Toll station traffic flow monitoring system based on digital twin
Through digital twin technology, digital twins of toll stations and roads are built, vehicle information is analyzed and vehicle traffic pressure is predicted, guidance and emergency strategies are generated, and vehicle congestion prediction and management problems are solved in highway toll stations, and the accuracy of traffic flow management is improved. and efficiency.
Patent Information
- Application Number
- CN202411297272.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-18
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-09-18
AI Technical Summary
The prior art is difficult to accurately predict and deal with vehicle congestion at highway toll stations, especially in emergencies, which lead to difficult to effectively manage traffic delays and traffic pressures.
Using a toll station traffic monitoring system based on digital twins, to build digital twins for toll stations and roads through servers and guidance boards, collect and analyze vehicle information, simulate vehicle traffic pressure and predict the number of vehicles to enter, and generate guidance and emergency response. Strategy, optimize lane allocation and traffic management.
It improves the congestion prediction capabilities of toll stations, provides more accurate guidance strategies, reduces traffic pressure, reduces traffic delays, and enhances the ability to respond to emergencies.
Smart Images

Figure CN118824017B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of highway intelligent management technology, and in particular to a toll station vehicle flow monitoring system based on digital twins. Background Art
[0002] Toll stations are facilities on highways that collect tolls from vehicles. The highway network toll collection system has been developed for nearly 20 years. With the requirements of the intelligent era, higher requirements have been placed on the toll station's traffic business and service capabilities. Traditional toll station business is only satisfied with the completion of the toll business, that is, to ensure that the tolls of vehicles are collected accurately and no billing errors occur, but the experience and efficiency of vehicle traffic are not particularly valued. As a key node for entering and exiting the highway, the toll station's traffic capacity is generally far less than the traffic capacity of the road section, becoming a "bottleneck point" on the highway. It is easy to form queues at the entrances and exits of the toll station, resulting in greatly increased traffic delays. If evacuation measures are not taken in time for the queued vehicles, the queues will continue to extend and lengthen, and even affect the traffic of vehicles on the main line of the highway or urban roads.
[0003] The congestion problem at toll stations usually needs to be addressed from three aspects: toll station transformation, emergency control strategies, and toll station service methods. First, the toll stations should be renovated and expanded to improve quality and efficiency. Then, control methods such as main line speed limit, entrance flow control, and toll lane numerical control can be used to accelerate the dissipation of congested traffic at toll stations, as well as flexible lane opening and closing methods. However, the environment of toll stations is usually more complex, and the existing technology only analyzes vehicles on the entrance road of toll stations. For vehicles on the road, it is difficult to determine whether they will enter the toll station. Therefore, it is mainly to analyze the traffic conditions on each lane after entering the toll station section, guide the vehicles, and summarize the regularity, analyze each traffic peak, and take emergency measures in advance before the peak period arrives. Although this method can cope with most scenarios, it relies too much on the development of events according to established rules. When emergencies or unexpected events occur, it is very likely that there will be congestion at the toll station due to the lack of emergency warning in advance. At the same time, in the existing technology, the research method on queuing at highway toll stations is mainly the M / G / K model, which cannot reflect certain differences between toll lanes and has certain limitations. Summary of the invention
[0004] The technical problem solved by the present invention is to provide a toll station traffic flow monitoring system based on digital twins, which can improve the congestion prediction ability of toll stations so as to make better response strategies.
[0005] The basic solution provided by the present invention is: a toll station vehicle flow monitoring system based on digital twin, including a server and a guide board;
[0006] The server includes a model building module, a traffic collection module, a vehicle analysis module, a vehicle prediction module, a vehicle guidance module and an emergency management module;
[0007] A model building module, used to collect digital point clouds of the toll station and the roads within a preset first distance from the toll station, and build digital twins of the toll station and the roads according to the digital point clouds;
[0008] A traffic collection module is used to collect first traffic information, wherein the first traffic information includes vehicle types, vehicle quantities, travel roads, vehicle speeds, and movement trajectories on roads within a first distance preset from a toll station, and map real-time vehicles on roads within the first distance to digital twins according to the vehicle information to generate a dynamic model of the toll station;
[0009] The vehicle analysis module simulates the number of vehicles passing through each lane per unit time in the toll station dynamic model according to the vehicle speed and number of vehicles on each lane, and determines the traffic pressure. Traffic pressure = number of vehicles on the lane / number of vehicles passing through per unit time;
[0010] a vehicle prediction module, configured to collect second traffic information, wherein the second traffic information includes vehicle information and vehicle behavior on a road between a first distance and a second distance preset from the toll station, wherein the second distance is greater than the first distance, and based on the second traffic information, analyzing the possibility of each vehicle on the road between the first distance and the second distance entering the toll station, and predicting the number of vehicles that will enter the first distance based on the possibility of each vehicle entering the toll station;
[0011] The vehicle guidance module is used to identify the number of vehicles that each lane can currently carry, where the number of vehicles that can be carried = the preset limit carrying capacity - unit time × the number of vehicles passing in the unit time - the number of vehicles on the current lane, and generate a guidance strategy based on the number of vehicles that each lane can carry, and map the vehicles that are about to enter the toll station to the toll station dynamic model simulation operation according to the guidance strategy to verify the feasibility of the guidance strategy. When the verification is passed, the guidance strategy is displayed through the guidance billboard set at the first distance;
[0012] The emergency management module generates an emergency strategy when the number of vehicles predicted to enter the first distance range is higher than a carrying threshold generated according to the number of vehicles that each lane can currently carry.
[0013] The principles and advantages of the present invention are as follows: First, the three-dimensional point cloud data of the toll station and the roads within a certain range at the toll station entrance are collected, and the digital twins of the toll station and the roads are established based on the point cloud data. The roads within the first distance here refer to roads that will inevitably enter the toll station and have no other branches. They are set according to the specific conditions of the toll station, that is, the vehicles on the roads within the first distance are vehicles that will inevitably enter the toll station. After the digital twins of the toll station and the road are established, the first traffic information within the first distance is collected. The first traffic information includes the vehicle type, the number of vehicles, the driving road, the vehicle speed, and the moving trajectory on the road within the first distance. According to the vehicle information, the real-time vehicles on the road within the first distance are mapped to the digital twin, thereby obtaining a dynamic model of the toll station, through the dynamic model of the toll station. That is, through the dynamic model of the toll station, the real-time traffic conditions near the toll station are restored. After that, the toll station dynamic model is used for simulation operation to obtain the traffic pressure of vehicles on each lane. The traffic pressure is based on the speed of vehicles on each lane and the number of vehicles. The number of vehicles passing through each lane per unit time is simulated in the toll station dynamic model, and then the traffic pressure is judged based on it. Specifically, traffic pressure = number of vehicles on the lane / number of vehicles passing through per unit time. The larger this value is, the greater the traffic pressure. Toll stations are usually divided into manual lanes, ETC lanes, and mixed lanes according to the type of toll collection. According to the type of vehicle, they are divided into lanes for passenger cars, free lanes for military and police, green lanes, load-bearing lanes, special lanes, etc. Different lanes have different traffic speeds according to different toll types and different passing vehicles. Therefore, in the real-time traffic environment, the traffic pressure they face may also be different. The toll station dynamic model generated by digital twin technology simulates the operation of each lane according to the real-time traffic conditions to obtain the traffic pressure of each lane.
[0014] After that, the vehicle information on the road between the first distance and the second distance from the toll station is obtained. The road between the first distance and the second distance is a road that does not necessarily enter the toll station. For example, the first distance is set at the ramp of the toll station, and the road within the first distance is the road from the ramp to the entrance of the toll station. The second distance is set outside the ramp. The road between the first distance and the second distance is one end of the road before entering the ramp. The vehicles on this road may enter the toll station from the ramp, or may not enter the ramp to continue driving. At this time, the second vehicle information on the road between the first distance and the second distance is collected. The second vehicle information includes vehicle information and vehicle behavior. According to the second vehicle information, the number of vehicles that are about to enter the toll station is predicted, so as to judge the permission on the subsequent toll station entrance road according to the predicted number of vehicles. Since the vehicles on the road between the first distance and the second distance are not necessarily the roads that will enter the toll station, there is no need to map the vehicles on the road between the first distance and the second distance to the digital twin of the toll station. Only after entering the road within the first distance will they be mapped to the digital twin.
[0015] According to the traffic pressure of each lane within the first distance, the vehicles entering the road within the first distance are guided by the guide signs set at the first distance, and the subsequent vehicles are guided to the lanes with less traffic pressure. When the number of vehicles predicted to enter the first distance is higher than the carrying threshold generated according to the number of vehicles that each lane can currently carry, it means that the number of vehicles currently predicted to enter the road within the first distance exceeds the bearing limit of the road obtained from the simulation operation. If left unchecked, it will cause road congestion, so an emergency strategy will be generated at this time.
[0016] Compared with existing technologies:
[0017] 1. In this solution, vehicles that are bound to enter the toll station within the first distance of the toll station are mapped to the digital twin for simulation operation, and the traffic pressure of each lane is obtained through real data. By analyzing the traffic pressure of different lanes, the differences between different lanes can be reflected, such as the difference between ETC lanes and manual lanes, and the difference between special lanes and ordinary lanes. Through the differences between different lanes, more accurate guidance strategies can be provided for guiding different types of vehicles that enter later, reducing traffic pressure.
[0018] 2. Analyze the second vehicle information on the road within the second distance and between the first distance, and determine the possibility of vehicles on the road between the second distance and the first distance entering the toll station based on the second vehicle information, so as to determine the number of vehicles that will subsequently enter the toll station, so as to understand the general situation of the vehicles that are about to enter the toll station.
[0019] 3. After predicting the vehicles that are about to enter the toll station and generating the guidance strategy, the predicted vehicles that are about to enter the toll station are mapped to the toll station dynamic model for simulation operation, so as to verify the feasibility of the guidance strategy and ensure that the guidance strategy is correct.
[0020] Further, the vehicle prediction module includes an information recognition module, a lane recognition module, a speed recognition module and a scoring module;
[0021] An information identification module is used to identify vehicle information of vehicles at a first distance and a second distance, wherein the vehicle information is license plate information, and whether the vehicle registration place is the location of the toll station is identified according to the license plate information, and the vehicles are divided into vehicles in the city, vehicles in the province, and vehicles from other places. An information coefficient X is generated according to the identification result. When the toll station is located at an out-of-city toll station, the information coefficient X of vehicles in the city is less than that of vehicles in the province and less than that of vehicles from other places. When the toll station is located at an in-city toll station, the information coefficient X of vehicles in the city is greater than that of vehicles in the province and greater than that of vehicles from other places.
[0022] The lane recognition module, wherein the vehicle behavior includes lane change behavior, is used to identify the driving lane of the vehicle at the first distance and the second distance, wherein the lanes are provided with increasing lane coefficients C from left to right, and according to the lane change behavior of the vehicle, when the vehicle changes lanes from a lane with a low lane coefficient C to a lane with a high lane coefficient, a correction lane coefficient is generated in is the lane coefficient corrected after the vehicle changes lanes. is the lane coefficient of the lane where the vehicle is when it enters the second distance, The lane coefficient of the lane where the vehicle is after changing lanes. When the vehicle changes lanes from a lane with a high lane coefficient C to a lane with a low lane coefficient, a correction lane coefficient is generated. ;
[0023] The speed recognition module, wherein the vehicle behavior includes a speed change behavior, is used to obtain and recognize the acceleration of the vehicle after entering the second distance and traveling a preset distance, and to generate an acceleration coefficient S according to the acceleration, wherein the smaller the acceleration, the larger the acceleration coefficient S;
[0024] The scoring module is used to generate a probability score based on the information coefficient X, the lane coefficient C, and the acceleration coefficient S:
[0025]
[0026] in , , They are the pre-examination weights of the information coefficient, lane coefficient and acceleration coefficient respectively.
[0027] The vehicles between the second distance and the first distance are vehicles that may enter the toll station, so it is necessary to analyze the possibility of their entering the toll station based on multiple factors. In this application, the analysis is conducted from three aspects: vehicle information, lane, and speed. The information recognition module identifies the vehicle license plate based on the license plate information, identifies the vehicle registration place based on the license plate, and identifies whether the vehicle registration place is the location of the toll station, thereby generating an information coefficient. For example, when the toll station is an entrance toll station, the vehicle coefficient of non-provincial vehicles will be higher than that of non-city vehicles and higher than that of local vehicles. If it is an entrance toll station, the vehicle coefficient of local vehicles will be higher than that of non-city vehicles and higher than that of non-provincial vehicles.
[0028] The lane recognition module identifies the vehicle's lane change behavior. When a vehicle enters a toll station, it needs to enter the toll station lane from the rightmost ramp, so it needs to change lanes to the rightmost lane or the lane close to the rightmost lane in advance. Therefore, the driving lanes of the vehicles at the first distance and the second distance are identified, and the lanes of each vehicle when entering the second distance are identified. The lanes are set with increasing lane coefficients from left to right, that is, the lane coefficient C is lower for the left lanes farther away from the toll station ramp, and the lane coefficient C is higher for the right lanes closer to the toll station ramp. At the same time, after entering the second distance, if the vehicle changes lanes, if it changes lanes to the left and moves away from the toll station ramp, the possibility of the vehicle getting off the toll station ramp will be reduced, and if it still changes lanes to the left when approaching the toll station, its possibility of getting off is even lower than that of the vehicle driving in the left lane. Therefore, the corrected lane coefficient generated will be smaller than the lane coefficient of its own lane. If it changes lanes to the right and approaches the toll station ramp, the vehicle changes lanes to the right when approaching the toll station, then its possibility of getting off will be higher than that of the vehicle driving in the right lane. Therefore, its corrected lane coefficient is calculated through , will be greater than .
[0029] The speed recognition module recognizes the acceleration of each vehicle after entering the second distance. Since the ramp of the toll station usually has a curve, the vehicle will decelerate when entering the ramp, so its acceleration is negative. By identifying the acceleration of the vehicle after entering the second distance and traveling a preset distance, such as the acceleration of traveling 400m, the acceleration analysis is performed to determine whether the vehicle is accelerating or decelerating, and the acceleration coefficient S is assigned to the vehicle according to the acceleration. The lower the acceleration, the higher the acceleration coefficient S.
[0030] Finally, based on the preset weights of each coefficient, the probability score of each vehicle entering the toll station is calculated.
[0031] Further, the vehicle prediction module also includes a quantity recognition module;
[0032] A quantity recognition module, used to recognize the number of vehicles within the first distance and the second distance;
[0033] The scoring module is also used to adjust the preset weights of each coefficient according to the number of vehicles within the first distance and the second distance. The more vehicles there are, the lower the preset weight of the acceleration coefficient is, and the higher the preset weight of the lane coefficient is.
[0034] Identify the number of vehicles within the first distance and the second distance, and adjust the weight of each coefficient according to the number of vehicles within the first distance and the second distance. When the number of vehicles within the first distance and the second distance is large, the change in the vehicle speed in the area is smaller, and when there are more vehicles, vehicles that need to enter the toll station will change lanes in advance. Therefore, the more vehicles within the first distance and the second distance, the higher the weight of the lane coefficient and the lower the weight of the acceleration coefficient when calculating the possibility score.
[0035] Further, the vehicle prediction module also includes a lane change recognition module;
[0036] a lane change recognition module, for identifying the number of lane changes of a vehicle within a first distance and a second distance;
[0037] The lane recognition module is also used to generate a lane change impact score M according to the lane change behavior when the number of lane changes of the vehicle exceeds a preset number threshold. The initial lane change impact score is 0. Each time the vehicle changes lanes to the left, the impact score is M-1. Each time the vehicle changes lanes to the right, the impact score is M+1. The lane change threshold is equal to the number of lanes.
[0038] Corrected lane coefficient
[0039] Where Q is the number of lane changes of the vehicle.
[0040] When driving on the road, there will be people who like to change lanes back and forth, so a number threshold is set to detect whether the lane changer is changing lanes back and forth, so as to avoid errors in the predicted vehicles that will enter the toll station. Set the number threshold between the first distance and the second distance. The number threshold is the number of lanes, that is, the maximum allowed vehicle to change lanes from the leftmost to the rightmost. At the same time, set the lane change impact score. Changing lanes to the left once will have an impact score of -1, and changing lanes to the right once will have an impact score of +1. Therefore, if the vehicle changes lanes to the left more times, the final impact score will be negative. If it changes lanes to the right more times, the impact score will be positive. If it changes lanes back and forth, it will be close to 0. By setting the impact score, it is determined which vehicles with more lane changes will actually enter the toll station.
[0041] Further, the vehicle prediction module also includes a vehicle marking module;
[0042] The vehicle marking module has a preset scoring standard for marking vehicles within the first distance and the second distance whose probability scores are higher than the scoring standard as vehicles about to enter the toll station. The more vehicles on the road within the first distance, the lower the scoring standard.
[0043] The more vehicles there are within the first distance, the lower the scoring standard. When there are more vehicles, the scoring standard is set low to identify more vehicles that may enter and reduce the possibility of triggering the warning. Since this solution predicts the possibility of vehicle entry, the number of vehicles that are about to enter the toll station is predicted as a predicted value, which may be too large or too small. Therefore, when there are a large number of vehicles in the toll station, the scoring standard is lowered to predict that more vehicles will enter to avoid the possibility of a small predicted value. If an early warning is generated, it is better to take emergency measures in advance. Even if the actual number of vehicles entering is less than the predicted value, it will not have a significant impact.
[0044] Further, the vehicle marking module includes a strategy generation module and a strategy simulation module;
[0045] A strategy generation module is used to identify the vehicle type and vehicle information of the vehicles marked as about to enter the toll station, determine the guidance type of each vehicle, and according to the number of vehicles of each guidance type, identify the number of vehicles that can be carried by each lane corresponding to the guidance type, and select the lane with the largest number of vehicles that can be carried for guidance one by one;
[0046] The strategy simulation module is used to map the vehicles that are about to enter the toll station to the dynamic model of the toll station, simulate the number of vehicles that each lane can carry after each vehicle enters the corresponding vehicle, identify the difference between the number of vehicles that the lane with the same guidance type can carry after simulated guidance and the current number of vehicles that can be carried, and identify whether the number difference of the lanes with the same guidance type is close. If it is close, guidance is carried out according to the current guidance strategy. If not, priority is given to guiding to the lane with the largest number difference.
[0047] When a vehicle is marked as a vehicle about to enter a toll station, the guidance type of each vehicle is determined according to its vehicle type and vehicle information. For example, the vehicle information is used to identify whether the vehicle is an ETC vehicle. If so, its guidance type is an ETC lane. If not, its guidance type is a manual lane. And according to the vehicle type, it is determined whether it is a dedicated lane for passenger cars, a free lane for military and police, a load-bearing lane, etc. After determining the guidance type of each vehicle, the number of vehicles of each guidance type is identified, such as 20 vehicles for ETC lanes and 15 vehicles for manual lanes, and then the maximum carrying capacity of lanes of each current guidance type is identified, and the lane with the largest number of vehicles that can be carried is selected one by one for guidance. For example, there are 3 ETC lanes, the maximum carrying capacity of ETC lane A is 9 vehicles, the maximum carrying capacity of ETC lane B is 8 vehicles, and the maximum carrying capacity of ETC lane C is 6 vehicles. The vehicle with the largest number of vehicles is ETC lane A, and the first vehicle with the guidance type of ETC lane is guided to ETC lane A. Then the maximum carrying capacity of ETC lane A is 8 vehicles, the maximum carrying capacity of ETC lane B is 8 vehicles, and the maximum carrying capacity of ETC lane C is 6 vehicles. At this time, the next vehicle with the guidance type of ETC lane can be guided to ETC lane A or lane B. According to the above logic, a complete guidance strategy is generated.
[0048] Then the guidance strategy is simulated. Since there are vehicles leaving the toll station at the same time as vehicles enter the toll station, the toll station dynamic model is used for simulation to identify the maximum load capacity of each lane after all the vehicles in each lane are guided in the simulation results. For example, after all 20 vehicles are guided, the maximum load capacity of ETC lane A is 7, the maximum load capacity of ETC lane B is 8, and the maximum load capacity of ETC lane C is 8. The difference in quantity before and after the guidance of ETC lane A is -2, the difference in quantity of ETC lane B is 0, and the difference in quantity of ETC lane C is 1. The larger the difference in quantity, the stronger the current vehicle's traffic capacity. At this time, first identify whether the quantity difference is close. In this application, the variance is calculated. When the variance is within the preset range, it is considered close. If it is not resolved, the guidance strategy is adjusted to give priority to guiding lanes with large quantity differences.
[0049] Furthermore, the emergency management module is used to mark the guidance type as a warning type when the number of vehicles entering a guidance type exceeds the number of vehicles that the corresponding lane can carry, and adjust the warning level according to the number of warning types. The more warning types there are, the higher the warning level. BRIEF DESCRIPTION OF THE DRAWINGS
[0050] Figure 1 It is a logic block diagram of an embodiment of a toll station vehicle flow monitoring system based on digital twins of the present invention;
[0051] Figure 2This is a logic block diagram of a vehicle prediction module in an embodiment of a toll station vehicle flow monitoring system based on digital twins of the present invention. DETAILED DESCRIPTION
[0052] The following is further described in detail through specific implementation methods:
[0053] The embodiment is basically as shown in the attached Figure 1 As shown:
[0054] A toll station vehicle flow monitoring system based on digital twin, including a server and a guide board;
[0055] The server includes a model building module, a traffic collection module, a vehicle analysis module, a vehicle prediction module, a vehicle guidance module and an emergency management module;
[0056] A model building module, used to collect digital point clouds of the toll station and the roads within a preset first distance from the toll station, and build digital twins of the toll station and the roads according to the digital point clouds;
[0057] A traffic collection module is used to collect first traffic information, wherein the first traffic information includes vehicle types, vehicle quantities, travel roads, vehicle speeds, and movement trajectories on roads within a first distance preset from a toll station, and map real-time vehicles on roads within the first distance to digital twins according to the vehicle information to generate a dynamic model of the toll station;
[0058] The vehicle analysis module simulates the number of vehicles passing through each lane per unit time in the toll station dynamic model according to the vehicle speed and number of vehicles on each lane, and determines the traffic pressure. Traffic pressure = number of vehicles on the lane / number of vehicles passing through per unit time;
[0059] a vehicle prediction module, configured to collect second traffic information, wherein the second traffic information includes vehicle information and vehicle behavior on a road between a first distance and a second distance preset from the toll station, wherein the second distance is greater than the first distance, and based on the second traffic information, analyzing the possibility of each vehicle on the road between the first distance and the second distance entering the toll station, and predicting the number of vehicles that will enter the first distance based on the possibility of each vehicle entering the toll station;
[0060] The vehicle guidance module is used to identify the number of vehicles that each lane can currently carry, where the number of vehicles that can be carried = the preset limit carrying capacity - unit time × the number of vehicles passing in the unit time - the number of vehicles on the current lane, and generate a guidance strategy based on the number of vehicles that each lane can carry, and map the vehicles that are about to enter the toll station to the toll station dynamic model simulation operation according to the guidance strategy to verify the feasibility of the guidance strategy. When the verification is passed, the guidance strategy is displayed through the guidance billboard set at the first distance;
[0061] The emergency management module generates an emergency strategy when the number of vehicles predicted to enter the first distance range is higher than a carrying threshold generated according to the number of vehicles that each lane can currently carry.
[0062] First, the three-dimensional point cloud data of the toll station and the roads within a certain range at the entrance of the toll station are collected, and the digital twins of the toll station and the roads are established based on the point cloud data. The roads within the first distance here refer to the roads that will inevitably enter the toll station and have no other branches. They are set according to the specific conditions of the toll station, that is, the vehicles on the roads within the first distance are the vehicles that will inevitably enter the toll station. After the digital twins of the toll station and the road are established, the first traffic information within the first distance is collected. The first traffic information includes the vehicle type, vehicle number, driving road, vehicle speed, and moving trajectory on the road within the first distance. According to the vehicle information, the real-time vehicles on the road within the first distance are mapped to the digital twin, so as to obtain the toll station dynamic model. That is, through the toll station dynamic model, the real-time traffic conditions near the toll station are restored. After that, the toll station dynamic model is simulated and run to obtain the traffic pressure of vehicles on each lane. Toll stations are usually divided into manual lanes, ETC lanes, and mixed lanes according to the toll type, and divided into small passenger car lanes, military and police free lanes, green lanes, load-bearing lanes, special lanes, etc. according to the vehicle type. Different lanes have different toll types, different vehicles, and different speeds, so they may face different traffic pressures in real-time traffic environments. The dynamic model of the toll station generated by digital twin technology simulates the operation of each lane based on real-time traffic conditions to obtain the traffic pressure of each lane.
[0063] After that, the vehicle information on the road between the first distance and the second distance from the toll station is obtained. The road between the first distance and the second distance is a road that does not necessarily enter the toll station. For example, the first distance is set at the ramp of the toll station, and the road within the first distance is the road from the ramp to the entrance of the toll station. The second distance is set outside the ramp. The road between the first distance and the second distance is one end of the road before entering the ramp. The vehicles on this road may enter the toll station from the ramp, or may not enter the ramp to continue driving. At this time, the second vehicle information on the road between the first distance and the second distance is collected. The second vehicle information includes vehicle information and vehicle behavior. According to the second vehicle information, the number of vehicles that are about to enter the toll station is predicted, so as to judge the permission on the subsequent toll station entrance road according to the predicted number of vehicles. Since the vehicles on the road between the first distance and the second distance are not necessarily the roads that will enter the toll station, there is no need to map the vehicles on the road between the first distance and the second distance to the digital twin of the toll station. Only after entering the road within the first distance will they be mapped to the digital twin.
[0064] According to the traffic pressure of each lane within the first distance, the vehicles entering the road within the first distance are guided by the guide signs set at the first distance, and the subsequent vehicles are guided to the lanes with less traffic pressure. When the number of vehicles predicted to enter the first distance is higher than the carrying threshold generated according to the number of vehicles that each lane can currently carry, it means that the number of vehicles currently predicted to enter the road within the first distance exceeds the bearing limit of the road obtained from the simulation operation. If left unchecked, it will cause road congestion, so an emergency strategy will be generated at this time.
[0065] In this solution, vehicles that are bound to enter the toll station within the first distance of the toll station are mapped to the digital twin for simulation operation, and the traffic pressure of each lane is obtained through real data. By analyzing the traffic pressure of different lanes, the differences between different lanes can be reflected, such as the difference between ETC lanes and manual lanes, and the difference between special lanes and ordinary lanes. Through the differences between different lanes, more accurate guidance strategies can be provided for guiding different types of vehicles that enter later, reducing traffic pressure.
[0066] And analyze the second vehicle information on the road within the second distance and between the first distance, judge the possibility of vehicles on the road between the second distance and the first distance entering the toll station based on the second vehicle information, thereby judging the number of vehicles that will subsequently enter the toll station, so as to understand the general situation of the vehicles that are about to enter the toll station.
[0067] After predicting the vehicles that are about to enter the toll station and generating the guidance strategy, the predicted vehicles that are about to enter the toll station are mapped to the toll station dynamic model for simulation operation, so as to verify the feasibility of the guidance strategy and ensure that the guidance strategy is correct.
[0068] The vehicle prediction module is as follows: Figure 2 The example shown includes an information recognition module, a lane recognition module, a speed recognition module, and a scoring module;
[0069] An information identification module is used to identify vehicle information of vehicles at a first distance and a second distance, wherein the vehicle information is license plate information, and whether the vehicle registration place is the location of the toll station is identified according to the license plate information, and the vehicles are divided into vehicles in the city, vehicles in the province, and vehicles from other places. An information coefficient X is generated according to the identification result. When the toll station is located at an out-of-city toll station, the information coefficient X of vehicles in the city is less than that of vehicles in the province and less than that of vehicles from other places. When the toll station is located at an in-city toll station, the information coefficient X of vehicles in the city is greater than that of vehicles in the province and greater than that of vehicles from other places.
[0070] The lane recognition module, wherein the vehicle behavior includes lane change behavior, is used to identify the driving lane of the vehicle at the first distance and the second distance, wherein the lanes are provided with increasing lane coefficients C from left to right, and according to the lane change behavior of the vehicle, when the vehicle changes lanes from a lane with a low lane coefficient C to a lane with a high lane coefficient, a correction lane coefficient is generated in is the lane coefficient corrected after the vehicle changes lanes. is the lane coefficient of the lane where the vehicle is when it enters the second distance, The lane coefficient of the lane where the vehicle is after changing lanes. When the vehicle changes lanes from a lane with a high lane coefficient C to a lane with a low lane coefficient, a correction lane coefficient is generated. ;
[0071] The speed recognition module, wherein the vehicle behavior includes a speed change behavior, is used to obtain and recognize the acceleration of the vehicle after entering the second distance and traveling a preset distance, and to generate an acceleration coefficient S according to the acceleration, wherein the smaller the acceleration, the larger the acceleration coefficient S;
[0072] The scoring module is used to generate a probability score based on the information coefficient X, the lane coefficient C, and the acceleration coefficient S:
[0073]
[0074] in , , They are the pre-examination weights of the information coefficient, lane coefficient and acceleration coefficient respectively.
[0075] The vehicles between the second distance and the first distance are vehicles that may enter the toll station, so it is necessary to analyze the possibility of their entering the toll station based on multiple factors. In this application, the analysis is conducted from three aspects: vehicle information, lane, and speed. The information recognition module identifies the vehicle license plate based on the license plate information, identifies the vehicle registration place based on the license plate, and identifies whether the vehicle registration place is the location of the toll station, thereby generating an information coefficient. For example, when the toll station is an entrance toll station, the vehicle coefficient of non-provincial vehicles will be higher than that of non-city vehicles and higher than that of local vehicles. If it is an entrance toll station, the vehicle coefficient of local vehicles will be higher than that of non-city vehicles and higher than that of non-provincial vehicles.
[0076] The lane recognition module identifies the vehicle's lane change behavior. When a vehicle enters a toll station, it needs to enter the toll station lane from the rightmost ramp, so it needs to change lanes to the rightmost lane or the lane close to the rightmost lane in advance. Therefore, the driving lanes of the vehicles at the first distance and the second distance are identified, and the lanes of each vehicle when entering the second distance are identified. The lanes are set with increasing lane coefficients from left to right, that is, the lane coefficient C is lower for the left lanes farther away from the toll station ramp, and the lane coefficient C is higher for the right lanes closer to the toll station ramp. At the same time, after entering the second distance, if the vehicle changes lanes, if it changes lanes to the left and moves away from the toll station ramp, the possibility of the vehicle getting off the toll station ramp will be reduced, and if it still changes lanes to the left when approaching the toll station, its possibility of getting off is even lower than that of the vehicle driving in the left lane. Therefore, the corrected lane coefficient generated will be smaller than the lane coefficient of its own lane. If it changes lanes to the right and approaches the toll station ramp, the vehicle changes lanes to the right when approaching the toll station, then its possibility of getting off will be higher than that of the vehicle driving in the right lane. Therefore, its corrected lane coefficient is calculated through , will be greater than .
[0077] The speed recognition module recognizes the acceleration of each vehicle after entering the second distance. Since the ramp of the toll station usually has a curve, the vehicle will decelerate when entering the ramp, so its acceleration is negative. By identifying the acceleration of the vehicle after entering the second distance and traveling a preset distance, such as the acceleration of traveling 400m, the acceleration analysis is performed to determine whether the vehicle is accelerating or decelerating, and the acceleration coefficient S is assigned to the vehicle according to the acceleration. The lower the acceleration, the higher the acceleration coefficient S.
[0078] Finally, based on the preset weights of each coefficient, the probability score of each vehicle entering the toll station is calculated.
[0079] Further, the vehicle prediction module also includes a quantity recognition module;
[0080] A quantity recognition module, used to recognize the number of vehicles within the first distance and the second distance;
[0081] The scoring module is also used to adjust the preset weights of each coefficient according to the number of vehicles within the first distance and the second distance. The more vehicles there are, the lower the preset weight of the acceleration coefficient is, and the higher the preset weight of the lane coefficient is.
[0082] Identify the number of vehicles within the first distance and the second distance, and adjust the weight of each coefficient according to the number of vehicles within the first distance and the second distance. When the number of vehicles within the first distance and the second distance is large, the change in the vehicle speed in the area is smaller, and when there are more vehicles, vehicles that need to enter the toll station will change lanes in advance. Therefore, the more vehicles within the first distance and the second distance, the higher the weight of the lane coefficient and the lower the weight of the acceleration coefficient when calculating the possibility score.
[0083] The vehicle prediction module also includes a lane change recognition module;
[0084] a lane change recognition module, for identifying the number of lane changes of a vehicle within a first distance and a second distance;
[0085] The lane recognition module is also used to generate a lane change impact score M according to the lane change behavior when the number of lane changes of the vehicle exceeds a preset number threshold. The initial lane change impact score is 0. Each time the vehicle changes lanes to the left, the impact score is M-1. Each time the vehicle changes lanes to the right, the impact score is M+1. The lane change threshold is equal to the number of lanes.
[0086] Corrected lane coefficient
[0087] Where Q is the number of lane changes of the vehicle.
[0088] When driving on the road, there will be people who like to change lanes back and forth, so a number threshold is set to detect whether the lane changer is changing lanes back and forth, so as to avoid errors in the predicted vehicles that will enter the toll station. Set the number threshold between the first distance and the second distance. The number threshold is the number of lanes, that is, the maximum allowed vehicle to change lanes from the leftmost to the rightmost. At the same time, set the lane change impact score. Changing lanes to the left once will have an impact score of -1, and changing lanes to the right once will have an impact score of +1. Therefore, if the vehicle changes lanes to the left more times, the final impact score will be negative. If it changes lanes to the right more times, the impact score will be positive. If it changes lanes back and forth, it will be close to 0. By setting the impact score, it is determined which vehicles with more lane changes will actually enter the toll station.
[0089] The vehicle prediction module also includes a vehicle marking module;
[0090] The vehicle marking module has a preset scoring standard for marking vehicles within the first distance and the second distance whose probability scores are higher than the scoring standard as vehicles about to enter the toll station. The more vehicles on the road within the first distance, the lower the scoring standard.
[0091] The more vehicles there are within the first distance, the lower the scoring standard. When there are more vehicles, the scoring standard is set low to identify more vehicles that may enter and reduce the possibility of triggering the warning. Since this solution predicts the possibility of vehicle entry, the number of vehicles that are about to enter the toll station is predicted as a predicted value, which may be too large or too small. Therefore, when there are a large number of vehicles in the toll station, the scoring standard is lowered to predict that more vehicles will enter to avoid the possibility of a small predicted value. If an early warning is generated, it is better to take emergency measures in advance. Even if the actual number of vehicles entering is less than the predicted value, it will not have a significant impact.
[0092] The vehicle marking module includes a strategy generation module and a strategy simulation module;
[0093] A strategy generation module is used to identify the vehicle type and vehicle information of the vehicles marked as about to enter the toll station, determine the guidance type of each vehicle, and according to the number of vehicles of each guidance type, identify the number of vehicles that can be carried by each lane corresponding to the guidance type, and select the lane with the largest number of vehicles that can be carried for guidance one by one;
[0094] The strategy simulation module is used to map the vehicles that are about to enter the toll station to the dynamic model of the toll station, simulate the number of vehicles that each lane can carry after each vehicle enters the corresponding vehicle, identify the difference between the number of vehicles that the lane with the same guidance type can carry after simulated guidance and the current number of vehicles that can be carried, and identify whether the number difference of the lanes with the same guidance type is close. If it is close, guidance is carried out according to the current guidance strategy. If not, priority is given to guiding to the lane with the largest number difference.
[0095] When a vehicle is marked as a vehicle about to enter a toll station, the guidance type of each vehicle is determined according to its vehicle type and vehicle information. For example, the vehicle information is used to identify whether the vehicle is an ETC vehicle. If so, its guidance type is an ETC lane. If not, its guidance type is a manual lane. And according to the vehicle type, it is determined whether it is a dedicated lane for passenger cars, a free lane for military and police, a load-bearing lane, etc. After determining the guidance type of each vehicle, the number of vehicles of each guidance type is identified, such as 20 vehicles for ETC lanes and 15 vehicles for manual lanes, and then the maximum carrying capacity of lanes of each current guidance type is identified, and the lane with the largest number of vehicles that can be carried is selected one by one for guidance. For example, there are 3 ETC lanes, the maximum carrying capacity of ETC lane A is 9 vehicles, the maximum carrying capacity of ETC lane B is 8 vehicles, and the maximum carrying capacity of ETC lane C is 6 vehicles. The vehicle with the largest number of vehicles is ETC lane A, and the first vehicle with the guidance type of ETC lane is guided to ETC lane A. Then the maximum carrying capacity of ETC lane A is 8 vehicles, the maximum carrying capacity of ETC lane B is 8 vehicles, and the maximum carrying capacity of ETC lane C is 6 vehicles. At this time, the next vehicle with the guidance type of ETC lane can be guided to ETC lane A or lane B. According to the above logic, a complete guidance strategy is generated.
[0096] Then the guidance strategy is simulated. Since there are vehicles leaving the toll station at the same time as vehicles enter the toll station, the toll station dynamic model is used for simulation to identify the maximum load capacity of each lane after all the vehicles in each lane are guided in the simulation results. For example, after all 20 vehicles are guided, the maximum load capacity of ETC lane A is 7, the maximum load capacity of ETC lane B is 8, and the maximum load capacity of ETC lane C is 8. The difference in quantity before and after the guidance of ETC lane A is -2, the difference in quantity of ETC lane B is 0, and the difference in quantity of ETC lane C is 1. The larger the difference in quantity, the stronger the current vehicle's traffic capacity. At this time, first identify whether the quantity difference is close. In this application, the variance is calculated. When the variance is within the preset range, it is considered close. If it is not resolved, the guidance strategy is adjusted to give priority to guiding lanes with large quantity differences.
[0097] The emergency management module is used to mark the guidance type as a warning type when the number of vehicles entering the guidance type exceeds the number of vehicles that the corresponding lane can carry, and adjust the warning level according to the number of warning types. The more warning types there are, the higher the warning level.
[0098] The above are only embodiments of the present invention. Common knowledge such as the known specific structures and characteristics in the scheme is not described in detail here. Ordinary technicians in the relevant field know all the common technical knowledge in the technical field to which the invention belongs before the application date or priority date, can obtain all the existing technologies in the field, and have the ability to apply conventional experimental means before that date. Ordinary technicians in the relevant field can improve and implement this scheme in combination with their own abilities under the enlightenment given by this application. Some typical known structures or known methods should not become obstacles for ordinary technicians in the relevant field to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can be made, which should also be regarded as the scope of protection of the present invention, and these will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.
Claims
1. The toll station traffic flow monitoring system based on digital twin is characterized by: Including server and guide board; The server includes a model building module, a traffic collection module, a vehicle analysis module, a vehicle prediction module, a vehicle guidance module and an emergency management module; A model building module, used to collect digital point clouds of the toll station and the roads within a preset first distance from the toll station, and build digital twins of the toll station and the roads according to the digital point clouds; A traffic collection module is used to collect first traffic information, wherein the first traffic information includes vehicle types, vehicle quantities, travel roads, vehicle speeds, and movement trajectories on roads within a first distance preset from a toll station, and map real-time vehicles on roads within the first distance to digital twins according to the vehicle information to generate a dynamic model of the toll station; The vehicle analysis module simulates the number of vehicles passing through each lane per unit time in the toll station dynamic model according to the vehicle speed and number of vehicles on each lane, and determines the traffic pressure. Traffic pressure = number of vehicles on the lane / number of vehicles passing through per unit time; a vehicle prediction module, configured to collect second traffic information, wherein the second traffic information includes vehicle information and vehicle behavior on a road between a first distance and a second distance preset from the toll station, wherein the second distance is greater than the first distance, and based on the second traffic information, analyzing the possibility of each vehicle on the road between the first distance and the second distance entering the toll station, and predicting the number of vehicles that will enter the first distance based on the possibility of each vehicle entering the toll station; The vehicle guidance module is used to identify the number of vehicles that each lane can currently carry, where the number of vehicles that can be carried = the preset limit carrying capacity - unit time × the number of vehicles passing in the unit time - the number of vehicles on the current lane, and generate a guidance strategy based on the number of vehicles that each lane can carry, and map the vehicles that are about to enter the toll station to the toll station dynamic model simulation operation according to the guidance strategy to verify the feasibility of the guidance strategy. When the verification is passed, the guidance strategy is displayed through the guidance billboard set at the first distance; The emergency management module issues an early warning when the number of vehicles predicted to enter the first distance is higher than the carrying threshold generated according to the number of vehicles that each lane can currently carry; The vehicle prediction module includes an information recognition module, a lane recognition module, a speed recognition module and a scoring module; An information identification module is used to identify vehicle information of vehicles at a first distance and a second distance, wherein the vehicle information is license plate information, and whether the vehicle registration place is the location of the toll station is identified according to the license plate information, and the vehicles are divided into vehicles in the city, vehicles in the province, and vehicles from other places. An information coefficient X is generated according to the identification result. When the toll station is located at an out-of-city toll station, the information coefficient X of vehicles in the city is less than that of vehicles in the province and less than that of vehicles from other places. When the toll station is located at an in-city toll station, the information coefficient X of vehicles in the city is greater than that of vehicles in the province and greater than that of vehicles from other places. The lane recognition module, wherein the vehicle behavior includes lane change behavior, is used to identify the driving lane of the vehicle at the first distance and the second distance, wherein the lanes are provided with increasing lane coefficients C from left to right, and according to the lane change behavior of the vehicle, when the vehicle changes lanes from a lane with a low lane coefficient C to a lane with a high lane coefficient, a correction lane coefficient is generated in is the lane coefficient corrected after the vehicle changes lanes. is the lane coefficient of the lane where the vehicle is when it enters the second distance, The lane coefficient of the lane where the vehicle is after changing lanes. When the vehicle changes lanes from a lane with a high lane coefficient C to a lane with a low lane coefficient, a correction lane coefficient is generated. ; The speed recognition module, wherein the vehicle behavior includes a speed change behavior, is used to obtain and recognize the acceleration of the vehicle after entering the second distance and traveling a preset distance, and to generate an acceleration coefficient S according to the acceleration, wherein the smaller the acceleration, the larger the acceleration coefficient S; The scoring module is used to generate a probability score based on the information coefficient X, the lane coefficient C, and the acceleration coefficient S: in , , are the preset weights of information coefficient, lane coefficient and acceleration coefficient respectively; The vehicle prediction module also includes a vehicle marking module; A vehicle marking module, having a preset scoring standard, for marking vehicles within the first distance and the second distance whose probability scores are higher than the scoring standard as vehicles about to enter the toll station, wherein the more vehicles there are on the road within the first distance, the lower the scoring standard; The vehicle marking module includes a strategy generation module and a strategy simulation module; A strategy generation module is used to identify the vehicle type and vehicle information of the vehicles marked as about to enter the toll station, determine the guidance type of each vehicle, and according to the number of vehicles of each guidance type, identify the number of vehicles that can be carried by each lane corresponding to the guidance type, and select the lane with the largest number of vehicles that can be carried for guidance one by one; The strategy simulation module is used to map the vehicles that are about to enter the toll station to the dynamic model of the toll station, simulate the number of vehicles that each lane can carry after each vehicle enters the corresponding vehicle, identify the difference between the number of vehicles that the lane with the same guidance type can carry after simulated guidance and the current number of vehicles that can be carried, and identify whether the number difference of the lanes with the same guidance type is close. If it is close, guidance is carried out according to the current guidance strategy. If not, priority is given to guiding to the lane with the largest number difference.
2. The toll station traffic flow monitoring system based on digital twin according to claim 1 is characterized by: The vehicle prediction module also includes a quantity recognition module; A quantity recognition module, used to recognize the number of vehicles within the first distance and the second distance; The scoring module is also used to adjust the preset weights of each coefficient according to the number of vehicles within the first distance and the second distance. The more vehicles there are, the lower the preset weight of the acceleration coefficient is, and the higher the preset weight of the lane coefficient is.
3. The toll station vehicle flow monitoring system based on digital twin according to claim 1 is characterized in that: The vehicle prediction module also includes a lane change recognition module; a lane change recognition module, for identifying the number of lane changes of a vehicle within a first distance and a second distance; The lane recognition module is also used to generate a lane change impact score M according to the lane change behavior when the number of lane changes of the vehicle exceeds a preset number threshold. The initial lane change impact score is 0. Each time the vehicle changes lanes to the left, the impact score is M-1. Each time the vehicle changes lanes to the right, the impact score is M+1. The lane change threshold is equal to the number of lanes. Corrected lane coefficient Where Q is the number of lane changes of the vehicle.
4. The toll station vehicle flow monitoring system based on digital twin according to claim 1 is characterized in that: The emergency management module is used to mark the guidance type as a warning type when the number of vehicles entering the guidance type exceeds the number of vehicles that the corresponding lane can carry, and adjust the warning level according to the number of warning types. The more warning types there are, the higher the warning level.
Citation Information
Cited By
Tide prediction-based toll station cloud warehouse management scheduling method, system and equipment
CN122658089A
Toll station cloud warehouse management scheduling method, system and equipment based on tide prediction
CN122658089B
