107719 A CN202011107719 A CN 202011107719A CN 112329208 A CN112329208 A CN 112329208A
Authority
CN
China
Prior art keywords
clock
model
logic
state
ccsl
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.)
Granted
Application number
CN202011107719.7A
Other languages
Chinese (zh)
Other versions
CN112329208B (en
Inventor
黄志球
王金永
谢健
王飞
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing University of Aeronautics and Astronautics
Original Assignee
Nanjing University of Aeronautics and Astronautics
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Nanjing University of Aeronautics and Astronautics filed Critical Nanjing University of Aeronautics and Astronautics
Priority to CN202011107719.7A priority Critical patent/CN112329208B/en
Publication of CN112329208A publication Critical patent/CN112329208A/en
Application granted granted Critical
Publication of CN112329208B publication Critical patent/CN112329208B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F8/00—Arrangements for software engineering
    • G06F8/20—Software design
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/04—Constraint-based CAD

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • Geometry (AREA)
  • Devices For Executing Special Programs (AREA)

Abstract

The invention discloses a method for realizing modeling and verification of a multi-clock constraint-oriented collaborative unmanned system, which comprises the following steps: providing a framework of the collaborative unmanned system, establishing a field modeling summary file SysML Profile of the collaborative unmanned system, and extracting professional terms, relations and operations of the professional terms and relations to the collaborative unmanned system; defining a single-clock and multi-clock system, and giving definition and relation semantics related to the clocks; modeling and verifying the collaborative unmanned system according to a model driving system development process; the automatic overtaking method comprises the steps of designing an automatic overtaking case, providing natural language requirements of a system, constructing a state machine model of the system, converting the state machine model into a corresponding CCSL model, and verifying the state machine model on a special simulation platform TimeSquare of a clock constraint protocol language. The method is safe and effective, and the modeling and the verification are carried out simultaneously, so that the method can be popularized and applied to other safety critical modeling fields of the multi-clock constraint system.

Description

Method for realizing modeling and verification of multi-clock-constrained collaborative unmanned system
Technical Field
The invention relates to a modeling and verification implementation method of a multi-clock constraint system, in particular to a modeling and verification implementation method of a multi-clock constraint cooperative unmanned system.
Background
Modeling and verification of a multi-clock constraint system, particularly modeling and verification oriented to a cooperative unmanned system, relates to natural language requirement modeling, a model driving development method of software, a model conversion technology and verification of a software model.
Modeling and verification of multi-clock restraint systems, particularly for safety-critical unmanned systems, there are three main types of current academic research and enterprise applications: 1, directly adopting a formal modeling language to carry out modeling for natural language requirements. The formal modeling language needs to have a certain formal modeling theoretical basis and is not beneficial to the communication among system interest correlators; 2, modeling is performed by using a common system modeling language sysml (system modeling language). The method is a general modeling language for all systems, and for a collaborative unmanned system, a modeler and an analyst need to provide a modeling summary file in the field, so that the model can be reused conveniently, and the modeling and verification analysis time can be saved; and 3, system modeling and system verification are carried out separately. Most defects of the system are generated in the system modeling and designing stage, and for the system with safety concern, the verification of the system is required to be carried out at the system designing stage, so that the defects of the system and the dangers of the system during operation are avoided as much as possible.
Disclosure of Invention
The purpose of the invention is as follows: the invention aims to provide a method for realizing modeling and verification of a multi-clock-constraint cooperative unmanned system.
The technical scheme is as follows: the invention discloses a method for realizing modeling and verification of a multi-clock constraint-oriented collaborative unmanned system, which comprises the following steps of:
s1, providing a framework of the collaborative unmanned system, establishing a field modeling summary file SysML Profile of the collaborative unmanned system, and extracting professional terms, relations and behavior operations of the collaborative unmanned system;
s2, defining a single-clock and multi-clock system, and giving definition and relation semantics related to the clocks;
s3, modeling the collaborative unmanned system according to the development process of the model driving system of the system;
and S4, for the cooperative unmanned system, constructing a state machine model of the system according to the natural language requirement of the system, converting the state machine model into a corresponding CCSL model, and verifying the state machine model on a special simulation platform TimeSquare of a clock constraint protocol language.
Further, the cooperative unmanned driving system in step S1 includes a cooperative sensing module, a brain-like driving module, and a vehicle control execution module, where the cooperative sensing module obtains information of surrounding vehicles, pedestrians, and roads through communication between a sensor and a vehicle, where the sensor includes a laser sensor, a radar sensor, and a camera, and the vehicle communication includes vehicle-to-vehicle communication, vehicle-to-road communication, and a global positioning system; the cooperative perception module transmits the acquired surrounding traffic environment information to a brain-like driving module, and the brain-like driving module comprises a scene understanding module and a behavior decision module; the scene understanding module carries out data fusion processing on the perception information obtained by the collaborative perception module and outputs information which can be processed by the brain-like driving module, the brain-like driving module invokes the behavior decision module according to the data fused by the scene understanding module, and the behavior decision module generates decision information; the behavior decision module transmits the decision information to the vehicle control execution module to generate corresponding vehicle control actions, and feeds back the behavior result to the surrounding traffic environment.
Further, the domain modeling summary file SysML Profile of the collaborative unmanned driving system in step S1 includes modeling terms, operations and relationships between modules of the system, the structural characteristics of the collaborative awareness module include sensor data, network communication, traffic snapshots and sensor tools, and the behavior characteristics of the obtained traffic information include vehicle-to-vehicle communication, vehicle-to-road communication and sensor awareness; the cooperative sensing module monitors the user and the traffic environment; the user structure characteristics comprise user identity identification, gender and age, and the user behavior characteristics comprise vehicle starting, starting point and end point information input and artificial control; the structural characteristics of the traffic environment comprise weather, road types, surrounding vehicles and surrounding pedestrians, and the behavior characteristics of the traffic environment comprise environment updating and data transmission; the cooperative sensing module triggers the scene understanding module, the cooperative sensing module is used for processing sensed user and environment information and outputting effective information which can be understood by a computer, the structural characteristics of the scene understanding module comprise traffic rules and prior traffic information, and the behavior characteristics of the scene understanding module comprise data fusion, traffic sign identification and observation; the output of the scene understanding module triggers a behavior decision module, a decision system generates behavior decision information of the vehicle by using the scene understanding information subjected to fusion processing, wherein the structural characteristics of the decision system comprise road application, road occupation, required acceleration, required speed, application withdrawal and occupation withdrawal, and the behavior characteristics of the decision system comprise road control, longitudinal control and transverse control; the input of the behavior decision information excites a vehicle control execution module, the vehicle control execution module controls the vehicle according to the input of the behavior decision information, the structural characteristics of the vehicle control execution module comprise space-time trajectory, advancing direction, transverse acceleration and emergency control, the behavior characteristics of the vehicle control execution module comprise brake control, steering control, accelerator control and gear control, and the vehicle control information is fed back to the user and the traffic environment, so that the user and the surrounding traffic environment can verify and control the vehicle execution information.
Further, in step S2, the single-clock system is defined as a main clock that needs to be established in the system, and the clock of each subsystem or sub-component must be referenced to the main clock, and samples are taken in the main clock to complete the clock system interacting with the overall system; the multi-clock system is defined as a clock system which has no global clock, each subsystem/sub-component obeys multiple clock frequencies, has an independent clock to control the action behavior of the subsystem/sub-component, and realizes the running of the global system through the partial order interaction relationship of the behavior among local subsystems.
Further, the clock constraint specification language CCSL in step S2 is a formal specification language based on a clock and clock constraints, the clock representing a trigger state of a signal at each discrete time point; the clock constraint describes a logic relationship between signal clocks, and the clock constraint includes a clock relationship and a clock expression, and specifically includes:
(1) logic clocks, clock scheduling and related definition of clock history;
logic clock: a logic clock c is defined as a sequence of finite or infinite clock jumps,
Figure BDA0002727539620000031
Figure BDA0002727539620000032
a set of non-0 natural numbers; for each one of ciE { tick, idle }, which indicates that the clock is in a trigger state tick or a non-trigger state idle in the state of the time i;
clock scheduling: a clock schedule sigma reflects the trigger state of all clocks at each moment in discrete time, and is a finite or infinite sequence
Figure BDA0002727539620000033
Figure BDA0002727539620000034
Is a set of 0 and positive integers; sigma (i) epsilon 2cRepresents a set of clocks that trigger at time i, formally defined as:
Figure BDA0002727539620000035
wherein C represents a set of finite logic clocks;
clock history: for a given clock schedule σ, the clock history records the number of triggers that each clock has expired to the current time in σ, formally defined as:
Figure BDA0002727539620000036
wherein Hσ(c, i) represents the number of triggers that a logic clock c expires to time i in clock schedule σ; hσ(c, i +1) stands for in clock scheduling σThe logic clock c is cut off to the triggering times of the time i + 1; hσ(c,0) represents the number of triggers for the logic clock c to expire at time 0 in clock schedule σ; i, j represents the time of the clock;
(2) clock relationship semantics;
the priority relationship is: logic clock c1Faster than the logic clock c2Triggering of (1);
Figure BDA0002727539620000037
wherein Hσ(c1I) represents the logic clock c in clock scheduling σ1The number of triggers until time i; hσ(c2I) represents the logic clock c in clock scheduling σ2The number of triggers until time i;
② mutual exclusion relationship: if logic clock c1Is triggered, then logic clock c2Not trigger and vice versa;
Figure BDA0002727539620000041
③ child-hour relationship: at any time, if the logic clock c1Triggered then logic clock c2Must be triggered;
Figure BDA0002727539620000042
fourthly, simultaneously relating: logic clock c1And logic clock c2Triggering at the same time;
Figure BDA0002727539620000043
a causal relationship: logic clock c1Is not slower than the logic clock c2Triggering of (1);
Figure BDA0002727539620000044
sixthly, the alternating relation is as follows: logic clock c1And logic clock c2Alternately triggering;
Figure BDA0002727539620000045
wherein, c1(i) Representing a logic clock c1At the i-th time, c2(i) Representing a logic clock c2At the i-th time, c1(i +1) represents a logic clock c1Time i + 1;
(3) a time-clock expression semantic;
combining clocks: triggering of clock and only if logic clock c1OR logic clock c2Triggering of (1);
Figure BDA0002727539620000046
clock crossing: triggering of clock crossing and only if logic clock c1And logic clock c2Simultaneously triggering;
Figure BDA0002727539620000047
clock delay: triggering and logic clock c of the clock2Are simultaneously triggered at c2Before, there is a logic clock c1And logic clock c2Trigger ratio logic clock c1D more triggering times;
Figure BDA0002727539620000048
clock period: triggering of this clock is with p logical clocks c1The triggering of (2) is periodic triggering;
Figure BDA0002727539620000051
wherein Hσ(c0I) represents the logic clock c in clock scheduling σ0The number of triggers until time i;
clock sampling: the clock is based on the clock and the logic clock c2For logic clock c1Sampling, in particular the toggling of the clock and the logic clock c2Triggered simultaneously at time i and preceded by a time j,0<j is less than or equal to i, the logic clock c of the moment1And at any time
Figure BDA0002727539620000052
Logic clock c2Not triggering;
Figure BDA0002727539620000053
sixthly, strictly sampling by a clock: the clock is based on the clock and the logic clock c2For logic clock c1Performing rigorous sampling, in particular, the triggering of the clock and the logic clock c2Triggered simultaneously at time i and preceded by a time j,0<j<i, the time logic clock c1And at any time
Figure BDA0002727539620000054
Logic clock c2Not triggering;
Figure BDA0002727539620000055
upper clock bound: the clock defines a logical clock c1And logic clock c2The fastest clock among the slow-triggering clocks;
Figure BDA0002727539620000056
clock lower bound: the clock defines a logical clock c1And logic clock c2The slowest clock among the triggered fast clocks;
Figure BDA0002727539620000057
further, the collaborative unmanned system model driven development process in step S3 specifically includes: firstly, analyzing system requirements in a natural language form by a system designer, extracting basic concepts of system design and relations among the concepts, and constructing a calculation irrelevant model; secondly, through model conversion, adding logic data related to application, and converting a calculation-independent model into a platform-independent model; then, converting the platform-independent model into a platform-dependent model by adding specific platform-dependent data; finally, converting the platform-related model into a system text code through conversion from the model to the text; in the process of model drive development of a software system, meta-models and corresponding summary files of a source model, a conversion model and a target model need to be constructed for conversion from a platform-independent model to a platform-dependent model, wherein the meta-models define the grammar and the semantics of a conversion language, and the summary files define the grammar and the semantics of a specific conversion language and take specific application platform and domain knowledge into consideration; the conversion model defines the conversion rule from the state machine diagram meta model to the formal clock constraint reduction language; the source model, the conversion model and the target model are respectively obeyed to the source meta-model, the conversion meta-model and the target meta-model, wherein the source model and the target model are respectively instantiated into a platform-independent model and a platform-dependent model, and model conversion from the platform-independent model to the platform-dependent model is completed.
Further, in step S3, coordinating formal definition of the meta model and the state machine diagram of the unmanned system;
(1) the meta-model of the collaborative unmanned system is a quadruple: AAM:: ═ BA,RA,EA,CA)；
①BARepresenting a system block definition diagram;
BA＝{Perception,Understanding,Planner,Actuator,User,Environment,KnowledgeBase}；
wherein a permission represents a Perception; underranging represents scene Understanding; planner represents behavioral decision-making; actuator denotes execution; user represents a User; environment represents an Environment; knowledgbasee represents a traffic knowledge base;
②RAthe relationship of system functions is defined;
Figure BDA0002727539620000061
wherein monitor represents monitoring; trigger represents trigger; precedence represents the precedence; stimulus stands for stimulation; response represents feedback; invoke represents a call;
③EArepresenting the system operating function, EA:BA×RA→BA；
④CAConstraint information, S, is definedA＝{Invariant,Precondition,Postcondition}；
Wherein Invariant represents an invariance; precondition represents a prerequisite; postcondition denotes a post condition;
(2) giving a formal definition of a state machine diagram;
the syntax for formalized representation of the state machine diagram is defined as the six-tuple SMD: (S, S)0,E,G,Act,Tr)；
S represents a state set in a state machine diagram;
②S0representing an initial set of states in a state machine diagram;
thirdly, E represents a set of trigger events in state transition;
g represents a set of satellite viewing conditions in state transition;
act represents a set of migration actions in state migration;
sixthly, Tr represents a set of migration relations in the state machine diagram,
Figure BDA0002727539620000071
wherein sv and tv are respectively a source state and a target state of a migration relationship, when a trigger event occurs and guard conditions are met, Tr executes a migration action to trigger the migration between the bodies in the state machine diagram;
the semantics of the formalized representation of the state machine diagram are as follows:
operating semantics in sequence:
Figure BDA0002727539620000072
wherein s issRepresenting the source state, stRepresenting a target state, e representing a trigger event in state transition, g representing a satellite condition in state transition, a representing a transition action in state transition, when the trigger event and the satellite condition are satisfied, the transition action is executed,
Figure BDA0002727539620000073
migration occurs;
selecting operation semantics: different migration conditions will cause different migration to occur;
Figure BDA0002727539620000074
wherein choice represents a selection operation, eiRepresents a trigger event in state transition i, giIndicating satellite Condition in State transition i, aiRepresenting a migration action in the state migration i;
Figure BDA0002727539620000075
wherein s iss1The first of the source states is represented,e1[g1]/a1,e2[g2]/a2andei[gi]/airepresenting different migration conditions, respectively representing trigger events, guard conditions and migration actions in the 1 st, 2 nd and i-th migration conditions, st1、st2And stiRespectively representing the 1 st, 2 nd and ith target states;
and thirdly, forking operation semantics: the bifurcation operation is that under the same migration condition, the migration target state is a composite state consisting of two or more orthogonal nodes;
Figure BDA0002727539620000076
wherein fork represents a fork operation, r1.s1int,r2.s2int,rn.snintRespectively representing target states of 1 st, 2 nd and ith orthogonal areas in the bifurcation operation;
Figure BDA0002727539620000081
wherein the content of the first and second substances,e[g]a trigger event and guard condition representing a migration condition;
fourthly, converging operation semantics: the merging operation is to be migrated to a target state by two or more mutually independent, concurrent and synchronous orthogonal source states;
join→(r1.s1fin|r2.s2fin|…|rn.snfin)×e×g×a×st；
where join represents a join operation, r1.s1fin,r2.s2finAnd rn.snfinRespectively representing initial states respectively belonging to the 1 st, 2 nd and ith orthogonal regions of the merging operation;
Figure BDA0002727539620000082
s33, defining a mapping rule between the meta-model and the state-based clock constraint specification language, and converting the formalized state machine diagram into a formalized clock constraint specification language CCSL according to the mapping rule;
the model conversion rules between the state machine graph meta-model and the clock constraint reduction language are as follows:
conversion between states: mapping the state in the state machine diagram to the state in the CCSL protocol;
Figure BDA0002727539620000083
wherein s issmdRepresents the state of the state machine diagram smd, SsmdDenotes ssmdSet of (1), sccslRepresenting states, S, in a clock constraint specification language, CCSLccslDenotes sccslA set of (a);
triggering the mapping between the event and the clock: mapping the trigger event in the state machine diagram to a logic clock in a CCSL protocol;
Figure BDA0002727539620000084
wherein E issmdRepresents a set of edges in the state machine diagram, λ (Tr) represents an action in the migration relationship in CCSL, TrccslA set representing migration relationships in the CCSL;
③ the mapping rule between guard condition and Boolean condition in the clock constraint reduction language: mapping guard conditions in a state machine to Boolean expressions in CCSL;
Figure BDA0002727539620000085
wherein G issmdRepresenting a set of guard conditions, Tr, in a state machine diagramchoice.BoolRepresents selection of Boolean condition, Tr, in CCSLccslRepresents Trchoice.BoolA set of (a);
fourthly, the mapping rule between the operation semantics of the sequential expression: mapping sequence expressions in a state machine diagram to Boolean sequence condition migration in a CCSL protocol;
Figure BDA0002727539620000091
wherein s is1、s2Representing two states, Tr, in a state machine diagramsmdRepresents a set s 'representing a migration relationship in CCSL'1、s’2Represents two states in the CCSL;
selecting a mapping rule between expression operation semantics: mapping a selection expression in a state machine diagram to Boolean selection condition migration in a CCSL protocol;
Figure BDA0002727539620000092
wherein S iss、StThe source state and the target state are respectively represented, and the CR.choice represents a set of selection operation states in a CCSL clock relationship;
sixthly, mapping rules among the operation semantics of the bifurcated expression are as follows: mapping a bifurcated expression in a state machine diagram to clock synchronization condition migration in a CCSL protocol;
Figure BDA0002727539620000093
wherein r is1,2,…,nRepresenting different orthogonal regions, s, in a state machine diagram1int,2int,…,nintRepresenting target states, s 'of the 1 st, 2 nd and nth orthogonal regions in a fork operation'sIs a source state in CCSL, tr'1,tr’2,tr’nRepresenting different target states after a fork operation in the CCSL;
seventhly, converging and selecting a mapping rule between expression operation semantics: mapping a confluent expression in a state machine diagram to conditional migration of a clock delay expression in a CCSL protocol;
Figure BDA0002727539620000101
wherein s is1fin,2fin,…,nfinIndicating a confluenceOperating initial states, s ', of the 1 st, 2 nd and i-th orthogonal regions'1、s’2、s’nRepresenting different initial states of the CCSL merge operation.
Has the advantages that: compared with the prior art, the invention has the following good technical effects:
(1) the invention provides a field modeling summary file SysML Profile of a collaborative unmanned system for the first time, and the summary file unifies field modeling terms, operations and relationships, thereby being beneficial to communication of field modeling personnel and saving modeling time.
(2) The invention converts the semi-formal model into the formal clock constraint reduction language, provides the meta-model of the state machine diagram, and defines the mapping rule between the meta-model and the state-based clock constraint language.
(3) The invention analyzes and verifies the multi-clock model in a model verification tool TimeSquare.
The method can be popularized and applied to other safety critical modeling fields of the multi-clock constraint system, and has wide application prospect.
Drawings
FIG. 1 is a schematic diagram of the general architecture of a collaborative unmanned system;
FIG. 2 is a schematic diagram of a collaborative unmanned driving domain modeling summary file;
FIG. 3 is a schematic diagram of a multi-clock binding system architecture;
FIG. 4 is a schematic diagram of clock relationship semantics;
FIG. 5 is a schematic diagram of the semantics of a clock expression;
FIG. 6 is a schematic diagram of a software system model driven development flow;
FIG. 7 is a schematic diagram of a collaborative unmanned system cut-in scenario;
FIG. 8 is a state machine diagram of a coordinated driverless system cut-in behavior;
FIG. 9 is a diagram illustrating simulation results according to an embodiment of the present invention.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments.
The invention discloses a method for realizing modeling and verification of a multi-clock constraint-oriented collaborative unmanned system, which specifically comprises the following steps:
s1, a framework of the collaborative unmanned system is provided, a field modeling summary file SysML Profile of the collaborative unmanned system is established, professional terms, relations and behavior operations of the collaborative unmanned system are extracted, communication between system design and analysts is facilitated, modeling and verification time is saved, and modeling accuracy is improved.
As shown in fig. 1, the cooperative unmanned driving system includes a cooperative sensing module, a brain-like driving module, and a vehicle control execution module, and forms a closed loop of "sensing-understanding-decision-execution-environment". The cooperative sensing module obtains information of surrounding vehicles, pedestrians and roads through communication between a sensor and the vehicles, wherein the sensor comprises a laser sensor, a radar sensor, a camera and the like, and the vehicle communication comprises vehicle-to-vehicle communication, vehicle-to-road communication, a global positioning system and the like; the cooperative perception module transmits the acquired surrounding traffic environment information to the brain-like driving module, the brain-like driving module comprises a scene understanding module and a behavior decision module, and the main functions of the brain-like driving module are scene understanding and behavior decision; the scene understanding module carries out data fusion processing on the perception information obtained by the cooperative perception module and outputs information which can be processed by the brain-like driving module, the brain-like driving module invokes the behavior decision module according to the data fused by the scene understanding module, and the behavior decision module generates decision information, namely safe driving control information; the behavior decision module transmits the decision information to the vehicle control execution module to generate corresponding vehicle control actions, and feeds back a behavior result to the surrounding traffic environment; the execution information of the vehicle control execution module mainly comprises longitudinal motion planning (acceleration, braking and the like) and transverse motion planning (steering and the like).
Through reading a large number of references, for the modeling capability of the collaborative unmanned system, the fact that the domain concepts of the existing general system modeling language SysML are unclear, the relationships among the concepts are disordered, and a general unified domain modeling extension language is lacked is found. Based on the problems, the invention expands and constructs the SysML Profile of the field modeling file, which is used for improving the modeling accuracy of the collaborative unmanned system and saving the modeling time.
As shown in fig. 2, the domain modeling Profile SysML Profile of the collaborative unmanned system mainly includes the relationships between modeling terms, operations, and modules of the system. The structural characteristics of the cooperative sensing module comprise sensor data, network communication, traffic snapshots, sensor tools and the like, and the behavior characteristics for acquiring traffic information comprise vehicle-to-vehicle communication, vehicle-to-road communication, sensor sensing and the like. The cooperative sensing module monitors the user and the traffic environment; the user structure characteristics mainly comprise user identity identification, gender, age and the like, and the user behavior characteristics mainly comprise vehicle starting, starting point and end point information input, artificial control and the like; the structural characteristics of the traffic environment comprise weather, road types, surrounding vehicles, surrounding pedestrians and the like, and the behavior characteristics of the traffic environment comprise environment updating, data transmission and the like; the cooperative sensing module triggers the scene understanding module, the cooperative sensing module is mainly used for processing sensed user and environment information and outputting effective information which can be understood by a computer, the structural characteristics of the scene understanding module comprise traffic rules, prior traffic information and the like, and the behavior characteristics of the scene understanding module comprise data fusion, traffic sign identification, observation and the like; the output of the scene understanding module triggers a behavior decision module, a decision system generates behavior decision information of the vehicle by using the scene understanding information subjected to fusion processing, wherein the structural characteristics of the decision system comprise road application, road occupation, required acceleration, required speed, application withdrawal, occupation withdrawal and the like, and the behavior characteristics of the decision system comprise road control, longitudinal control, transverse control and the like; the input of the behavior decision information excites a vehicle control execution module, the vehicle control execution module controls the vehicle according to the input of the behavior decision information, the structural characteristics of the vehicle control execution module comprise space-time trajectory, advancing direction, transverse acceleration, emergency control and the like, the behavior characteristics of the vehicle control execution module comprise brake control, steering control, accelerator control, gear control and the like, and the vehicle control information is fed back to the user and the traffic environment, so that the user and the surrounding traffic environment verify and control the vehicle execution information. Fig. 2 also lists two enumeration modules, and the traffic snapshot structure characteristics include an occupied road number, an applied road number, a position, a speed, an acceleration and a safety distance; enumerated sensor tool names include laser sensors, radar sensors, cameras, and vision sensors.
S2, defining a single-clock and multi-clock system, and giving definition and relation semantics related to the clocks. The method specifically comprises the following steps:
s21, defining a single clock system and a multi-clock system;
the present invention describes a distributed embedded system as a single clock system or a multi-clock system, as shown in fig. 3, the single clock system is defined as a clock system in which a master clock needs to be established, and the clock of each subsystem or sub-component must be referred to the master clock, and sampling is performed in the master clock to complete the interaction with the overall system. The multi-clock system is defined as a clock system which has no global clock, each subsystem/sub-component obeys multiple clock frequencies, has an independent clock to control the action behavior of the subsystem/sub-component, and realizes the running of the global system through the partial order interaction relationship of the behavior among local subsystems. The multi-clock system is a globally asynchronous locally synchronous distributed system and is suitable for describing a distributed and highly concurrent collaborative unmanned system.
S22, giving definition of clock correlation and definition and semantics of clock relation and clock expression of clock Constraint Specification language CCSL (clock Constraint Specification language);
the clock constraint specification language CCSL is a formal specification language based on clocks and clock constraints, and the clocks represent the trigger state of a signal at each discrete time point; the clock constraints describe the logical relationship between the signal clocks, and include the clock relationship and the clock expression, whose specific semantic graph representation is shown in fig. 4 and 5.
(1) Logic clocks, clock scheduling and related definition of clock history;
logic clock: a logic clock c being defined as consisting of a finite or infinite clock jitterThe sequence of the sequence is determined by the sequence,
Figure BDA0002727539620000121
Figure BDA0002727539620000122
is a set of natural numbers (excluding 0). For each one of ciAn element { tick, idle } indicates that the clock can be either a triggered state tick or an un-triggered state idle in the state at time i.
Clock scheduling: a clock schedule sigma reflects the trigger state of all clocks at each moment in discrete time, and is a finite or infinite sequence
Figure BDA0002727539620000131
Figure BDA0002727539620000132
Is a set of 0 and positive integers. Sigma (i) epsilon 2cRepresents a set of clocks that trigger at time i, formally defined as:
Figure BDA0002727539620000133
where C represents a set of finite logic clocks.
Clock history: for a given clock schedule σ, the clock history records the number of triggers that each clock has expired to the current time in σ, formally defined as:
Figure BDA0002727539620000134
wherein Hσ(c, i) represents the number of triggers that a logic clock c expires to time i in clock schedule σ; hσ(c, i +1) represents the number of triggers for the logic clock c to expire at time i +1 in the clock schedule σ; hσ(c,0) represents the number of triggers for the logic clock c to expire at time 0 in clock schedule σ; i. j represents the time of the clock.
(2) Clock relationship semantics, as shown in FIG. 4;
the priority relationship is: logic clock c1Faster than the logic clock c2Is triggered.
Figure BDA0002727539620000135
Wherein Hσ(c1I) represents the logic clock c in clock scheduling σ1The number of triggers until time i; hσ(c2I) represents the logic clock c in clock scheduling σ2By the number of triggers to time i.
② mutual exclusion relationship: if logic clock c1Is triggered, then logic clock c2Not trigger and vice versa.
Figure BDA0002727539620000136
③ child-hour relationship: at any time, if the logic clock c1Triggered then logic clock c2Must be triggered.
Figure BDA0002727539620000137
Fourthly, simultaneously relating: logic clock c1And logic clock c2And simultaneously triggered.
Figure BDA0002727539620000138
A causal relationship: logic clock c1Is not slower than the logic clock c2Is triggered.
Figure BDA0002727539620000141
Sixthly, the alternating relation is as follows: logic clock c1And logic clock c2And (4) alternately triggering.
Figure BDA0002727539620000142
Wherein, c1(i) Representing a logic clock c1At the i-th time, c2(i) Representing a logic clock c2At the i-th time, c1(i +1) represents a logic clock c1At time i + 1.
(3) The clock expression semantics, as shown in fig. 5;
combining clocks: triggering of clock and only if logic clock c1OR logic clock c2Is triggered.
Figure BDA0002727539620000143
Clock crossing: triggering of clock crossing and only if logic clock c1And logic clock c2While simultaneously triggering.
Figure BDA0002727539620000144
Clock delay: triggering and logic clock c of the clock2Are simultaneously triggered at c2Before, there is a logic clock c1And logic clock c2Trigger ratio logic clock c1The number of triggers of (d) is large.
Figure BDA0002727539620000145
Clock period: triggering of this clock is with p logical clocks c1Is triggered periodically.
Figure BDA0002727539620000146
Wherein Hσ(c0I) stands for scheduling at clockIn sigma, logic clock c0By the number of triggers to time i.
Clock sampling: the clock is based on the clock and the logic clock c2For logic clock c1Sampling, in particular the toggling of the clock and the logic clock c2Triggered simultaneously at time i and preceded by a time j,0<j is less than or equal to i (including the moment i), and the moment logic clock c1And at any time
Figure BDA0002727539620000147
Logic clock c2And not triggered.
Figure BDA0002727539620000148
Sixthly, strictly sampling by a clock: the clock is based on the clock and the logic clock c2For logic clock c1Performing rigorous sampling, in particular, the triggering of the clock and the logic clock c2Triggered simultaneously at time i and preceded by a time j,0<j<i (excluding the time i), the time logic clock c1And at any time
Figure BDA0002727539620000151
Logic clock c2And not triggered.
Figure BDA0002727539620000152
Upper clock bound: the clock defines a ratio of logic clocks c1And logic clock c2The fastest clock among the slow-triggering clocks.
Figure BDA0002727539620000153
Clock lower bound: the clock defines a ratio of logic clocks c1And logic clock c2Of the clock that triggers fastA clock.
Figure BDA0002727539620000154
And S3, modeling the collaborative unmanned system according to the model driving system development process of the system.
S31, driving a development process by the aid of a collaborative unmanned system model;
as shown in fig. 6, model-driven development of software systems, especially for safety-critical systems, is necessary and critical, and can effectively reduce the defects of the systems. Firstly, analyzing system requirements in a natural language form by a system designer, extracting basic concepts of system design and relations among the concepts, and constructing a calculation irrelevant model; secondly, through model conversion, adding logic data related to application, and converting a calculation-independent model into a platform-independent model; then, converting the platform-independent model into a platform-dependent model by adding specific platform-dependent data; and finally, converting the platform-related model into a system text code through conversion from the model to the text. In the process of model drive development of a software system, the key point is the conversion from a platform-independent model to a platform-dependent model, meta-models of a source model, a conversion model and a target model and corresponding summary files need to be constructed, the meta-models define the grammar and the semantics of a conversion language, the summary files define the grammar and the semantics of a specific conversion language, and specific application platform and domain knowledge are considered; the transformation model defines the transformation rules from the state machine graph meta-model to the formalized clock constraint specification language. The source model, the conversion model and the target model are respectively subject to a source meta-model, a conversion meta-model and a target meta-model, wherein the source model and the target model can be respectively instantiated into a platform-independent model and a platform-dependent model, the model conversion from the platform-independent model to the platform-dependent model is completed, and the specific flow of the model conversion refers to fig. 6. The invention is based on the development process of model drive, and carries out modeling and verification on the cooperative unmanned system.
S32, providing formal definitions of the meta model and the state machine diagram of the collaborative unmanned system;
(1) the meta-model of the collaborative unmanned system is a quadruple: AAM:: ═ BA,RA,EA,CA)；
①BARepresenting a system block definition diagram;
BA＝{Perception,Understanding,Planner,Actuator,User,Environment,KnowledgeBase}；
wherein a permission represents a Perception; underranging represents scene Understanding; planner represents behavioral decision-making; actuator denotes execution; user represents a User; environment represents an Environment; knowledgbasee represents a traffic knowledge base.
②RAThe relationship of system functions is defined;
Figure BDA0002727539620000161
wherein monitor represents monitoring; trigger represents trigger; precedence represents the precedence; stimulus stands for stimulation; response represents feedback; invoke represents a call.
③EARepresenting the system operating function, EA:BA×RA→BA；
④CAConstraint information, S, is definedA＝{Invariant,Precondition,Postcondition}；
Wherein Invariant represents an invariance; precondition represents a prerequisite; postcondition denotes a Postcondition.
(2) Giving a formal definition of a state machine diagram;
the syntax for formalized representation of the state machine diagram is defined as the six-tuple SMD: (S, S)0,E,G,Act,Tr)；
S represents a state set in a state machine diagram;
②S0representing an initial set of states in a state machine diagram;
thirdly, E represents a set of trigger events in state transition;
g represents a set of satellite viewing conditions in state transition;
act represents a set of migration actions in state migration;
sixthly, Tr represents a set of migration relations in the state machine diagram,
Figure BDA0002727539620000162
and when a trigger event occurs and guard conditions are met, Tr executes a migration action to trigger the migration between the bodies in the state machine diagram.
The semantics of the formalized representation of the state machine diagram are as follows:
operating semantics in sequence:
Figure BDA0002727539620000171
wherein s issRepresenting the source state, stRepresenting a target state, e representing a trigger event in state transition, g representing a satellite condition in state transition, a representing a transition action in state transition, when the trigger event and the satellite condition are satisfied, the transition action is executed,
Figure BDA0002727539620000172
migration occurs;
selecting operation semantics: different migration conditions will cause different migration to occur;
Figure BDA0002727539620000173
wherein choise denotes a select operation, ssRepresenting the source state, stRepresenting the target state, eiRepresents a trigger event in state transition i, giIndicating satellite Condition in State transition i, aiIndicating a migration action in state migration i.
Figure BDA0002727539620000174
Wherein s iss1Representing a first source state, e1[g1]/a1,e2[g2]/a2And ei[gi]/aiRepresenting different migration conditions, respectively representing trigger events, guard conditions and migration actions in the 1 st, 2 nd and i-th migration conditions, st1,st2And stiRepresenting the 1 st, 2 nd and ith target states.
And thirdly, forking operation semantics: the bifurcation operation is that under the same migration condition, the migration target state is a composite state consisting of two or more orthogonal nodes.
Figure BDA0002727539620000175
Wherein fork represents a fork operation, r1.s1int,r2.s2int,rn.snintIndicating the target states of the 1 st, 2 nd and ith orthogonal regions in the fork operation.
Figure BDA0002727539620000176
Wherein the content of the first and second substances,e[g]trigger events and guard conditions representing migration conditions.
Fourthly, converging operation semantics: the join operation is a migration from two or more mutually independent, concurrent and synchronous orthogonal source states to one target state.
join→(r1.s1fin|r2.s2fin|…|rn.snfin)×e×g×a×st；
Where join represents a join operation, r1.s1fin,r2.s2finAnd rn.snfinIndicating the initial states belonging to the 1 st, 2 nd and ith orthogonal regions of the merging operation, respectively.
Figure BDA0002727539620000181
S33, defining a mapping rule between the meta-model and the state-based clock constraint specification language, and converting the formalized state machine diagram into a formalized clock constraint specification language CCSL according to the mapping rule;
the model conversion rules between the state machine graph meta-model and the clock constraint reduction language are as follows:
conversion between states: mapping the state in the state machine diagram to the state in the CCSL protocol;
Figure BDA0002727539620000182
wherein s issmdRepresents the state of the state machine diagram smd, SsmdDenotes ssmdSet of (1), sccslRepresenting states, S, in a clock constraint specification language, CCSLccslDenotes sccslA collection of (a).
Triggering the mapping between the event and the clock: mapping the trigger event in the state machine diagram to a logic clock in a CCSL protocol;
Figure BDA0002727539620000183
wherein E issmdRepresents a set of edges in the state machine diagram, λ (Tr) represents an action in the migration relationship in CCSL, TrccslRepresenting a set of migration relationships in the CCSL.
③ the mapping rule between guard condition and Boolean condition in the clock constraint reduction language: mapping guard conditions in a state machine to Boolean expressions in CCSL;
Figure BDA0002727539620000184
wherein G issmdRepresenting a set of guard conditions, Tr, in a state machine diagramchoice.BoolRepresents selection of Boolean condition, Tr, in CCSLccslRepresents Trchoice.BoolA collection of (a).
Fourthly, the mapping rule between the operation semantics of the sequential expression: mapping sequence expressions in a state machine diagram to Boolean sequence condition migration in a CCSL protocol;
Figure BDA0002727539620000191
wherein s is1、s2Representing two states, Tr, in a state machine diagramsmdRepresents a set s 'representing a migration relationship in CCSL'1、s’2Representing two states in the CCSL.
Selecting a mapping rule between expression operation semantics: mapping a selection expression in a state machine diagram to Boolean selection condition migration in a CCSL protocol;
Figure BDA0002727539620000192
wherein S iss、StChoice represents the set of selected operating states in the CCSL clock relationship.
Sixthly, mapping rules among the operation semantics of the bifurcated expression are as follows: mapping a bifurcated expression in a state machine diagram to clock synchronization condition migration in a CCSL protocol;
Figure BDA0002727539620000193
wherein r is1,2,…,nRepresenting different orthogonal regions, s, in a state machine diagram1int,2int,…,nintRepresenting target states, s 'of the 1 st, 2 nd and nth orthogonal regions in a fork operation'sIs a source state in CCSL, tr'1,tr’2,tr’nIndicating different target states after a fork operation in the CCSL.
Seventhly, converging and selecting a mapping rule between expression operation semantics: mapping a confluent expression in a state machine diagram to conditional migration of a clock delay expression in a CCSL protocol;
Figure BDA0002727539620000194
wherein s is1fin,2fin,…,nfinDenotes an initial state, s'1,’s2,s’nRepresenting different initial states of the CCSL merge operation.
And S4, for the cooperative unmanned system, constructing a state machine model of the system according to the natural language requirement of the system, converting the state machine model into a corresponding CCSL model, and verifying the state machine model on a special simulation platform TimeSquare of a clock constraint protocol language.
Example (b): according to the modeling and verification method, the requirements of the overtaking behavior of the unmanned automobile are modeled and verified in the expressway scene.
First, a description of the requirements of the collaborative unmanned overtaking system is given, as shown in fig. 7: when the current self-vehicle on the road has the intention of overtaking, the turn light is firstly required to be turned on to prompt that the surrounding vehicle has the intention of overtaking, and at the moment, three safety-critical things need to be confirmed: the first is the need to confirm whether the vehicle is within a safe distance from the adjacent lane-change vehicle; secondly, after the surrounding vehicles observe the lane change signal, the rear vehicles cannot accelerate; and thirdly, sending a passing signal to the front vehicle. After receiving the overtaking request signal, the Ego-front-vehicle needs to reply to the agreement or refusal within a certain time (reaction time, supposing 4 seconds), if Ego-vehicle does not receive reply information or refusal information within the reaction time, the Ego-vehicle Ego-vehicle cannot implement lane change overtaking, and needs to keep the original state to advance; lane change overtaking may also not be implemented if the safety distance constraint is not satisfied with surrounding on-demand road vehicles. Lane change overtaking can be implemented only when the agreement information is received within the appointed reaction time and the safety distance is met. The overtaking action can be decomposed into two times of lane change, firstly, the first lane change is carried out to the applied road, the acceleration is needed to advance after the lane change, in the process, the previously agreed vehicles cannot accelerate, when the relative spatial position of the self vehicle ego-vehicle exceeds the self vehicle ego-front-vehicle in front of the self vehicle and the safe distance of the lane change is met, the self vehicle carries out the second lane change, and the time interval of the two lane change overtaking cannot exceed overtaking specified time, such as 10 seconds.
Next, a state machine description of the case system is given in conjunction with the extended domain modeling summary file in step S1 and the formal definition of the meta model and the state machine diagram of the collaborative unmanned system in step S3 of the modeling and verification method described above, as shown in fig. 8.
The natural language requirements are processed with reference to the domain extension file and the extraction introduction of the meta model, and the state machine diagram comprises 10 states, an initial state, a final state and states s 1-s 8 (respectively, a left-turn signal, a lane change application, a lane change disallowance, collision detection, a first lane change, an overtaking implementation, a right-turn signal and a second lane change). Where the s4 collision detection state is a composite state, containing two sub-states for collision detection of the preceding vehicle and potential collision detection of the following vehicle, respectively. The state S2 includes a state invariance specifying that the waiting time for the lane change request does not exceed 4S, the state S6 includes a state invariance specifying that the overtaking time does not exceed 10S, and the preceding overtaken vehicle cannot accelerate during the overtaking process. In the transition condition from the state S6 to the state S7, it is specified that the overtaking vehicle position and the overtaken vehicle position are separated by a distance greater than the safety distance, the overtaking time is about 10S, and the triggered transition action of the overtaking vehicle is an accelerated overtaking.
Then, a model of a clock constraint reduction language for the co-unmanned system vehicle passing case is constructed in combination with the above steps S2 and S3.
First, we will introduce the meaning of clock abbreviations in the CCSL protocol: an ego vehicle steering signal ego _ turn _ signal: ets, an application road is in accordance with a safe distance application _ lane _ safe distance: asd, a vehicle behind the ego vehicle can not accelerate ego _ before _ no _ access: ebna, an ego vehicle application lane change ego _ send _ lanechange: esl, a vehicle ahead of the ego vehicle replies to an agreement ego _ front _ age: efa, a vehicle ahead of the ego vehicle replies to an agreement ego _ front _ distance: efd, an ego vehicle can not overtake ego _ no _ overrating: eno, a vehicle ahead of the ego vehicle can not accelerate ego _ front _ no _ access: efna, and an ego vehicle replies to an overtake completion ego _ send _ overlance: eso.
CCSL1：ets～(asd*ebna)；
The ego-vehicle steering signal ets alternates with the clock crossing (if and only if the clock asd and the clock ebna trigger simultaneously).
CCSL2：(asd∨ebna)≤esl；
The upper clock bound asd V ebna is the fastest clock among the clocks that are slower than the triggering of clock asd and clock ebna, which triggered or caused the occurrence of clock esl.
Figure BDA0002727539620000211
The toggling of clock efa + efd (toggling of clock efa or clock efd) is slower by 4s than the toggling of clock esl.
CCSL4：efd≡eno；
The clock efd and the clock eno occur simultaneously, namely when the self-vehicle receives an unexpected signal returned by the front vehicle coupler, the self-vehicle is triggered to not overtake.
CCSL5：(efa∨asd)≡efna；
The upper clock bound efa V-old is the fastest clock among the clocks that are slower than the triggers of clock efa and clock asd, whose occurrence triggers the coincidence of clock efna, i.e., the vehicle in front cannot accelerate.
Figure BDA0002727539620000212
The activation of the clock efa is faster than the activation of the clock eso and the ego vehicle completes the overtaking action eso within the former vehicle's consent efa overtaking 10 s.
Finally, by combining the verification guiding idea of the step S4, in the simulation verification tool timessquare, simulation verification analysis is performed on the constructed clock constraint reduction language model, and the result shows that the multi-clock constraint relationship satisfies the requirement description of the system, as shown in fig. 9. Table 1 shows the correspondence between the CCSL specifications and the validation tool TimeSquare expression.
TABLE 1 CCSL protocol and TimeSquare expression
Figure BDA0002727539620000221
For any collaborative unmanned system, according to the method provided by the invention, firstly, natural language requirements are analyzed, concepts of the created system, logical relations among the concepts and required function operations are extracted according to the field modeling summary file which is established in an expanding way in the invention, and a state machine model of the system is established; then converting the state machine model into a multi-clock-protocol constraint language according to the semantic mapping rule of the model conversion to obtain a CCSL multi-clock-protocol model; and finally, performing clock relation analysis on the created model by using a special verification tool of the multi-clock protocol model.

Claims (7)

1. The method for realizing the modeling and verification of the multi-clock constraint-oriented collaborative unmanned system is characterized by comprising the following steps of:
s1, providing a framework of the collaborative unmanned system, establishing a field modeling summary file SysML Profile of the collaborative unmanned system, and extracting professional terms, relations and behavior operations of the collaborative unmanned system;
s2, defining a single-clock and multi-clock system, and giving definition and relation semantics related to the clocks;
s3, modeling the collaborative unmanned system according to the development process of the model driving system of the system;
and S4, for the cooperative unmanned system, constructing a state machine model of the system according to the natural language requirement of the system, converting the state machine model into a corresponding CCSL model, and verifying the state machine model on a special simulation platform TimeSquare of a clock constraint protocol language.
2. The method for modeling and verifying the collaborative unmanned aerial system facing multiple clock constraints according to claim 1, wherein the collaborative unmanned aerial system in step S1 comprises a collaborative sensing module, a brain-like driving module and a vehicle control execution module, the collaborative sensing module obtains surrounding vehicles, pedestrians and road information through sensor and vehicle communication, wherein the sensor comprises a laser sensor, a radar sensor and a camera, and the vehicle communication comprises vehicle-to-vehicle communication, vehicle-to-road communication and a global positioning system; the cooperative perception module transmits the acquired surrounding traffic environment information to a brain-like driving module, and the brain-like driving module comprises a scene understanding module and a behavior decision module; the scene understanding module carries out data fusion processing on the perception information obtained by the collaborative perception module and outputs information which can be processed by the brain-like driving module, the brain-like driving module invokes the behavior decision module according to the data fused by the scene understanding module, and the behavior decision module generates decision information; the behavior decision module transmits the decision information to the vehicle control execution module to generate corresponding vehicle control actions, and feeds back the behavior result to the surrounding traffic environment.
3. The method for implementing modeling and verification of a multi-clock constraint-oriented collaborative unmanned aerial vehicle system according to claim 1, wherein the SysML Profile of the domain modeling Profile of the collaborative unmanned aerial vehicle system in step S1 includes modeling terms, operations and relationships between modules of the system, the structural characteristics of the collaborative awareness modules include sensor data, network communication, traffic snapshots and sensor tools, and the behavior characteristics of obtaining traffic information include vehicle-to-vehicle communication, vehicle-to-road communication and sensor awareness; the cooperative sensing module monitors the user and the traffic environment; the user structure characteristics comprise user identity identification, gender and age, and the user behavior characteristics comprise vehicle starting, starting point and end point information input and artificial control; the structural characteristics of the traffic environment comprise weather, road types, surrounding vehicles and surrounding pedestrians, and the behavior characteristics of the traffic environment comprise environment updating and data transmission; the cooperative sensing module triggers the scene understanding module, the cooperative sensing module is used for processing sensed user and environment information and outputting effective information which can be understood by a computer, the structural characteristics of the scene understanding module comprise traffic rules and prior traffic information, and the behavior characteristics of the scene understanding module comprise data fusion, traffic sign identification and observation; the output of the scene understanding module triggers a behavior decision module, a decision system generates behavior decision information of the vehicle by using the scene understanding information subjected to fusion processing, wherein the structural characteristics of the decision system comprise road application, road occupation, required acceleration, required speed, application withdrawal and occupation withdrawal, and the behavior characteristics of the decision system comprise road control, longitudinal control and transverse control; the input of the behavior decision information excites a vehicle control execution module, the vehicle control execution module controls the vehicle according to the input of the behavior decision information, the structural characteristics of the vehicle control execution module comprise space-time trajectory, advancing direction, transverse acceleration and emergency control, the behavior characteristics of the vehicle control execution module comprise brake control, steering control, accelerator control and gear control, and the vehicle control information is fed back to the user and the traffic environment, so that the user and the surrounding traffic environment can verify and control the vehicle execution information.
4. The method for modeling and verifying the collaborative unmanned system based on multi-clock constraint according to claim 1, wherein the single-clock system is defined in step S2 as a master clock needs to be established in the system, and the clock of each subsystem or sub-component must be referenced to the master clock, and the sampling is performed in the master clock to complete the clock system interacting with the overall system; the multi-clock system is defined as a clock system which has no global clock, each subsystem/sub-component obeys multiple clock frequencies, has an independent clock to control the action behavior of the subsystem/sub-component, and realizes the running of the global system through the partial order interaction relationship of the behavior among local subsystems.
5. The method for modeling and validating a multi-clock constraint oriented collaborative unmanned aerial system according to claim 1, wherein the Clock Constraint Specification Language (CCSL) in step S2 is a formal specification language based on clocks and clock constraints, and the clocks represent trigger states of a signal at each discrete time point; the clock constraint describes a logic relationship between signal clocks, and the clock constraint includes a clock relationship and a clock expression, and specifically includes:
(1) logic clocks, clock scheduling and related definition of clock history;
logic clock: a logic clock c is defined as a sequence of finite or infinite clock jumps,
Figure FDA0002727539610000021
Figure FDA0002727539610000022
a set of non-0 natural numbers; for each one of ciE { tick, idle }, which indicates that the clock is in a trigger state tick or a non-trigger state idle in the state of the time i;
clock scheduling: a clock schedule sigma, which reflects the trigger state of all clocks at each moment in discrete time, is a finite or infinite sequence sigma (0) sigma (1) … sigma (i) …,
Figure FDA0002727539610000023
Figure FDA0002727539610000024
is a set of 0 and positive integers; sigma (i) epsilon 2cRepresents a set of clocks that trigger at time i, formally defined as:
Figure FDA0002727539610000025
wherein C represents a set of finite logic clocks;
clock history: for a given clock schedule σ, the clock history records the number of triggers that each clock has expired to the current time in σ, formally defined as:
Figure FDA0002727539610000031
wherein Hσ(c, i) represents the number of triggers that a logic clock c expires to time i in clock schedule σ; hσ(c, i +1) represents the number of triggers for the logic clock c to expire at time i +1 in the clock schedule σ; hσ(c,0) represents the number of triggers for the logic clock c to expire at time 0 in clock schedule σ; i, j represents the time of the clock;
(2) clock relationship semantics;
the priority relationship is: logic clock c1Faster than the logic clock c2Triggering of (1);
Figure FDA0002727539610000032
wherein Hσ(c1I) represents the logic clock c in clock scheduling σ1The number of triggers until time i; hσ(c2I) represents the logic clock c in clock scheduling σ2The number of triggers until time i;
② mutual exclusion relationship: if logic clock c1Is triggered, then logic clock c2Not trigger and vice versa;
Figure FDA0002727539610000033
③ child-hour relationship: at any time, if the logic clock c1Triggered then logic clock c2Must be triggered;
Figure FDA0002727539610000034
fourthly, simultaneously relating: logic clock c1And logic clock c2Triggering at the same time;
Figure FDA0002727539610000035
a causal relationship: logic clock c1Is not slower than the logic clock c2Triggering of (1);
Figure FDA0002727539610000036
sixthly, the alternating relation is as follows: logic clock c1And logic clock c2Alternately triggering;
Figure FDA0002727539610000037
wherein, c1(i) Representing a logic clock c1At the i-th time, c2(i) Representing a logic clock c2At the i-th time, c1(i +1) represents a logic clock c1Time i + 1;
(3) a time-clock expression semantic;
combining clocks: triggering of clock and only if logic clock c1OR logic clock c2Triggering of (1);
Figure FDA0002727539610000041
clock crossing: triggering of clock crossing and only if logic clock c1And logic clock c2Simultaneously triggering;
Figure FDA0002727539610000042
clock delay: triggering and logic clock c of the clock2Are simultaneously triggered at c2Before, there is a logic clock c1Is triggered byAnd logic clock c2Trigger ratio logic clock c1D more triggering times;
Figure FDA0002727539610000043
clock period: triggering of this clock is with p logical clocks c1The triggering of (2) is periodic triggering;
Figure FDA0002727539610000044
wherein Hσ(c0I) represents the logic clock c in clock scheduling σ0The number of triggers until time i;
clock sampling: the clock is based on the clock and the logic clock c2For logic clock c1Sampling, in particular the toggling of the clock and the logic clock c2Triggered simultaneously at time i and preceded by a time j,0<j is less than or equal to i, the logic clock c of the moment1And at any time
Figure FDA0002727539610000048
Logic clock c2Not triggering;
Figure FDA0002727539610000045
sixthly, strictly sampling by a clock: the clock is based on the clock and the logic clock c2For logic clock c1Performing rigorous sampling, in particular, the triggering of the clock and the logic clock c2Triggered simultaneously at time i and preceded by a time j,0<j<i, the time logic clock c1And at any time
Figure FDA0002727539610000046
Logic clock c2Do not touchHair is sent;
Figure FDA0002727539610000047
upper clock bound: the clock defines a logical clock c1And logic clock c2The fastest clock among the slow-triggering clocks;
Figure FDA0002727539610000051
clock lower bound: the clock defines a logical clock c1And logic clock c2The slowest clock among the triggered fast clocks;
Figure FDA0002727539610000052
6. the method for implementing modeling and verification of a multi-clock constraint-oriented collaborative unmanned aerial system according to claim 1, wherein the collaborative unmanned aerial system model-driven development process in step S3 is specifically as follows: firstly, analyzing system requirements in a natural language form by a system designer, extracting basic concepts of system design and relations among the concepts, and constructing a calculation irrelevant model; secondly, through model conversion, adding logic data related to application, and converting a calculation-independent model into a platform-independent model; then, converting the platform-independent model into a platform-dependent model by adding specific platform-dependent data; finally, converting the platform-related model into a system text code through conversion from the model to the text; in the process of model drive development of a software system, meta-models and corresponding summary files of a source model, a conversion model and a target model need to be constructed for conversion from a platform-independent model to a platform-dependent model, wherein the meta-models define the grammar and the semantics of a conversion language, and the summary files define the grammar and the semantics of a specific conversion language and take specific application platform and domain knowledge into consideration; the conversion model defines the conversion rule from the state machine diagram meta model to the formal clock constraint reduction language; the source model, the conversion model and the target model are respectively obeyed to the source meta-model, the conversion meta-model and the target meta-model, wherein the source model and the target model are respectively instantiated into a platform-independent model and a platform-dependent model, and model conversion from the platform-independent model to the platform-dependent model is completed.
7. The method for modeling and verifying the collaborative unmanned aerial system based on multi-clock constraints as claimed in claim 1, wherein in step S3, meta-models and state machine diagrams of the collaborative unmanned aerial system are formally defined;
(1) the meta-model of the collaborative unmanned system is a quadruple: AAM:: ═ BA,RA,EA,CA)；
①BARepresenting a system block definition diagram;
BA＝{Perception,Understanding,Planner,Actuator,User,Environment,KnowledgeBase}；
wherein a permission represents a Perception; underranging represents scene Understanding; planner represents behavioral decision-making; actuator denotes execution; user represents a User; environment represents an Environment; knowledgbasee represents a traffic knowledge base;
②RAthe relationship of system functions is defined;
Figure FDA0002727539610000061
wherein monitor represents monitoring; trigger represents trigger; precedence represents the precedence; stimulus stands for stimulation; response represents feedback; invoke represents a call;
③EArepresenting the system operating function, EA:BA×RA→BA；
④CAConstraint information, S, is definedA＝{Invariant,Precondition,Postcondition}；
Wherein Invariant represents an invariance; precondition represents a prerequisite; postcondition denotes a post condition;
(2) giving a formal definition of a state machine diagram;
the syntax for formalized representation of the state machine diagram is defined as the six-tuple SMD: (S, S)0,E,G,Act,Tr)；
S represents a state set in a state machine diagram;
②S0representing an initial set of states in a state machine diagram;
thirdly, E represents a set of trigger events in state transition;
g represents a set of satellite viewing conditions in state transition;
act represents a set of migration actions in state migration;
sixthly, Tr represents a set of migration relations in the state machine diagram,
Figure FDA0002727539610000062
wherein sv and tv are respectively a source state and a target state of a migration relationship, when a trigger event occurs and guard conditions are met, Tr executes a migration action to trigger the migration between the bodies in the state machine diagram;
the semantics of the formalized representation of the state machine diagram are as follows:
operating semantics in sequence:
Figure FDA0002727539610000063
e belongs to E, G belongs to G, a belongs to Act, wherein s belongs to E, G belongs to G, a belongs to ActsRepresenting the source state, stRepresenting a target state, e representing a trigger event in state transition, g representing a satellite condition in state transition, a representing a transition action in state transition, when the trigger event and the satellite condition are satisfied, the transition action is executed,
Figure FDA0002727539610000064
migration occurs;
selecting operation semantics: different migration conditions will cause different migration to occur;
Figure FDA0002727539610000071
wherein choice represents a selection operation, eiRepresents a trigger event in state transition i, giIndicating satellite Condition in State transition i, aiRepresenting a migration action in the state migration i;
Figure FDA0002727539610000072
wherein s iss1Representing a first source state, e1[g1]/a1,e2[g2]/a2And ei[gi]/aiRepresenting different migration conditions, respectively representing trigger events, guard conditions and migration actions in the 1 st, 2 nd and i-th migration conditions, st1、st2And stiRespectively representing the 1 st, 2 nd and ith target states;
and thirdly, forking operation semantics: the bifurcation operation is that under the same migration condition, the migration target state is a composite state consisting of two or more orthogonal nodes;
Figure FDA0002727539610000073
wherein fork represents a fork operation, r1.s1int,r2.s2int,rn.snintRespectively representing target states of 1 st, 2 nd and ith orthogonal areas in the bifurcation operation;
Figure FDA0002727539610000074
wherein e [ g ]]A trigger event and guard condition representing a migration condition;
fourthly, converging operation semantics: the merging operation is to be migrated to a target state by two or more mutually independent, concurrent and synchronous orthogonal source states;
join→(r1.s1fin|r2.s2fin|…|rn.snfin)×e×g×a×st；
where join represents a join operation, r1.s1fin,r2.s2finAnd rn.snfinRespectively representing initial states respectively belonging to the 1 st, 2 nd and ith orthogonal regions of the merging operation;
Figure FDA0002727539610000075
s33, defining a mapping rule between the meta-model and the state-based clock constraint specification language, and converting the formalized state machine diagram into a formalized clock constraint specification language CCSL according to the mapping rule;
the model conversion rules between the state machine graph meta-model and the clock constraint reduction language are as follows:
conversion between states: mapping the state in the state machine diagram to the state in the CCSL protocol;
Figure FDA0002727539610000081
wherein s issmdRepresents the state of the state machine diagram smd, SsmdDenotes ssmdSet of (1), sccslRepresenting states, S, in a clock constraint specification language, CCSLccslDenotes sccslA set of (a);
triggering the mapping between the event and the clock: mapping the trigger event in the state machine diagram to a logic clock in a CCSL protocol;
Figure FDA0002727539610000082
wherein E issmdRepresenting a set of edges, λ, in a state machine diagram(Tr) represents an action in a migration relationship in CCSL, TrccslA set representing migration relationships in the CCSL;
③ the mapping rule between guard condition and Boolean condition in the clock constraint reduction language: mapping guard conditions in a state machine to Boolean expressions in CCSL;
Figure FDA0002727539610000083
wherein G issmdRepresenting a set of guard conditions, Tr, in a state machine diagramchoice.BoolRepresents selection of Boolean condition, Tr, in CCSLccslRepresents Trchoice.BoolA set of (a);
fourthly, the mapping rule between the operation semantics of the sequential expression: mapping sequence expressions in a state machine diagram to Boolean sequence condition migration in a CCSL protocol;
Figure FDA0002727539610000084
wherein s is1、s2Representing two states, Tr, in a state machine diagramsmdRepresents a set s 'representing a migration relationship in CCSL'1、s'2Represents two states in the CCSL;
selecting a mapping rule between expression operation semantics: mapping a selection expression in a state machine diagram to Boolean selection condition migration in a CCSL protocol;
Figure FDA0002727539610000091
wherein S iss、StThe source state and the target state are respectively represented, and the CR.choice represents a set of selection operation states in a CCSL clock relationship;
sixthly, mapping rules among the operation semantics of the bifurcated expression are as follows: mapping a bifurcated expression in a state machine diagram to clock synchronization condition migration in a CCSL protocol;
Figure FDA0002727539610000092
wherein r is1,2,…,nRepresenting different orthogonal regions, s, in a state machine diagram1int,2int,…,nintRepresenting target states, s 'of the 1 st, 2 nd and nth orthogonal regions in a fork operation'sIs the source state in CCSL, tr1’,tr’2,tr'nRepresenting different target states after a fork operation in the CCSL;
seventhly, converging and selecting a mapping rule between expression operation semantics: mapping a confluent expression in a state machine diagram to conditional migration of a clock delay expression in a CCSL protocol;
Figure FDA0002727539610000093
wherein s is1fin,2fin,…,nfinDenotes an initial state, s'1、s'2、s'nRepresenting different initial states of the CCSL merge operation.
CN202011107719.7A 2020-10-16 2020-10-16 Modeling and verification implementation method of multi-clock constraint cooperative unmanned system Active CN112329208B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011107719.7A CN112329208B (en) 2020-10-16 2020-10-16 Modeling and verification implementation method of multi-clock constraint cooperative unmanned system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011107719.7A CN112329208B (en) 2020-10-16 2020-10-16 Modeling and verification implementation method of multi-clock constraint cooperative unmanned system

Publications (2)

Publication Number Publication Date
CN112329208A true CN112329208A (en) 2021-02-05
CN112329208B CN112329208B (en) 2024-05-17

Family

ID=74313900

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011107719.7A Active CN112329208B (en) 2020-10-16 2020-10-16 Modeling and verification implementation method of multi-clock constraint cooperative unmanned system

Country Status (1)

Country Link
CN (1) CN112329208B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115431904A (en) * 2022-08-05 2022-12-06 智己汽车科技有限公司 Vehicle function control method and device, vehicle and storage medium
WO2023016268A1 (en) * 2021-08-09 2023-02-16 华为技术有限公司 Vehicle control method and apparatus
CN115758789A (en) * 2022-12-01 2023-03-07 金航数码科技有限责任公司 Software architecture design and architecture transmission method of complex real-time embedded system

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060190867A1 (en) * 2005-02-03 2006-08-24 Baumgartner Jason R Method for reconfiguration of random biases in a synthesized design without recompilation
CN104915514A (en) * 2015-06-25 2015-09-16 华东师范大学 Time requirement modeling and verification method based on problem frame method
CN108919672A (en) * 2018-07-13 2018-11-30 华东师范大学 A kind of implementation method of complex industrial control system

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060190867A1 (en) * 2005-02-03 2006-08-24 Baumgartner Jason R Method for reconfiguration of random biases in a synthesized design without recompilation
CN104915514A (en) * 2015-06-25 2015-09-16 华东师范大学 Time requirement modeling and verification method based on problem frame method
CN108919672A (en) * 2018-07-13 2018-11-30 华东师范大学 A kind of implementation method of complex industrial control system

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023016268A1 (en) * 2021-08-09 2023-02-16 华为技术有限公司 Vehicle control method and apparatus
CN115431904A (en) * 2022-08-05 2022-12-06 智己汽车科技有限公司 Vehicle function control method and device, vehicle and storage medium
CN115431904B (en) * 2022-08-05 2024-05-14 智己汽车科技有限公司 Control method and device for vehicle functions, vehicle and storage medium
CN115758789A (en) * 2022-12-01 2023-03-07 金航数码科技有限责任公司 Software architecture design and architecture transmission method of complex real-time embedded system
CN115758789B (en) * 2022-12-01 2023-11-17 金航数码科技有限责任公司 Software architecture design and architecture transfer method of complex real-time embedded system

Also Published As

Publication number Publication date
CN112329208B (en) 2024-05-17

Similar Documents

Publication Publication Date Title
CN112329208B (en) Modeling and verification implementation method of multi-clock constraint cooperative unmanned system
Feng et al. Testing scenario library generation for connected and automated vehicles, part I: Methodology
Mubeen et al. Supporting timing analysis of vehicular embedded systems through the refinement of timing constraints
Korssen et al. Systematic model-based design and implementation of supervisors for advanced driver assistance systems
Damm et al. A formal semantics for traffic sequence charts
Elrofai et al. Scenario-based safety validation of connected and automated driving
Obergfell et al. Model-based resource analysis and synthesis of service-oriented automotive software architectures
Passerone et al. A methodology for the design of safety-compliant and secure communication of autonomous vehicles
Firl Probabilistic maneuver recognition in traffic scenarios
Kibalama et al. AV/ADAS Safety-Critical Testing Scenario Generation from Vehicle Crash Data
Meyer et al. Scenario-and model-based systems engineering procedure for the sotif-compliant design of automated driving functions
Hou et al. Twin scenarios establishment for autonomous vehicle digital twin empowered SOTIF assessment
Paulweber Validation of highly automated safe and secure systems
Furda et al. Towards increased road safety: Real-time decision making for driverless city vehicles
Barthauer et al. Coupling traffic and driving simulation: Taking advantage of SUMO and SILAB together
Khatun et al. An optimization and validation method to detect the collision scenarios and identifying the safety specification of highly automated driving vehicle
Van Acker et al. Managing heterogeneity in model-based systems engineering of cyber-physical systems
Wang et al. Multiclock constraint system modelling and verification for ensuring cooperative autonomous driving safety
Höfner et al. Algebraic view reconciliation
Kaiser et al. An AEBS use case for model-based system design integrating safety analyses and simulation
Zhang Specifying and modeling automotive cyber physical systems
CN113272744A (en) System for monitoring appropriate behavior of autonomous vehicle and method thereof
Möstl et al. Controlling Concurrent Change-A Multiview Approach Toward Updatable Vehicle Automation Systems
Santa et al. Relations of UML and OETPN Models
Elgharbawy A big testing framework for automated truck driving

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
GR01 Patent grant
GR01 Patent grant