Intelligent traffic guiding device based on machine learning

Abstract

The invention relates to the technical field of traffic guidance, in particular to an intelligent traffic guidance device based on machine learning, which comprises an intelligent system and a guidance device body, and is characterized in that the intelligent system comprises the following modules: an acquisition module used for acquiring environment information; the analysis module is used for analyzing causes of the environment information to obtain an analysis result; the processing module is used for processing the environment information to obtain a processing result; the learning module is used for creating a machine learning model and training the machine learning model according to the processing result; the prediction module is used for predicting future traffic conditions through a machine learning model to obtain a prediction result and sending the prediction result to the control module; and the control module is used for adjusting the guiding strategy of the guiding device body according to the prediction result. According to the method, the machine learning model is established and trained, and the future traffic condition is predicted through the machine learning model, so that the traffic pressure is effectively relieved, and the road passing efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic guidance, and in particular to an intelligent traffic guidance device based on machine learning. Background Art

[0002] With the acceleration of urbanization, the urban population continues to increase, and the number of vehicles has increased dramatically, leading to increasingly serious traffic congestion problems. Traditional traffic management methods and technologies are difficult to cope with increasingly complex traffic conditions and cannot achieve real-time and efficient traffic scheduling, so machine learning technology came into being. Machine learning is a branch of artificial intelligence that enables computers to learn from data and improve their performance without explicit programming. The core of machine learning is to enable computer systems to learn from data through algorithms and statistical models, and make predictions or decisions based on what they have learned.

[0003] At present, most existing traffic guidance devices rely on pre-set routes and traffic signals, cannot be dynamically adjusted according to real-time traffic conditions, and still have the problem of insufficient flexibility.

[0004] In summary, most existing traffic guidance devices still have the problem of insufficient flexibility, which has become a difficult problem that needs to be solved urgently in this field. Therefore, it is necessary to propose an intelligent traffic guidance device based on machine learning. Summary of the invention

[0005] To solve the above problems, the present invention provides an intelligent traffic guidance device based on machine learning. By establishing and training a machine learning model and predicting future traffic conditions through the machine learning model, traffic pressure is effectively alleviated and road traffic efficiency is improved.

[0006] In order to achieve the above-mentioned purpose, the technical solution of the present invention is as follows: an intelligent traffic guidance device based on machine learning, including an intelligent system and a guidance device body, the intelligent system including a collection module, an analysis module, a processing module, a learning module, a prediction module and a control module.

[0007] The acquisition module is used to collect current environmental information and transmit the collected environmental information to the analysis module.

[0008] The analysis module is used to analyze the causes of environmental information, obtain analysis results, and transmit the analysis results to the processing module.

[0009] The processing module is used to process the collected environmental information according to the analysis result of the analysis module, obtain the processing result, and transmit the processing result to the learning module.

[0010] The learning module is used to create a machine learning model and train the machine learning model according to the processing results of the processing module.

[0011] The prediction module is used to predict future traffic conditions through a machine learning model, obtain prediction results, and send the prediction results to the control module.

[0012] The control module is used to adjust the guidance strategy of the guidance device body according to the prediction result.

[0013] Furthermore, the environmental information includes the current traffic volume and the traffic density per unit time.

[0014] Furthermore, when processing environmental information, the processing module first collects the external index of the data set through the Internet big model, and then performs a clustering operation on it. The specific formula of this process is as follows: D={x1,x2,…,x m}(1).

[0015] Where D is the external index of the data set, x1, x2 and x m All samples were collected.

[0016] During the clustering process, the clustering results are compared and evaluated with the reference model. The reference model is derived from the division results provided by experts, and its evaluation results are recorded as external indicators. At the same time, the clustering results are evaluated by direct evaluation, and its evaluation results are recorded as internal indicators.

[0017] The external indicator evaluation process is as follows: The clusters given by clustering are divided into , the cluster division given by the reference model is , pair the samples in pairs and make the following definitions: .

[0018] .

[0019] .

[0020] .

[0021] .

[0022] in, Representation sample The categories in the clustering model, Representation sample In the reference model, a represents the number of elements in the set SS. The SS set is contained in belong to the same cluster and in The sample pairs in belong to the same cluster, b represents the number of elements in the set SD, and the SD set is included in belong to the same cluster and in The sample pairs belonging to different clusters in , c represents the number of elements in the set DS, and the DS set is included in belong to different clusters and in The sample pairs in the same cluster belong to the same cluster, d represents the number of elements in the set DD, and the DD set is included in belong to different clusters and in The sample pairs in belong to different clusters, m represents the total number of samples, and i and j represent the sequence numbers of the samples.

[0023] .

[0024] Among them, JC is the Jaccard coefficient, which is in the range of [0,1]. The consistency between the clustering results and the actual situation is positively correlated with the Jaccard index.

[0025] .

[0026] Among them, RI is the Rand Index, which ranges from [0,1].

[0027] .

[0028] Among them, ARI is the adjusted Rand index, which ranges from [-1,1]. The consistency between the clustering results and the actual situation is positively correlated with the adjusted Rand index.

[0029] The internal indicator evaluation formula is as follows: .

[0030] Among them, SSE is the sum of squares of errors, which represents the sum of squares of the distance loss from the data in the class to the class center.

[0031] .

[0032] .

[0033] Where SC is the silhouette coefficient, represents the silhouette coefficient of the ith sample, Representation sample The average distance to other samples in the cluster, express The average distance of all samples in the nearest cluster; the range of SC is [-1, 1], and its value is positively correlated with the clustering result.

[0034] .

[0035] .

[0036] .

[0037] Among them, CH is the Calinski-Harabasz index, and is the covariance matrix of the data between and within clusters, and Represent the center points of cluster q and data set D respectively, represents the number of data sets belonging to cluster q, represents the trace of the matrix in the between-cluster covariance matrix, represents the trace of the matrix in the intra-cluster covariance matrix; the CH indicator represents the ratio between the distance between clusters and the distance within a cluster, and its ratio is positively correlated with the clustering effect; k represents the number of clusters.

[0038] The technical principles of the above scheme are as follows: The acquisition module first collects environmental information of the current environment and transmits the environmental information to the analysis module. The analysis module analyzes the environmental information to obtain the analysis results, and sends the analysis results to the processing module. The processing module processes the environmental information according to the analysis results to obtain the processing results. The processing module sends the processing results to the learning module, and establishes a machine learning model through the learning module. The machine learning model is trained according to the processing results. The prediction module predicts future traffic conditions through the machine learning model, obtains the prediction results, and sends the prediction results to the control module, which adjusts the strategy of the guidance device body.

[0039] The above scheme has the following beneficial effects: 1. The present invention can quickly identify changes in traffic conditions through real-time collection and analysis of current environmental information. At the same time, combined with the predictive ability of the machine learning model, it can predict possible traffic congestion or smooth conditions in advance, so as to adjust the guidance strategy in time, effectively alleviate traffic pressure and improve road traffic efficiency.

[0040] 2. The present invention can timely discover and warn potential traffic safety hazards, such as sudden increase in traffic volume or abnormal driving behavior, through real-time monitoring of environmental information. By taking measures in advance, the risk of traffic accidents can be effectively reduced.

[0041] 3. The present invention optimizes traffic guidance, reduces waiting time and unnecessary driving of vehicles, thereby reducing fuel consumption and tail gas emissions, improving air quality and protecting the environment.

[0042] Furthermore, the guiding device body includes a base plate, a fixing plate is fixedly connected to the top of the base plate, a control box is fixedly connected to the top of the fixing plate, a controller and an alarm are fixedly connected to the bottom wall of the control box, the controller is used to control the operation of the alarm, a plurality of cameras are fixedly connected to the base plate along the circumference of its side wall, the controller is used to control the operation of the camera, and a guiding component for displaying guiding information and a transmission component for driving the guiding component are provided on the top of the base plate.

[0043] Beneficial effects: The camera can monitor the surrounding traffic environment in real time to ensure a comprehensive understanding of the traffic conditions. Combined with the controller's control of the camera, it can accurately capture the traffic conditions in different areas. At the same time, when the guidance device itself is abnormal, the controller can quickly activate the alarm, send out a warning signal, and inform the staff to go for maintenance.

[0044] Furthermore, the transmission assembly includes a driving frame fixedly connected to one side of the base plate, a first driving member is fixedly connected to the bottom wall of the driving frame, a controller is used to control the operation of the first driving member, an output shaft of the first driving member is coaxially fixedly connected to a transmission gear, the transmission gear is meshed with a first gear ring, the first gear ring is rotatably matched with the top of the base plate, a side of the first gear ring away from the transmission gear is meshed with a plurality of moving wheels, a side of the moving wheel away from the first gear ring is meshed with a second gear ring, and the second gear ring is fixedly connected to the side wall of the fixed plate.

[0045] Beneficial effects: The transmission gear, the first gear ring, the moving wheel and the second gear ring form a stable transmission chain, so that the moving wheel can move or adjust its position according to a predetermined trajectory and speed.

[0046] Furthermore, the guide assembly includes a moving rod coaxially fixedly connected to the moving wheel, and a plurality of guide frames are arranged above the base plate. The tops of adjacent moving rods are rotatably matched with the bottoms of the adjacent guide frames, and the guide frames are provided with replacement components for changing different guide modes.

[0047] Beneficial effects: Replacing components can easily change different guidance modes according to actual needs or changes in traffic conditions. This flexibility ensures that the guidance components can adapt to different traffic environments and needs, and improves the pertinence and effectiveness of traffic guidance.

[0048] Furthermore, the replacement component includes a second driving member fixedly connected to the inner wall of the guide frame, the controller is used to control the operation of the second driving member, the output shaft of the second driving member is coaxially fixedly connected with a connecting rod, and the connecting rod is fixedly connected to a limiting block at one end away from the output shaft of the second driving member.

[0049] The inner wall of the guide frame is rotatably matched with a rotating shaft, a limiting wheel is coaxially fixedly connected to the rotating shaft, a plurality of limiting grooves are equidistantly provided on the limiting wheel, and the limiting blocks are slidably matched with the limiting grooves.

[0050] A number of first bevel gears are coaxially and equidistantly fixedly connected to the rotating shaft, the first bevel gears are all meshed with second bevel gears, the second bevel gears are coaxially and fixedly connected to a reversing rod, the reversing rod is fixedly connected to a guide column, and the end of the reversing rod away from the second bevel gear is rotatably matched with the top wall of the guide frame.

[0051] Beneficial effect: Driven by the second driving member, the connecting rod and the limit block can move accurately and engage with the limit groove on the limit wheel. This design ensures that the replacement process of the guide mode is both fast and accurate, and reduces the display errors of the guide information caused by improper replacement. At the same time, due to the meshing relationship between the limit block and the limit groove, and the linkage effect of the rotating shaft, the first bevel gear, the second bevel gear and the reversing rod, the entire replacement process is more stable and reliable, reducing the replacement failure caused by mechanical failure.

[0052] Furthermore, the guide pillars are all in the shape of triangular pillars.

[0053] Beneficial effects: The triangular column structure is relatively simple and easy to manufacture and process. At the same time, due to its regular shape, it is also easy to install and debug, thereby reducing the production cost and time of the traffic guide device and improving its economy and practicality.

[0054] Furthermore, the side walls of the guide columns are fixedly connected with guide plates representing different guide strategies.

[0055] Beneficial effects: The guide board can intuitively display different guidance strategies, such as direction instructions, lane allocation, speed limit, etc. The driver only needs to quickly glance to understand the current traffic guidance information, without the trouble of interpreting complex signs or symbols. This intuitiveness helps to improve the driver's attention and reaction speed, especially in heavy traffic or limited vision, so that the driver can quickly make the right driving decisions.

[0056] Furthermore, a plurality of universal wheels are fixedly connected to the bottom of the base plate.

[0057] Beneficial effects: The design of universal wheels makes the traffic guiding device more convenient during transportation. At the same time, during deployment, the staff can quickly adjust the position and angle of the device to adapt to different traffic environments and needs.

[0058] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a structural schematic diagram of the intelligent system of the present invention.

[0060] Figure 2 This is an axonometric schematic diagram of the intelligent traffic guidance device based on machine learning of the present invention.

[0061] Figure 3 This is an axonometric schematic diagram of replacing components in the intelligent traffic guidance device based on machine learning of the present invention.

[0062] Figure 4 It is a front cross-sectional schematic diagram of the driving frame in the intelligent traffic guidance device based on machine learning of the present invention.

[0063] The figure marks in the drawings of the specification include: 1, base plate; 2, fixed plate; 3, control box; 4, camera; 5, drive frame; 6, transmission gear; 7, first ring gear; 8, moving wheel; 9, second ring gear; 10, moving rod; 11, guide frame; 12, connecting rod; 13, limit block; 14, rotating shaft; 15, limit wheel; 16, first bevel gear; 17, second bevel gear; 18, reversing rod; 19, guide column; 20, guide plate; 21, universal wheel. DETAILED DESCRIPTION

[0064] The following is further described in detail through specific implementation methods: Embodiment 1: As attached Figure 1 and Figure 2 As shown: An intelligent traffic guidance device based on machine learning includes an intelligent system and a guidance device body, the intelligent system includes a collection module for collecting environmental information, an analysis module for analyzing environmental information, a processing module for processing environmental information, a learning module for establishing and training a machine learning model, a prediction module for predicting future traffic conditions, and a control module for adjusting the guidance strategy of the guidance device body.

[0065] The following is a detailed explanation of the functions of each module: The acquisition module is used to collect current environmental information and transmit the collected environmental information to the analysis module, wherein the environmental information includes the current vehicle flow and vehicle flow density per unit time.

[0066] The analysis module is used to analyze the causes of environmental information, obtain analysis results, and transmit the analysis results to the processing module.

[0067] The processing module is used to process the collected environmental information according to the analysis result of the analysis module, obtain the processing result, and transmit the processing result to the learning module.

[0068] The learning module is used to create a machine learning model and train the machine learning model according to the processing results of the processing module.

[0069] The prediction module is used to predict future traffic conditions through a machine learning model, obtain prediction results, and send the prediction results to the control module.

[0070] The control module is used to adjust the guidance strategy of the guidance device body according to the prediction result.

[0071] When processing environmental information, the processing module first collects the external index of the data set through the Internet big model, and then performs clustering operations on it. The specific formula of this process is as follows: D={x1,x2,…,x m}(1).

[0072] Where D is the external index of the data set, x1, x2 and x m All samples were collected.

[0073] During the clustering process, the clustering results are compared and evaluated with the reference model. The reference model is derived from the division results provided by experts, and its evaluation results are recorded as external indicators. At the same time, the clustering results are evaluated by direct evaluation, and its evaluation results are recorded as internal indicators.

[0074] The external indicator evaluation process is as follows: The clusters given by clustering are divided into , the cluster division given by the reference model is , pair the samples in pairs and make the following definitions: .

[0075] .

[0076] .

[0077] .

[0078] .

[0079] in, Representation sample The categories in the clustering model, Representation sample In the reference model, a represents the number of elements in the set SS. The SS set is contained in belong to the same cluster and in The sample pairs in belong to the same cluster, b represents the number of elements in the set SD, and the SD set is included in belong to the same cluster and in The sample pairs belonging to different clusters in , c represents the number of elements in the set DS, and the DS set is included in belong to different clusters and in The sample pairs in the same cluster belong to the same cluster, d represents the number of elements in the set DD, and the DD set is included in belong to different clusters and in The sample pairs in belong to different clusters, m represents the total number of samples, and i and j represent the sequence numbers of the samples.

[0080] .

[0081] Among them, JC is the Jaccard coefficient, which is in the range of [0,1]. The consistency between the clustering results and the actual situation is positively correlated with the Jaccard index.

[0082] .

[0083] Among them, RI is the Rand Index, which ranges from [0,1].

[0084] .

[0085] Among them, ARI is the adjusted Rand index, which ranges from [-1,1]. The consistency between the clustering results and the actual situation is positively correlated with the adjusted Rand index.

[0086] The internal indicator evaluation formula is as follows: .

[0087] Among them, SSE is the sum of squares of errors, which represents the sum of squares of the distance loss from the data in the class to the class center.

[0088] .

[0089] .

[0090] Where SC is the silhouette coefficient, represents the silhouette coefficient of the ith sample, Representation sample The average distance to other samples in the cluster, express The average distance of all samples in the nearest cluster; the range of SC is [-1, 1], and its value is positively correlated with the clustering result.

[0091] .

[0092] .

[0093] .

[0094] Among them, CH is the Calinski-Harabasz index, and is the covariance matrix of the data between and within clusters, and Represent the center points of cluster q and data set D respectively, represents the number of data sets belonging to cluster q, represents the trace of the matrix in the between-cluster covariance matrix, represents the trace of the matrix in the intra-cluster covariance matrix; the CH indicator represents the ratio between the distance between clusters and the distance within a cluster, and its ratio is positively correlated with the clustering effect; k represents the number of clusters.

[0095] The specific implementation process is as follows: the acquisition module first collects the vehicle flow in the current environment and the vehicle flow density per unit time as environmental information. In this embodiment, 5 minutes is used as a unit time. The acquisition module sends the collected environmental information to the analysis module, which analyzes the causes of the environmental information, obtains the analysis results, and sends the analysis results to the processing module, which processes the environmental information.

[0096] During the processing, the processing module first collects the external index of the data set through the Internet big model, and then performs clustering operation on the environmental information. The specific formula of this process is as follows: D={x1,x2,…,x m}(1).

[0097] Where D is the external index of the data set, x1, x2 and x m All samples were collected.

[0098] At this time in the clustering process, a reference model is first established based on the division results provided by experts, and the clustering results are compared and evaluated with the reference model, and the evaluation results are recorded as external indicators. At the same time, a direct evaluation method is also used to directly evaluate the clustering results without relying on the reference model, and the evaluation results are recorded as internal indicators.

[0099] The external indicator evaluation process is as follows: The clusters given by clustering are divided into , the cluster division given by the reference model is , pair the samples in pairs and make the following definitions: .

[0100] .

[0101] .

[0102] .

[0103] .

[0104] in, Representation sample The categories in the clustering model, Representation sample In the reference model, a represents the number of elements in the set SS. The SS set is contained in belong to the same cluster and in The sample pairs in belong to the same cluster, b represents the number of elements in the set SD, and the SD set is included in belong to the same cluster and in The sample pairs belonging to different clusters in , c represents the number of elements in the set DS, and the DS set is included in belong to different clusters and in The sample pairs in the same cluster belong to the same cluster, d represents the number of elements in the set DD, and the DD set is included in belong to different clusters and in The sample pairs in belong to different clusters, m represents the total number of samples, and i and j represent the sequence numbers of the samples.

[0105] .

[0106] Among them, JC is the Jaccard coefficient, which is in the range of [0,1]. The consistency between the clustering results and the actual situation is positively correlated with the Jaccard index.

[0107] .

[0108] Among them, RI is the Rand Index, which ranges from [0,1].

[0109] .

[0110] Among them, ARI is the adjusted Rand index, and its interval is in the range of [-1,1]. The consistency between the clustering results and the actual situation is positively correlated with the adjusted Rand index. The larger the value, the more consistent the clustering results are with the actual situation.

[0111] The internal indicator evaluation formula is as follows: .

[0112] Among them, SSE is the sum of squares of errors, which represents the sum of squares of the distance loss from the data in the class to the class center.

[0113] .

[0114] .

[0115] Where SC is the silhouette coefficient, represents the silhouette coefficient of the ith sample, Representation sample The average distance to other samples in the cluster, express The average distance of all samples in the cluster closest to it; the range of SC is [-1, 1], and its value is positively correlated with the clustering result. The larger the value, the better the clustering effect. At this time, the distance between samples in the same cluster is smaller, while the distance between samples in different clusters is larger.

[0116] .

[0117] .

[0118] .

[0119] Among them, CH is the Calinski-Harabasz index, and is the covariance matrix of the data between and within clusters, and Represent the center points of cluster q and data set D respectively, represents the number of data sets belonging to cluster q, represents the trace of the matrix in the between-cluster covariance matrix, represents the trace of the matrix in the intra-cluster covariance matrix; the CH indicator represents the ratio between the distance between clusters and the distance within a cluster, and its ratio is positively correlated with the clustering effect; k represents the number of clusters.

[0120] After processing the environmental information, the processing module obtains the processing result and sends the processing result to the learning module. The learning module first creates a machine learning model and trains it through the processing result of the processing module. After the training is completed, the prediction module predicts the future traffic conditions through the machine learning model, obtains the prediction result, and sends the prediction result to the control module, which adjusts the guidance strategy of the guidance device body.

[0121] Embodiment 2: As attached Figure 2 , Figure 3 and Figure 4 As shown, the difference from Example 1 is that the guiding device body includes a base plate 1, a fixing plate 2 is integrally formed on the top of the base plate 1, a control box 3 is bolted fixedly connected to the top of the fixing plate 2, a controller and an alarm are bolted fixedly connected to the bottom wall of the control box 3, the controller is used to control the operation of the alarm, a plurality of cameras 4 are screwed fixedly connected to the base plate 1 along the circumferential side wall thereof, the controller is used to control the operation of the cameras 4, and a guiding component for displaying guiding information and a transmission component for driving the guiding component to operate are provided on the top of the base plate 1.

[0122] The transmission assembly includes a driving frame 5 welded to one side of the base plate 1, and the first driving member is fixedly connected to the bottom wall of the driving frame 5 by bolts. The controller is used to control the operation of the first driving member. In this embodiment, the first driving member is preferably a first stepper motor, and the output shaft of the first stepper motor is coaxially fixed with a transmission gear 6, and the transmission gear 6 is meshed with a first gear ring 7. The first gear ring 7 is rotatably matched with the top of the base plate 1, and a plurality of moving wheels 8 are meshed on the inner side of the first gear ring 7, and a second gear ring 9 is meshed on the inner side of the moving wheel 8, and the second gear ring 9 is welded to the side wall of the fixed plate 2.

[0123] The guide assembly includes a moving rod 10 coaxially fixedly connected to the moving wheel 8, and a plurality of guide frames 11 are arranged above the base plate 1. The tops of adjacent moving rods 10 are rotatably matched with the bottoms of the adjacent guide frames 11, and the guide frames 11 are provided with replacement components for changing different guide modes.

[0124] The replacement component includes a second driving member fixedly connected to the inner wall of the guide frame 11 by bolts, and the controller is used to control the operation of the second driving member. In this embodiment, the second driving member is a second stepper motor, and the output shaft of the second stepper motor is coaxially bolted and fixedly connected with a connecting rod 12, and a limiting block 13 is integrally formed at the lower end of the connecting rod 12.

[0125] The inner wall of the guide frame 11 is rotatably matched with a rotating shaft 14, and a limiting wheel 15 is coaxially fixedly clamped on the rotating shaft 14. The limiting wheel 15 is provided with a plurality of limiting grooves at equal intervals, and the limiting blocks 13 are slidably matched with the limiting grooves.

[0126] A number of first bevel gears 16 are coaxially and equidistantly fixed on the rotating shaft 14, each of the first bevel gears 16 is meshed with a second bevel gear 17, each of the second bevel gears 17 is coaxially and fixedly connected with a reversing rod 18, a triangular column-shaped guide column 19 is integrally formed on the reversing rod 18, the top end of the reversing rod 18 is rotatably matched with the inner top wall of the guide frame 11, and the side walls of the guide columns 19 are fixedly connected with guide plates 20 representing different guiding strategies by screws.

[0127] The specific implementation process is as follows: the staff starts the first stepper motor through the controller, the first stepper motor output shaft rotates, driving the transmission gear 6 to rotate, the transmission gear 6 drives the first gear ring 7 meshing with it to rotate, and the first gear ring 7 drives the moving wheel 8 meshing with it to rotate. Since the moving wheel 8 is meshing with the second gear ring 9, and the second gear ring 9 is welded to the side wall of the fixed plate 2, when the moving wheel 8 rotates, it will rotate around the second gear ring 9, and then the moving wheel 8 drives the moving rod 10 to rotate around the second gear ring 9. The moving rod 10 will drive the adjacent guide frame 11 to move during the movement process, and the staff can adjust the position of the guide frame 11 by controlling the start and stop of the first stepper motor.

[0128] When the guidance strategy needs to be adjusted, the staff starts the second stepper motor through the controller, the output shaft of the stepper motor rotates, driving the connecting rod 12 to rotate, and the connecting rod 12 drives the limit block 13 integrally formed therewith to rotate together. Since the limit wheel 15 is provided with a plurality of limit slots at equal intervals, the limit block 13 and the limit slots are slidably matched, and the limit block 13 will toggle the limit wheel 15 during the rotation process, and the limit wheel 15 drives the rotating shaft 14 to rotate, and the rotating shaft 14 drives the first bevel gear 16 to rotate, and the first bevel gear 16 drives the second bevel gear 17 to rotate, and the second bevel gear 17 drives the reversing rod 18 to rotate, and the reversing rod 18 drives the guide column 19 integrally formed therewith to rotate. At this time, since the side walls of the guide column 19 are all screw-fixedly connected with guide plates 20 representing different guidance strategies, the guide column 19 will drive the guide plate 20 to rotate together when rotating, and the staff can realize the conversion of the guidance strategy by rotating the guide plate 20.

[0129] The guiding device body collects information about the current environment's vehicle flow and vehicle flow density per unit time through the camera 4.

[0130] When the guidance device body is damaged, the alarm will sound an alarm in time to inform the staff to go for repair.

[0131] Embodiment 3: As attached Figure 2 As shown, the difference from the embodiment 2 is that a plurality of universal wheels 21 are fixedly connected with bolts at the bottom of the base plate 1 .

[0132] The specific implementation process is as follows: when the guiding device body needs to be moved, the design of the universal wheel 21 enables the staff to push it easily.

[0133] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

Claims

1. An intelligent traffic guidance device based on machine learning, characterized in that: It includes an intelligent system and a guiding device body, wherein the intelligent system includes a collection module, an analysis module, a processing module, a learning module, a prediction module and a control module; The acquisition module is used to collect current environmental information and transmit the collected environmental information to the analysis module; The analysis module is used to analyze the causes of environmental information, obtain analysis results, and transmit the analysis results to the processing module; The processing module is used to process the collected environmental information according to the analysis result of the analysis module, obtain the processing result, and transmit the processing result to the learning module; A learning module, used to create a machine learning model and train the machine learning model according to the processing results of the processing module; The prediction module is used to predict future traffic conditions through a machine learning model, obtain prediction results, and send the prediction results to the control module; The control module is used to adjust the guidance strategy of the guidance device body according to the prediction result.

2. The intelligent traffic guidance device based on machine learning according to claim 1, characterized in that: Environmental information includes the current traffic volume and traffic density per unit time.

3. The intelligent traffic guidance device based on machine learning according to claim 2 is characterized in that: When processing environmental information, the processing module first collects the external index of the data set through the Internet big model, and then performs clustering operations on it. The specific formula of this process is as follows: D={x1,x2,…,x m }(1); Where D is the external index of the data set, x1, x2 and x m All are samples collected; In the clustering process, the clustering results are compared and evaluated with the reference model. The reference model is obtained by the division results provided by experts, and its evaluation results are recorded as external indicators. At the same time, the clustering results are evaluated by direct evaluation, and its evaluation results are recorded as internal indicators. The external indicator evaluation process is as follows: The clusters given by clustering are divided into , the cluster division given by the reference model is , pair the samples in pairs and make the following definitions: ； ； ； ； ； in, Representation sample The categories in the clustering model, Representation sample In the reference model, a represents the number of elements in the set SS. The SS set is contained in belong to the same cluster and in The sample pairs in belong to the same cluster, b represents the number of elements in the set SD, and the SD set is included in belong to the same cluster and in The sample pairs belonging to different clusters in , c represents the number of elements in the set DS, and the DS set is included in belong to different clusters and in The sample pairs in the same cluster belong to the same cluster, d represents the number of elements in the set DD, and the DD set is included in belong to different clusters and in The sample pairs belonging to different clusters in , m represents the total number of samples, i and j represent the sequence numbers of the samples; ； Among them, JC is the Jaccard coefficient, which is in the range of [0,1]. The consistency between the clustering results and the actual situation is positively correlated with the Jaccard index; ； Among them, RI is the Rand Index, which ranges from [0,1]; ； Among them, ARI is the adjusted Rand index, which is in the range of [-1,1]. The consistency between the clustering results and the actual situation is positively correlated with the adjusted Rand index; The internal indicator evaluation formula is as follows: ; Among them, SSE is the sum of squares of errors, which represents the sum of squares of the distance loss from the data in the class to the class center; ； ； Where SC is the silhouette coefficient, represents the silhouette coefficient of the ith sample, Representation sample The average distance to other samples in the cluster, express The average distance of all samples in the nearest cluster; the range of SC is [-1, 1], and its value is positively correlated with the clustering result; ； ； ； Among them, CH is the Calinski-Harabasz index, and is the covariance matrix of the data between and within clusters, and Represent the center points of cluster q and data set D respectively, represents the number of data sets belonging to cluster q, represents the trace of the matrix in the between-cluster covariance matrix, represents the trace of the matrix in the intra-cluster covariance matrix; the CH indicator represents the ratio between the distance between clusters and the distance within a cluster, and its ratio is positively correlated with the clustering effect; k represents the number of clusters.

4. The intelligent traffic guidance device based on machine learning according to claim 3 is characterized in that: The guiding device body comprises a bottom plate (1), the top of the bottom plate (1) is fixedly connected to a fixing plate (2), and the top of the fixing plate (2) is fixedly connected to a control box (3); A controller and an alarm are fixedly connected to the inner bottom wall of the control box (3), and the controller is used to control the operation of the alarm; A plurality of cameras (4) are fixedly connected to the bottom plate (1) along the circumference of its side wall, and the controller is used to control the operation of the cameras (4); A guide component for displaying guide information and a transmission component for driving the guide component to operate are provided on the top of the base plate (1).

5. The intelligent traffic guidance device based on machine learning according to claim 4 is characterized in that: The transmission assembly comprises a driving frame (5) fixedly connected to one side of the bottom plate (1); a first driving member is fixedly connected to the inner bottom wall of the driving frame (5); a controller is used to control the operation of the first driving member; a transmission gear (6) is coaxially fixedly connected to the output shaft of the first driving member; the transmission gear (6) is meshed with a first gear ring (7); the first gear ring (7) is rotatably matched with the top of the bottom plate (1); a side of the first gear ring (7) away from the transmission gear (6) is meshed with a plurality of moving wheels (8); a side of the moving wheel (8) away from the first gear ring (7) is meshed with a second gear ring (9); the second gear ring (9) is fixedly connected to the side wall of the fixed plate (2).

6. The intelligent traffic guidance device based on machine learning according to claim 5, characterized in that: The guide assembly comprises a moving rod (10) coaxially fixedly connected to the moving wheel (8); A plurality of guide frames (11) are arranged above the bottom plate (1), the tops of adjacent moving rods (10) are rotatably matched with the bottoms of the adjacent guide frames (11), and replacement components for replacing different guide modes are arranged in the guide frames (11).

7. The intelligent traffic guidance device based on machine learning according to claim 6, characterized in that: The replacement assembly comprises a second driving member fixedly connected to the inner wall of the guide frame (11), and the controller is used to control the operation of the second driving member; the output shaft of the second driving member is coaxially fixedly connected to a connecting rod (12), and the connecting rod (12) is fixedly connected to a limiting block (13) at one end away from the output shaft of the second driving member; The inner wall of the guide frame (11) is rotatably matched with a rotating shaft (14), a limiting wheel (15) is coaxially fixedly connected to the rotating shaft (14), a plurality of limiting grooves are equidistantly provided on the limiting wheel (15), and the limiting blocks (13) are slidably matched with the limiting grooves; A plurality of first bevel gears (16) are coaxially and equidistantly fixedly connected to the rotating shaft (14); each of the first bevel gears (16) is meshed with a second bevel gear (17); each of the second bevel gears (17) is coaxially and fixedly connected to a reversing rod (18); each of the reversing rods (18) is fixedly connected to a guide column (19); and one end of the reversing rod (18) away from the second bevel gear (17) is rotatably engaged with an inner top wall of the guide frame (11).

8. The intelligent traffic guidance device based on machine learning according to claim 7, characterized in that: The guide columns (19) are all in the shape of a triangular column.

9. The intelligent traffic guidance device based on machine learning according to claim 8, characterized in that: The side walls of the guide columns (19) are fixedly connected with guide plates (20) representing different guide strategies.

10. The intelligent traffic guidance device based on machine learning according to claim 9, characterized in that: A plurality of universal wheels (21) are fixedly connected to the bottom of the base plate (1).

Images