CN107561932B - CPS anti-collision control method based on differential dynamic logic - Google Patents
CPS anti-collision control method based on differential dynamic logic Download PDFInfo
- Publication number
- CN107561932B CN107561932B CN201710609623.2A CN201710609623A CN107561932B CN 107561932 B CN107561932 B CN 107561932B CN 201710609623 A CN201710609623 A CN 201710609623A CN 107561932 B CN107561932 B CN 107561932B
- Authority
- CN
- China
- Prior art keywords
- collision
- verification
- decision
- cps
- safety
- 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.)
- Active
Links
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention discloses a CPS anti-collision control method based on differential dynamic logic, which comprises the steps of firstly carrying out anti-collision decision verification based on dL, predicting the safety of a system at a certain time in the future if the verification is successful, carrying out tentative forward operation and prompting and alarming if the verification fails or dangers are predicted, then screening a static optimal decision instruction according to a certain original or obtaining a dynamic optimal decision instruction based on dL if a system suggestion is adopted, otherwise judging whether the system takes over or continues to prompt and alarm according to the triggering condition of a minimum distance event trigger, and finally starting a braking device to avoid collision. The method carries out modeling and safety anti-collision decision verification based on dL, can dynamically acquire the optimal decision instruction, can predict the safety of a system at a certain time or a certain continuous or discontinuous time period in the future, and has the advantages of predictability, flexibility, precise instruction system and the like.
Description
Technical Field
The invention belongs to the field of intelligent navigation automatic control of CPS technology, and mainly relates to a CPS anti-collision control method based on differential dynamic logic.
Background
At present, most of even all vehicles cannot intelligently predict safety and flexibly navigate possible collision accidents by using collision prevention technologies, and various factors such as weather or human factors cause frequent collision accidents, so that life safety of people is threatened, and major economic loss is caused, and therefore, a more intelligent differential dynamic logic (dL) -based CPS collision prevention control method is very important. For example, some anti-collision devices such as radars or Beidou navigation devices adopted by ships have poor effects and cannot be automatically controlled, and meanwhile, due to the fact that the system has long delay time and large mass and inertia, emergency braking of the ships is very difficult when collision danger occurs; most of the airplane adopts an airborne collision avoidance system which has poor flexibility and low automatic control precision, and simultaneously, the airplane easily enters a dangerous area to fly and possibly causes collision accidents due to the fact that the airplane has high speed, large inertia and lacks of a safety prediction function.
Disclosure of Invention
The purpose of the invention is as follows: aiming at overcoming the defects in the prior art and solving the problem that a driver cannot avoid collision in time due to various factors such as weather or human factors in the driving process, the CPS anti-collision control method based on the differential dynamic logic is provided with predictability, flexibility and precise and simplified instruction system.
The technical scheme is as follows: in order to achieve the purpose, the invention provides a CPS anti-collision control method based on differential dynamic logic, which comprises the steps of firstly carrying out anti-collision decision verification based on dL, predicting the safety of a system at a certain time in the future if the verification is successful, carrying out tentative forward operation and prompting and alarming if the verification is failed or the danger is predicted, then screening a static optimal decision instruction according to a certain original or obtaining a dynamic optimal decision instruction based on dL if the verification is failed or the danger is predicted, otherwise judging whether the system takes over or continues to prompt and alarm according to the triggering condition of a minimum distance event trigger, and finally starting a braking device to avoid collision.
The method specifically comprises the following steps:
Step 1: determining security of a system
1.1 real-time modeling and validation
1.1.1 interaction, Collection and processing of information
The CPS node equipment is used for information interaction, acquisition and processing, and information such as acceleration, speed, distance and the like is acquired, and whether collision targets exist or not is analyzed, such as obstacles like reefs, piers, ships, airplanes, automobiles and the like.
1.1.2 dL-based real-time modeling and anti-collision decision verification
If the target is not found, the CPS node equipment acquisition information is returned, otherwise, the acquired information is used as an initial condition to construct a model and carry out anti-collision decision verification, if the verification is successful, the model is predicted to be modeled, and if the verification is not successful, a time trigger is started.
1.2 predictive modeling and validation
if the target is found and the anti-collision decision verification is successful, calculating information such as speed, distance and the like at a certain moment in the future according to the real-time information, constructing a model according to the information, and performing safety decision verification based on dL, if the verification is successful, returning CPS node equipment to acquire information, otherwise, starting a time trigger, and running in a tentative mode.
1.3 prompting and alerting
If the anti-collision decision verification fails or danger is predicted, starting a time trigger, displaying an acceptance button and a rejection button through a computer screen, and simultaneously giving an alarm by voice; in the case of a moving obstacle, the two or more parties cooperate to avoid the collision.
Step 2: selection acceptance button
2.1 screening static best decision Instructions
firstly, applying a Markov decision process optimization system instruction list to the current early warning event to screen out a plurality of ideal decision instructions, then applying a dynamic decision algorithm to calculate the respective expected values of the ideal decision instructions, and finally selecting the instruction with the maximum expected value.
2.2 obtaining dynamic best decision Instructions
2.2.1 update information
2.2.2 modeling based on dL and iteratively performing safety decision test validation
Building a model according to the updated information and carrying out security decision verification, if the verification fails, readjusting the target speed and modifying the model parameters, carrying out security decision verification again, and repeating the steps until the verification succeeds; and then assigning the acquired speed of the collision avoidance target to a dynamic instruction.
And step 3: selection rejection button
3.1 updating information and real-time modeling
And constructing a model according to the real-time information acquired by the CPS node equipment.
3.2 Untriggered minimum distance event trigger
And performing anti-collision decision verification based on the dL, returning CPS node acquisition information and reconstructing the model if the verification is successful, performing anti-collision decision verification based on the dL again, and prompting and alarming until the verification fails.
3.3 triggering minimum distance event triggers
And performing anti-collision decision verification based on dL, taking over by the system if the verification fails, otherwise, updating information, modeling at a certain future moment and performing security decision verification at the same time, if the verification succeeds, continuously updating the information, predicting the modeling and performing security decision verification at the same time, and repeating until the verification fails, and taking over by the system at the moment to obtain a super decision instruction.
And 4, step 4: and starting a braking device to avoid collision.
The system of the invention comprises: the system comprises a control center, CPS node equipment and an actuator, wherein the control center is responsible for receiving information and sending instructions, the CPS node equipment is responsible for collecting, processing, interacting and forwarding the information, and the actuator is responsible for steering, accelerating and decelerating.
The CPS node device includes: the system comprises an acceleration and speed sensor for measuring the acceleration and the speed of the system, a laser ranging sensor for measuring the distance between the system and static or moving close-range obstacles such as reefs, piers and the like, an interrogator and a responder for acquiring information such as the acceleration, the speed, the distance between the remote moving obstacles and the like. The three attributes can be used as input conditions for the verification process.
Has the advantages that: the invention carries out modeling and safety anti-collision decision verification based on dL and can dynamically acquire the optimal decision instruction, predict the safety of a system at a certain time or a certain continuous or discontinuous time period in the future, and simultaneously, two or more systems negotiate with each other to avoid collision, compared with the prior art, the invention has the following advantages:
1) Predicting the safety of a system at a future time or over a continuous or discontinuous period of time
The existing methods have not proposed the safety of predictive systems based on dL.
the safety prediction necessity examples are modeling according to information collected by a CPS node at the 1 st second, performing anti-collision decision verification based on dL, and verifying safety, but the safety critical point is at the moment, namely collision danger exists at the 2 nd second and later, if the time interval for collecting the information by the CPS node is 4 seconds, the 5 th second verification fails, the system already advances for 4 seconds in a dangerous area, and the personnel reflection (5 seconds) and the system delay (2 seconds) are added, the system starts a braking device to avoid collision at the 13 th second, but the system already advances for 12 seconds in the dangerous area at the moment, which possibly causes very serious consequences, if a safety prediction function is adopted and the safety prediction time interval delta τ is 5 seconds, the safety is verified at the 1 st second, but the safety prediction fails at the 6 th second, the system can perform safety prompt and alarm at the 1 st second, and therefore, the dead zone of collision danger of the system from the safety point to the safety prediction time interval delta τ is compensated.
2) The system is more flexible and intelligent
The instruction system of the existing anti-collision equipment is complex, the screening process is long, and the complexity and the emergent scene cannot be freely coped with, so that the invention provides a dynamic instruction and a method for acquiring the dynamic instruction based on dL, which is expected to make the system more intelligent and reduce the interference of human factors as much as possible.
Examples of flexibility necessities: in the existing instruction system, the acceleration and the target speed are fixed, a minimum collision avoidance navigation route is provided, if the calculated collision avoidance target speed is greater than the target speed fixed by the optimal decision instruction, human factors are required to participate, otherwise collision is ensured; if the dynamic instruction is adopted, the calculated speed of the collision avoidance target is only assigned to the dynamic instruction, and the participation of human factors is not needed at all.
3) can realize cooperative collision avoidance
At present, most of vehicles with information interaction capability exchange information through a third party, and a small part of vehicles with autonomous interaction capability acquire information such as distance between the vehicles, and can not mutually negotiate through two or more anti-collision systems to avoid collision.
Examples of synergistic necessities: under the conditions that the collision occurs in sudden scenes such as the minimum safe distance and corners and the maximum mechanical braking force also occurs, or in the normal driving process, two or more vehicles share collision-prevention decision information mutually, so that the collision is avoided easily. If the automobile b detects the collision danger based on dL and receives the super decision instruction information of the automobile a at the same time, the automobile b can calculate a collision avoidance measure or accelerate forwards at the maximum acceleration according to the instruction information of the automobile a, so that the occurrence of rear-end collision accidents is avoided.
Drawings
FIG. 1 is a flow diagram of the authentication system security of the present invention;
FIG. 2 is a flow chart of an instruction for obtaining an optimal decision according to the present invention;
FIG. 3 is a flow chart of a get super decision instruction of the present invention;
Fig. 4 is an example of a ship collision avoidance application scenario of the present invention;
Fig. 5 is a modular framework diagram of the collision avoidance system of the present invention.
Detailed Description
The present invention is further illustrated by the following figures and specific examples, which are to be understood as illustrative only and not as limiting the scope of the invention, which is to be given the full breadth of the appended claims and any and all equivalent modifications thereof which may occur to those skilled in the art upon reading the present specification.
An embodiment of the invention is described in further detail below with reference to fig. 1 to 5, as shown in fig. 4, in a scenario where a boat ownship is traveling freely to the left, is traveling horizontally at time τ 0 and finds a stationary boat intutter directly in front, is safe at time τ 1 but predicts a collision risk at time τ 1, assuming a CPS node device information acquisition period T o of 1S, a safety prediction time interval Δ τ of 5S, and a minimum distance S min of 1000 m.
The embodiment comprises the following specific steps:
step 1) judging the safety of the system, as shown in the attached figure 1:
At time τ 0, the ownship finds the target and performs modeling and collision avoidance decision validation based on dL and predicts safety at time τ 0 + Δ τ because T o is 1s, then each time τ 0 + NT o (N is 1, 2.) will be based on the safety of the dL validation system and predict the safety of the time τ 0 + Δ τ + NT o (N is 1, 2.) system at time τ 1 validates safety, but a hazard is predicted at time τ 1 + Δ τ, at which point the ownship sails forward, the computer screen displays accept and reject buttons and alerts the crew, while negotiating a collision avoidance with the ship intaker through the CPS node device, and the ship intaker starts sailing forward horizontally at time τ 1.
step 2) selecting an accept button, as shown in fig. 2:
At time τ 1, the target speed of the ship ownship collision avoidance is v o and the fixed speed of the screened static optimal decision command is v min, but v o > v min, if no human factor participates in the process, the risk of collision still exists, namely the screening of the static optimal decision command fails, therefore, a model is built according to the information collected at time τ 1 + NT o (N is 0, 1.), anti-collision decision verification is carried out, if the verification fails, the target speed is readjusted and the verification is carried out again, the process is repeated until the verification is successful at time τ 2, the collision avoidance target speed v o is obtained, v o is assigned to the dynamic command, at the moment, the system sails forward heuristically (τ 2 - τ 1) for seconds, and then the process goes to step 4 below.
Step 3) selecting a reject button, as shown in fig. 3:
At time T 1, the horizontal distance between the two ships is S real and S real > S min, at which time the ship ownship constructs a model from the information acquired at time T 1 + NT o (N0, 1.), and performs collision avoidance decision verification, if verification succeeds, the ship ownship constructs a model from the information acquired at time T 1 + NT o (N1, 2.), and performs collision avoidance decision verification, and if the verification fails, the prompting and warning are continued until time T 2 is not adopted, but at this time S real is S min, that is, a minimum distance trigger is triggered (of course, the time from T 1 to T 2 is as long as S real is S min immediately triggers the minimum distance trigger), so that collision avoidance decision verification is performed at time T 2, if verification fails, the system takes over and turns to the following step 4), otherwise, the information at time T 2 + Δ 2 is calculated (and the model is constructed from these information and the safety decision verification is performed after time T 2 + N 2, if the safety decision verification fails, the safety decision is continued until time T 2 + N 2 + 2 is calculated (N 2).
And 4) starting a braking device to avoid collision.
Claims (7)
1. CPS anti-collision control method based on differential dynamic logic is characterized in that: the method comprises the following steps:
step 1: performing information interaction, acquisition and processing through CPS node equipment, performing anti-collision decision verification based on differential dynamic logic, predicting the safety of the system at a certain future moment if the verification is successful, and performing tentative forward operation and prompting and alarming if the verification fails or a danger is predicted;
Step 2: adopting system suggestions, and screening a static optimal decision instruction according to a certain principle or acquiring a dynamic optimal decision instruction based on differential dynamic logic;
And step 3: rejecting system suggestion, and judging whether the system takes over or continues to prompt and alarm according to the triggering condition of the minimum distance event trigger;
And 4, step 4: and starting a braking device to avoid collision.
2. the differential dynamic logic based CPS collision avoidance control method according to claim 1, wherein: the specific process for judging the system security in the step 1 is as follows:
Step 1.1: performing modeling and anti-collision decision verification based on differential dynamic logic;
Step 1.2: if the anti-collision decision verification is successful, modeling and safety decision verification are carried out on a certain future moment, namely the safety of the system at the certain future moment is predicted;
Step 1.3: if the anti-collision decision verification fails or collision danger is predicted, starting a time trigger, displaying an acceptance button and a rejection button through a screen, and simultaneously performing voice alarm; in the case of a moving obstacle, the two or more parties cooperate to avoid the collision.
3. The differential dynamic logic based CPS collision avoidance control method according to claim 1, wherein: the specific process for screening the optimal decision instruction in the step 2 is as follows:
Step 2.1: firstly, applying a Markov decision process optimization system instruction list to a current early warning event to screen out a plurality of ideal decision instructions, then calculating respective expected values of the ideal decision instructions by applying a dynamic decision algorithm, and finally selecting the instruction with the maximum expected value;
step 2.2: and if the static optimal decision instruction fails to be screened, updating information and constructing a model, repeatedly adjusting the target speed and repeatedly carrying out safety test verification until the verification is successful, and assigning the acquired collision avoidance target speed to the dynamic instruction.
4. The differential dynamic logic based CPS collision avoidance control method according to claim 1, wherein: the specific process for screening the super decision instruction in the step 3 is as follows:
step 3.1: updating information and modeling in real time;
Step 3.2: if the minimum distance event trigger is not triggered, real-time modeling and anti-collision decision verification are repeatedly carried out until the verification fails, prompting and alarming are carried out at the moment, and otherwise CPS node acquisition information is returned;
Step 3.3: and triggering a minimum distance event trigger, carrying out real-time modeling and anti-collision decision verification, taking over by the system if the verification fails, otherwise predicting the safety of the system in a certain continuous or discontinuous time period, and taking over by the system if the collision danger is predicted.
5. The differential dynamic logic based CPS collision avoidance control method according to any of claims 1-4, wherein: the system comprises the following components: the system comprises a control center, CPS node equipment and an actuator, wherein the control center is responsible for receiving information and sending instructions, the CPS node equipment is responsible for collecting, processing, interacting and forwarding the information, and the actuator is responsible for steering, accelerating and decelerating.
6. the differential dynamic logic based CPS collision avoidance control method of claim 5, wherein: the CPS node device includes: the system comprises an acceleration and speed sensor for measuring acceleration and speed of the system, a laser ranging sensor for measuring the distance between the system and a short-distance obstacle, an interrogator for acquiring information of the long-distance moving obstacle and a transponder.
7. The differential dynamic logic based CPS collision avoidance control method of claim 2, wherein: the obstacle in step 1.3 is a moving obstacle, and at this time, collision is avoided by cooperation of two or more parties.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710609623.2A CN107561932B (en) | 2017-07-25 | 2017-07-25 | CPS anti-collision control method based on differential dynamic logic |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710609623.2A CN107561932B (en) | 2017-07-25 | 2017-07-25 | CPS anti-collision control method based on differential dynamic logic |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107561932A CN107561932A (en) | 2018-01-09 |
| CN107561932B true CN107561932B (en) | 2019-12-10 |
Family
ID=60974583
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710609623.2A Active CN107561932B (en) | 2017-07-25 | 2017-07-25 | CPS anti-collision control method based on differential dynamic logic |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107561932B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111951610B (en) * | 2020-08-31 | 2021-09-28 | 上海海事大学 | Ship-bridge collision avoidance active anti-collision intelligent early warning system and method |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9014915B2 (en) * | 2011-07-25 | 2015-04-21 | GM Global Technology Operations LLC | Active safety control for vehicles |
| CN102426522B (en) * | 2011-10-28 | 2014-12-10 | 东南大学 | CPS (Cyber Physical Systems) modeling and verification method based on transformation from Hybrid UML (Unified Modeling Language) to DAP (Differential-Algebraic Program) |
| CN102426521B (en) * | 2011-10-28 | 2014-04-16 | 东南大学 | CPS (Cyber Physical Systems) adaptability verification method based on Hybrid UML (Unified Modeling Language) and theorem proving |
| CN102722593B (en) * | 2011-10-28 | 2014-08-06 | 东南大学 | Cyber physical system (CPS) attribute verification method based on differential algebra timing sequence dynamic logic (DATL) |
| CN105321362B (en) * | 2015-10-30 | 2017-10-13 | 湖南大学 | A kind of intelligent coordinated passing method of intersection vehicles |
| CN106681322B (en) * | 2016-12-21 | 2020-03-13 | 华东师范大学 | Ground autonomous mobile robot safety navigation method based on formal description |
-
2017
- 2017-07-25 CN CN201710609623.2A patent/CN107561932B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN107561932A (en) | 2018-01-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111507372B (en) | Method and apparatus for multi-agent sensor in performing collaborative driving | |
| EP3361466B1 (en) | Risk-based driver assistance for approaching intersections of limited visibility | |
| US10437247B2 (en) | Multi-stage operation of autonomous vehicles | |
| US11273826B2 (en) | Vehicle control system, vehicle control method, and storage medium | |
| US10011276B2 (en) | Collision avoidance device | |
| US8788134B1 (en) | Autonomous driving merge management system | |
| US9358976B2 (en) | Method for operating a driver assistance system of a vehicle | |
| US11543825B2 (en) | Human supervision of an automated driving system | |
| US20210221402A1 (en) | Prediction control device | |
| JP2020166667A (en) | Vehicle control system | |
| CN110023168A (en) | Vehicle control system, control method for vehicle and vehicle control program | |
| US10836254B2 (en) | Vehicle control system, vehicle control method, and vehicle control program | |
| WO2023101717A9 (en) | Devices and methods for assisting operation of vehicles based on situational assessment fusing expoential risks (safer) | |
| CN102509476A (en) | Short-period flight conflict relieving method | |
| CN107561932B (en) | CPS anti-collision control method based on differential dynamic logic | |
| WO2023201563A1 (en) | Control method and apparatus, and means of transportation | |
| BE1028777B1 (en) | System and method for detecting inconsistencies in the outputs of perception systems of autonomous vehicles | |
| Tarish et al. | Proposed collision avoidance system in driverless cars | |
| EP2232297B1 (en) | Tracking coordinator for air-to-air and air-to- ground tracking | |
| KR101571972B1 (en) | Aircraft Collision Prediction system | |
| CN116761998A (en) | Safety test method of vehicle system and vehicle for testing | |
| Zeng et al. | Novel vehicle crash risk detection based on vehicular sensory system | |
| Nishizaki et al. | Process of judgment error on collision risk by modeling judgment patterns of officers | |
| CN104842877B (en) | Based on the self-propelled vehicle Emergency avoidance platform of the communication technology | |
| CN106515580A (en) | Method and system for implementing security control |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| CB03 | Change of inventor or designer information |
Inventor after: Qiu Jianpeng Inventor after: Duan Pengfei Inventor after: Zhou Ying Inventor before: Li Bixin Inventor before: Qiu Jianpeng Inventor before: Duan Pengfei Inventor before: Zhou Ying |
|
| CB03 | Change of inventor or designer information | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |