Disclosure of Invention
In order to solve the problem that the intelligent road cone in the prior art has single function and can not intelligently adjust the warning strategy according to the real-time road condition and construction environment change, the invention provides the intelligent road cone traffic early warning system which comprises an intelligent road cone, a control center and a user terminal;
the intelligent road cone is used for being deployed in a road construction area to warn vehicles, collecting road environment information through a sensor, and carrying out risk assessment and early warning on the vehicles according to the detected vehicle information;
the control center is used for integrating the monitoring information of the intelligent road cone and dynamically adjusting the placement scheme and the early warning strategy of the intelligent road cone;
And the user terminal is used for receiving the early warning information from the intelligent road cone and the control center and sending a warning to a user.
Further, the intelligent road cone comprises a sensing module, a data processing module, a communication module, an early warning module and a mobile module;
the sensing module is used for acquiring the real-time position information of the intelligent road cone and sensing the surrounding environment information in real time, wherein the surrounding environment information comprises the surrounding intelligent road cone position, lane line information and traffic signs;
the data processing module is used for processing and analyzing the original perception data acquired by the perception module and judging potential risks;
The communication module is used for communicating with the control center, uploading sensing data and risk information, receiving control instructions, and simultaneously carrying out real-time data interaction with other intelligent road cones to share vehicle information and self state;
the early warning module is used for carrying out differential early warning on the vehicles entering the early warning range according to the data processing result;
the mobile module is used for automatically moving to a road appointed position according to the control instruction acquired by the communication module.
Further, the judging of the potential risk by the data processing module specifically includes:
S101, defining fuzzy sets for each key vehicle behavior, wherein the key vehicle behavior comprises lane changing, acceleration, deceleration and normal running, and setting corresponding membership functions for the fuzzy sets;
S102, the data processing module calculates membership of each vehicle behavior feature to the fuzzy set based on the vehicle information acquired by the sensing module in real time;
S103, based on a fuzzy rule base, reasoning the comprehensive membership of the current state of the vehicle to various vehicle behaviors;
s104, converting the comprehensive membership of the current vehicle state into definite behavior judgment by using a maximum membership method;
s105, screening the vehicle behaviors with the comprehensive membership exceeding the behavior recognition threshold according to the behavior recognition threshold;
And S106, carrying out priority ranking on the vehicle behaviors corresponding to the comprehensive membership according to the comprehensive membership, and generating comprehensive behavior description.
Further, the control center comprises a data receiving module, a traffic situation sensing module, an early warning strategy optimizing module and an instruction sending module;
The data receiving module is used for receiving perception data, state information and risk information monitored by the intelligent road cone, and preprocessing the perception data and the state information;
The traffic situation awareness module is used for accessing real-time road condition data of a traffic management department and carrying out prediction analysis on traffic flow of a construction road section by combining with the historical road condition data;
The early warning strategy optimization module is used for dynamically adjusting the placement scheme and the early warning strategy of the intelligent road cone according to real-time road conditions, traffic prediction results and risk information;
The instruction sending module is used for sending the placement scheme and the early warning strategy of the intelligent road cone to the intelligent road cone and sending early warning information to the user terminal.
Further, the predicting and analyzing the traffic flow of the construction road section includes:
s201, the traffic situation awareness module receives real-time road condition information and historical traffic information acquired by a traffic management department;
s202, performing data cleaning, standardization and time alignment operation on the received data;
s203, extracting traffic characteristics from the data, wherein the traffic characteristics comprise the number of lanes, the width, the traffic flow change rule and the speed limit of a road construction area;
s204, the traffic situation awareness module predicts the traffic flow change trend of the construction area in future time by adopting a machine learning algorithm and combining traffic characteristics extracted from real-time road condition information and historical traffic information;
S205, according to the traffic flow change trend and the traffic characteristics of the road construction area, estimating the congestion risk and the accident risk of each lane by setting risk factors.
Further, the early warning strategy optimization module dynamically adjusts the early warning strategy specifically includes:
s301, establishing a multi-factor model according to the traffic flow prediction result of the traffic situation awareness module and the characteristics and safety requirements of a road construction area, and calculating the risk coefficient of each road section;
s302, carrying out standardized processing on risk coefficients of different road sections, dividing the road sections in a road construction area into different groups according to the risk data by using a clustering algorithm, and determining risk grades;
s303, the early warning strategy optimization module calculates the early warning strategy priority of each intelligent road cone in real time according to the risk information and the vehicle position information uploaded by the intelligent road cone, the risk area dividing result and the storage information in the early warning rule base;
S304, selecting an early warning strategy with the highest risk score as an execution strategy of the intelligent road cone at the current moment, and issuing the execution strategy to the corresponding intelligent road cone.
Further, the pre-warning strategy priority of each intelligent road cone is calculated in real time, namely, the risk scores of all pre-warning strategies corresponding to each intelligent road cone are calculated, and the risk scores are calculatedIs calculated as follows:
;
Wherein,Risk scoring of the early warning strategy corresponding to each intelligent road cone; Risk coefficients set for different risk level areas; the rule matching degree is used for representing the matching degree of the current traffic condition and the early warning rule; And the weight coefficient is used for adjusting the importance degree of different early warning rules.
Further, the early warning strategy optimization module further comprises a scheme for dynamically adjusting the placement of the intelligent road cone:
s401, initially planning the initial layout of the intelligent road cone by using a genetic algorithm according to the characteristics of a road construction area, safety requirements and traffic flow change trend before the road construction is started;
s402, calculating an optimal deployment path of each intelligent road cone by using an ant colony algorithm based on the initial layout, and generating an intelligent road cone deployment scheme;
s403, deploying the intelligent road cone based on the intelligent road cone deployment scheme, monitoring the deployment process in real time, and dynamically adjusting the deployment strategy through a reinforcement learning algorithm until the intelligent road cone is deployed at a corresponding position;
s404, after construction is finished, the early warning strategy optimization module simulates different recovery sequences and paths by using a deep reinforcement learning algorithm according to the construction completion condition and the real-time traffic condition.
Further, the user terminal comprises a data communication module and a user interface;
the data communication module is used for receiving early warning information sent by the intelligent road cone and the control center;
the user interface is used for displaying the early warning information received by the data communication module.
Further, the intelligent road cone traffic early warning method comprises the following steps:
S1, determining an intelligent road cone arranging scheme according to a construction position and traffic flow at the beginning of road construction;
s2, the intelligent road cone automatically moves to a target position along a route planned by the intelligent road cone swing and placement scheme according to a control instruction sent by a control center, and deployment early warning is carried out;
s3, the intelligent road cone detects the distance between the front vehicle and the rear vehicle in real time through the sensing module and the communication module, judges the risk coefficient between the vehicles, and adopts early warning when the risk is overlarge;
s4, when the vehicle enters the early warning range, updating early warning measures in real time, and continuously monitoring traffic flow and road conditions by a control center and adjusting an early warning strategy in real time;
And S5, after the road construction is completed, the control center updates the early warning rules, controls the intelligent road cone to gradually retract and recycle, and also maintains early warning operation on surrounding vehicles in the intelligent road cone recycling process.
Advantageous effects
1. Meanwhile, by knowing the specific behavior of the vehicle, a control center can evaluate which behavior can cause higher risk more accurately, thereby optimizing the sending time and content of the alarm;
2. According to the method, the control center integrates risk assessment information, traffic situation awareness information and historical data from each intelligent road cone, the overall risk level of a current construction road section can be judged more accurately, an early warning strategy can be adjusted in a targeted mode, when the risk level of a certain area is high, a vehicle driver can be reminded of safety by increasing the early warning frequency and the early warning intensity of the intelligent road cone of the area, and when the risk level of a certain time is generally increased, the vehicle can be guided to bypass or decelerate by issuing early warning information in advance, so that congestion and accidents are avoided.
Detailed Description
The invention is further described below with reference to the drawings and examples.
As shown in figure 1, the invention discloses an intelligent road cone traffic early warning system which comprises an intelligent road cone, a control center and a user terminal.
In this embodiment, the intelligent road cone is configured to be deployed on a road construction or an accident site to warn a vehicle, collect road environment information through a sensor, and perform risk assessment and early warning on the vehicle according to detected vehicle information. In this embodiment, as shown in fig. 3, the intelligent road cone is made of green environment-friendly materials, and comprises a shell 1, a warning lamp 2, a base 3 and a roller 4, wherein the shell 1 is integrally conical, the upper part is narrow, the lower part is wide, the stability of the intelligent road cone is effectively improved, the warning lamp 2 is located around the shell 1 and used for warning and prompting a vehicle through vivid lamplight colors, the base 3 is located at the bottom of the shell 1, the additional functional module of the intelligent road cone is deployed in the base 3, and meanwhile, the roller 4 is installed at the lower side of the base 3, so that the intelligent road cone automatically moves to a designated position after receiving the deployment and recovery scheme of a control center.
The intelligent road cone comprises a sensing module, a data processing module, a communication module, an early warning module and a mobile module.
The sensing module is provided with various sensors and is used for acquiring real-time position information of the intelligent road cone and real-time sensing surrounding environment information, wherein the surrounding environment information comprises but is not limited to positions, lane line information and traffic marks of other surrounding intelligent road cones, and meanwhile, the sensing module also comprises the step of acquiring vehicle information, and the vehicle information comprises but is not limited to vehicle positions, speeds, accelerations, running tracks, vehicle types and the like.
The data processing module is used for processing and analyzing the original perception data acquired by the perception module and judging the potential risk.
Specifically, the data processing module determines risk by identifying vehicle behavior, specifically including:
S101, defining fuzzy sets for each key vehicle behavior, wherein the key vehicle behavior comprises, but is not limited to lane changing, acceleration, deceleration, too close of a vehicle distance, normal running and the like, setting corresponding membership functions for the fuzzy sets, namely, when the key vehicle behavior is too close of the vehicle distance, judging indexes of the key vehicle behavior are the vehicle distance, so that the fuzzy sets can be set to be { close, safe and far }, and the membership functions of the fuzzy sets can be calculated by adopting one of Gaussian, triangular or trapezoidal types.
S102, the data processing module calculates membership of each vehicle behavior feature to the fuzzy set based on the vehicle information acquired by the sensing module in real time. The critical vehicle behavior with too close a distance as set in S101, the too close membership calculation may use a gaussian membership function:
;
Wherein,For the actual distance between the vehicles to be measured,In order to ensure that the distance between vehicles is safe,To adjust the standard deviation.
S103, based on the fuzzy rule base, reasoning the comprehensive membership of the current state of the vehicle to various behaviors. The fuzzy rule base is used for describing risk levels corresponding to different behavior combinations, and the reasoning process can be that if the vehicle speed is high and the vehicle distance is very close, the risk level is extremely high.
S104, converting the comprehensive membership of the current vehicle state into explicit behavior judgment by using a maximum membership method, namely selecting the maximum value in the comprehensive membership as a judgment result.
And S105, screening the vehicle behaviors with the membership degree exceeding the behavior recognition threshold according to the behavior recognition threshold. After selecting the maximum value of the comprehensive membership degrees as the judgment result in the step S104, the rest comprehensive membership degrees still need to be calculated, and the behaviors with the comprehensive membership degrees exceeding a certain standard are screened out by setting a behavior identification threshold.
And S106, carrying out priority ranking on the vehicle behaviors corresponding to the membership according to the membership, and generating comprehensive behavior description, wherein the behavior with high membership is more urgent or important under the current condition, so that the behavior needs to be responded first, and other behaviors in the comprehensive behavior description can help the intelligent road cone and the control center to determine the next action.
Specifically, the intelligent road cone can realize differential early warning by identifying the behavior of the vehicle, rather than simply directly starting an alarm after detecting that the vehicle enters the monitoring range. Because different vehicle behaviors, including but not limited to acceleration, deceleration, and lane changes, require different warning response times, vehicles approaching a construction area, for example, require more urgent and noticeable warnings, while slow-running vehicles may require milder warnings. The intelligent road cone is used for identifying the behavior of the vehicle, so that the accident risk can be accurately judged, false alarm or too frequent alarm can be avoided, and a control center can evaluate which behavior can cause higher risk more accurately by knowing the specific behavior of the vehicle, so that the sending time and content of the alarm are optimized.
The communication module is used for carrying out real-time data interaction with other intelligent road cones, sharing vehicle information and self state, simultaneously communicating with the control center, uploading sensing data, receiving control instructions, uploading risk assessment information of the data processing module to the control center, and enabling the control center to master more comprehensive road condition information.
And the early warning module is used for carrying out differential early warning on the vehicles entering the early warning range through various modes such as lamplight flickering, voice broadcasting, V2X communication and the like according to the data processing result.
And the mobile module is used for automatically moving to the specified position of the road according to the optimal placement scheme of the intelligent road cone acquired by the communication module.
Specifically, after the intelligent road cone detects a vehicle in a monitoring range and acquires vehicle information of the vehicle, the behavior of the vehicle is identified through the data processing module, the risk level is estimated, the early warning module carries out preliminary early warning on the vehicle through the data processing module, meanwhile, the intelligent road cone sends the vehicle information to the control center in real time through the data communication module, the control center further analyzes the vehicle risk to judge whether a further early warning strategy is needed, and if so, the intelligent road cone carries out further early warning and reminding on the vehicle.
Therefore, in this embodiment, the control center is configured to integrate monitoring information of the intelligent road cone, and dynamically adjust a placement scheme and an early warning strategy of the intelligent road cone. The control center comprises a data receiving module, a traffic situation sensing module, an early warning strategy optimizing module and an instruction sending module.
The data receiving module is used for receiving the perception data, the state information and the risk information which are monitored by the intelligent road cone, and carrying out preprocessing operations such as cleaning, filtering and integrating on the perception data and the state information.
The traffic situation awareness module is used for accessing real-time road condition data of a traffic management department, grasping global traffic conditions of the current road, and carrying out predictive analysis on traffic flow of a construction road section by combining historical road condition data and a machine learning algorithm. The prediction analysis of the traffic flow of the construction road section comprises the following steps:
s201, the traffic situation awareness module receives real-time road condition information and historical traffic information acquired by a traffic management department;
s202, performing data cleaning, standardization and time alignment operation on the received data;
S203, extracting traffic characteristics from the data, wherein the traffic characteristics comprise, but are not limited to, identifying lane data of a construction area and the width of each lane based on the position information and lane line information of an intelligent road cone, analyzing the traffic flow change rule of the construction area in different time periods based on historical traffic flow data, identifying speed limit and lane change prohibition areas of the road construction area based on traffic sign information and traffic rules and the like;
s204, the traffic situation awareness module predicts the traffic flow change trend of the construction area in future time by adopting a machine learning algorithm and combining traffic characteristics extracted from real-time road condition information and historical traffic information;
S205, according to the traffic flow change trend and the traffic characteristics of the road construction area, setting risk factors, distributing weights for the risk factors to calculate comprehensive risk scores, and evaluating the congestion risk and accident risk of each lane according to the comprehensive risk scores in a preset threshold range.
The control center integrates risk assessment information, traffic situation awareness information and historical data from each intelligent road cone, can judge the overall risk level of the current construction road section more accurately, and adjusts an early warning strategy in a targeted mode, when the risk level of a certain area is high, the early warning frequency and the early warning intensity of the intelligent road cone of the area can be increased to remind a vehicle driver of paying attention to safety, and when the risk level generally rises in a certain time, the early warning information can be issued in advance to guide the vehicle to detour or decelerate, so that congestion and accidents are avoided.
Specifically, a traffic situation sensing module is added in the control center, and the control center can adjust the placement scheme of the intelligent road cone and the early warning strategy for the vehicle in advance by predicting traffic flow, congestion situation and the like, so that the vehicle is guided to change the road or reduce the speed in advance, congestion and accidents are avoided, and the intelligent road cone is not passively responded when the risk occurs.
And the early warning strategy optimization module is used for dynamically adjusting the placement scheme and the early warning strategy of the intelligent road cone according to the real-time road condition, the traffic prediction result and the risk information. The early warning strategy optimization module dynamically adjusts the early warning strategy specifically comprises the following steps:
S301, building a multi-factor model according to the traffic flow prediction result of the traffic situation awareness module and the characteristics and safety requirements of the road construction area, and calculating the risk coefficient of each road section.
The multi-factor model comprehensively considers factors including but not limited to traffic flow, vehicle speed, road geometry, weather conditions and the like, gives weight to each factor according to the influence degree of the factors on traffic safety, then carries out weighted summation, and calculates the risk score of each road section:
;
Wherein,For the risk score of each road segment,For the weight coefficient of each factor,To be the factor after quantizationIs a numerical value of (2).
And S302, carrying out standardized processing on risk coefficients of different road sections, dividing the road sections in the road construction area into different groups according to the risk data by using a clustering algorithm, and determining risk grades.
S303, the early warning strategy optimization module calculates the early warning strategy priority of each intelligent road cone in real time according to the risk information and the vehicle position information uploaded by the intelligent road cone, the risk area dividing result and the storage information in the early warning rule base.
The priority of the early warning strategy of each intelligent road cone is calculated, namely the risk scores of all the early warning strategies corresponding to each intelligent road cone are calculated, and the risk scores are calculatedThe calculation formula of (2) is as follows:;
Wherein,Risk scoring of the early warning strategy corresponding to each intelligent road cone; Risk coefficients set for different risk level areas; The weight coefficient is used for adjusting the importance degree of different early warning rules; The rule matching degree is used for representing the matching degree of the current traffic condition and the early warning rule.
In risk scoring, rule matching degreeThe matching degree of the current traffic condition and a specific strategy stored in the early warning rule base is measured, so that a plurality of factors are needed to be considered in calculation, and firstly, characteristic values related to the early warning strategy are extracted from real-time data, including but not limited to the current speed of the vehicleIntelligent road cone distance of vehicleDensity of traffic flow in current laneAnd the current occupancy of the lane;
The characteristic values of different units and magnitudes are then normalized to a uniform range, i.e., to within the [0,1] range, to ensure that the different indicators are comparable in calculation, e.g., the current speed of the vehicleWithin the range of [0,1], a specific unified approach may beWhereinFor the value after the current speed of the vehicle is unified,A vehicle current speed minimum speed value for setting a speed for a current road,Setting a maximum speed value of the current speed of the vehicle at the speed for the current road;
then weighting and summing the standardized characteristic values to obtain the rule matching degree:
;
Wherein,The preset weight is used for representing the importance of the characteristic value; For the matching degree required by the speed characteristic value and the early warning rule base, interpolation reciprocal or normal distribution can be adopted for calculation, and the calculation of other characteristic values is the same; for the current speed warning threshold of the vehicle specified in the warning rule base,And the warning threshold value is the warning threshold value of the distance between the vehicle and the intelligent road cone in the warning rule.
S304, selecting an early warning strategy with the highest risk score as an execution strategy of the intelligent road cone at the current moment, and issuing the execution strategy to the corresponding intelligent road cone.
The early warning strategy optimization module dynamically adjusts the placement scheme and the playback scheme of the intelligent road cone before the road construction starts and ends, and specifically comprises the following steps:
S401, initially planning an initial layout of the intelligent road cone by using a genetic algorithm according to the characteristics of a road construction area, safety requirements and traffic flow change trend before road construction begins, wherein the initial layout comprises, but is not limited to, specific position coordinates, placement angles and intervals of each intelligent road cone.
And S402, converting the initial layout into specific coordinates and angles of each formulated road cone, calculating an optimal deployment path of each intelligent road cone by using an ant colony algorithm, minimizing deployment time and influence on traffic, determining a staged deployment strategy, and combining the initial layout, the optimal deployment path and the staged deployment strategy to generate an intelligent road cone deployment scheme.
S403, deploying the intelligent road cone based on the intelligent road cone deployment scheme, monitoring the deployment process in real time, and dynamically adjusting the deployment strategy through a reinforcement learning algorithm to cope with the emergency until the intelligent road cone is deployed at the corresponding position.
S404, after the road construction is finished, the early warning strategy optimization module simulates different recovery sequences and paths by using a deep reinforcement learning algorithm according to the construction completion condition and the real-time traffic condition, and prepares a safe and efficient intelligent road cone recovery scheme, and generates a safe recovery route for each intelligent road cone.
Specifically, in the recycling process of the intelligent road cone, a dynamic cluster of the intelligent road cone is formed through a cooperative algorithm, the intelligent road cone detects surrounding environment information in real time through a sensing module and realizes surrounding environment information sharing among all modules through a communication module, and the safety recycling index of each intelligent road cone is calculated and updated through a data processing module, so that the recycling action is coordinated, and the safety of the recycling process is ensured.
The intelligent road cone automatically moves according to the appointed route and sequence after receiving the deployment scheme or the playback scheme of the control center, and simultaneously, the intelligent road cone starts to monitor the vehicle and the vehicle behavior in the road in real time in the automatic moving process and transmits the vehicle and the vehicle behavior to the control center, and the intelligent road cone is prevented from being unnoticed by vehicles in the deployment or recovery process by dynamically adjusting the early warning strategy.
And the instruction sending module is used for sending the early warning strategy and the control instruction to the intelligent road cone.
Specifically, when the control center judges that the vehicle needs to be further warned and reminded, the intelligent road cone communicates with infrastructure in the vehicle through a V2X technology, and warning information of the control center is sent to the user terminal.
And the user terminal is used for receiving the early warning information from the intelligent road cone and the control center, sending an early warning prompt to a user in a voice and image mode and bypassing the road construction area according to the received early warning information. The user terminal comprises a data communication module and a user interface;
And the data communication module is used for receiving early warning information sent by the intelligent road cone and the control center.
And the user interface is used for displaying the early warning information received by the data communication module.
In this embodiment, an intelligent road cone traffic early warning method is further included, as shown in fig. 2, including:
S1, determining an intelligent road cone placing scheme according to the road construction position and traffic flow at the beginning of road construction, wherein the intelligent road cone placing scheme needs to ensure that an intelligent road cone reasonably covers a lane to provide accurate early warning.
S2, the intelligent road cone automatically moves to a target position along a route planned by the intelligent road cone swing and placement scheme according to a control instruction sent by a control center, deployment early warning is carried out, in the process that the intelligent road cone automatically moves to the target position according to the planned route, vehicles on a construction lane are decelerated and lane change reminded through measures such as voice warning, and vehicles on a normal lane are reminded of lane change of the vehicles on the side.
And S3, the intelligent road cone detects the distance between the front vehicle and the rear vehicle in real time through the sensing module and the communication module, judges the risk coefficient between the vehicles, adopts early warning when the risk is excessive, and provides different early warning information for the vehicles directly entering the early warning area and the subsequent traffic flow respectively so as to ensure safe driving.
And S4, when the vehicle enters the early warning range, updating early warning measures in real time, and continuously monitoring traffic flow and road conditions by the control center and adjusting an early warning strategy in real time.
And S5, after the road construction is completed, the control center updates the early warning rules, controls the intelligent road cone to gradually retract and recycle, and also maintains the early warning operation on surrounding vehicles in the intelligent road cone recycling process so as to prevent the sudden appearance of the recycled vehicles and the sudden self-movement of the intelligent road cone from influencing traffic.
By the traffic early warning method, effective safety early warning and guiding of vehicles passing through a road construction area are ensured during construction of the expressway, accident risks are reduced, and traffic management efficiency is improved.
The foregoing examples have shown only the preferred embodiments of the invention, which are described in more detail and are not to be construed as limiting the scope of the invention. It should be noted that modifications, improvements and substitutions can be made by those skilled in the art without departing from the spirit of the invention, which are all within the scope of the invention. Accordingly, the scope of protection of the present invention is to be determined by the appended claims.