Vehicle scheduling method of traffic station center road passenger transport management system

Abstract

The invention discloses a vehicle scheduling method of a traffic station center road passenger transport management system, and relates to the technical field of traffic control. The method comprises the following steps: acquiring running state information and road traffic information of each vehicle in real time; based on the running state information of each vehicle and the road traffic information, respectively evaluating the passenger bad experience factor of each vehicle at the current moment, and based on the running state information of each vehicle, respectively evaluating the driving risk of each vehicle at the current moment; based on the passenger bad experience factor and the driving risk of each vehicle, determining a scheduling priority factor of each vehicle; and performing resource scheduling on each vehicle based on the scheduling priority factor of each vehicle to obtain a vehicle scheduling result. According to the method, the problem of unreasonable scheduling caused by neglect of individual differences of vehicles in a traditional scheduling method is effectively solved, so that the rationality of vehicle scheduling is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic control, and in particular to a vehicle dispatching method for a road passenger transport management system in a traffic station center. Background Art

[0002] Against the backdrop of today's growing and complex transportation demands, the vehicle dispatching approach within the road passenger transport management system employed by transportation hubs and terminals has become a key measure for improving passenger service quality and operational efficiency. This approach, powered by multi-source data fusion and cutting-edge intelligent technologies, is dedicated to building a comprehensive, standardized, automated, and visualized management system for road passenger transport operations, from source planning to terminal services.

[0003] In the current vehicle dispatching method, dynamic updating of vehicle dispatching results based on real-time data is an essential step, so as to be able to respond to various emergencies and changing traffic conditions in a timely manner.

[0004] However, due to the differences in the real-time status of passengers in each vehicle, the driving status of the vehicle itself, and the traffic conditions encountered in reality, these differences have led to unreasonable vehicle scheduling arrangements to a certain extent. Summary of the Invention

[0005] The embodiment of the present invention provides a vehicle dispatching method for a road passenger transport management system in a transportation station center, which can improve the rationality of vehicle dispatching.

[0006] A first aspect of an embodiment of the present invention provides a vehicle dispatching method for a road passenger transport management system in a transportation terminal center, the method comprising: Obtain the operating status information and road traffic information of each vehicle in real time; Based on the operating status information and road traffic information of each vehicle, the passenger adverse experience factor of each vehicle at the current moment is evaluated, and based on the operating status information of each vehicle, the driving risk of each vehicle at the current moment is evaluated; Determine the dispatch priority factor for each vehicle based on its passenger adverse experience factor and driving risk; Based on the scheduling priority factor of each vehicle, resource scheduling is performed on each vehicle to obtain the vehicle scheduling result.

[0007] Furthermore, the present invention also proposes to evaluate the passenger adverse experience factor of each vehicle at the current moment based on the operating status information and road traffic information of each vehicle, including: Determining the passenger anxiety level of each passenger on each vehicle based on the operating status information and road traffic information of each vehicle, and predicting the future congestion impact factor of each vehicle based on the operating status information and road traffic information of each vehicle; Determine the current congestion impact factor of each vehicle based on the passenger anxiety level of each passenger on each vehicle and the passenger's elapsed time on each vehicle. The elapsed time is used to represent the length of time from the moment the passenger boarded the vehicle to the current moment. The future congestion impact factor of each vehicle is averaged with the current congestion impact factor to obtain the passenger adverse experience factor of each vehicle at the current moment.

[0008] Furthermore, the present invention also proposes that the operation status information includes the time when passengers board the vehicle, the number of passengers, and the vehicle's speed, and the road traffic information includes the congestion level of each route segment; Based on the operating status information of each vehicle and the road traffic information, the passenger anxiety level of each passenger in each vehicle is determined separately, including: The time period from the boarding time to the current time of each passenger on the target vehicle is determined as the riding time of each passenger, and the target vehicle is any vehicle; The congestion degree of each passenger on the target vehicle during the riding period is constructed in chronological order to obtain a congestion degree sequence for each passenger; The ratio of the number of passengers at each moment to the maximum load is constructed in chronological order to obtain the passenger ratio sequence of the target vehicle; Taking the passenger proportion sequence as the weight sequence, perform linear fitting on each congestion degree sequence to obtain the slope of the fitted line corresponding to each passenger; The passenger anxiety level of each passenger is determined based on the target congestion level mean of each passenger, the slope of the fitted line, and the vehicle speed at each moment. The target congestion level mean is the mean of all congestion levels in the passenger's congestion level sequence.

[0009] Furthermore, the present invention also proposes to determine the passenger anxiety level of each passenger based on the target congestion mean value of each passenger, the slope of the fitted line, and the vehicle speed at each moment, including: The product of the target congestion mean of each passenger and the slope of the fitted straight line is determined as the congestion evaluation value of each passenger; The ratio of the vehicle speed at each moment to the road speed limit is constructed in chronological order to obtain the speed ratio sequence of the target vehicle; The passenger anxiety level of each passenger is determined using the travel distance, congestion evaluation value, travel time and speed proportion sequence of each passenger. The travel distance is used to represent the distance traveled by the vehicle from the moment the passenger gets on the vehicle to the current moment.

[0010] Furthermore, the present invention also proposes that the operating status information includes a preset total driving time, a preset driving route, an elapsed driving time, and a current location of the vehicle, and the road traffic information includes the congestion level of each route segment and the predicted continued driving time; Based on the operating status information and road traffic information of each vehicle, the future congestion impact factors of each vehicle are predicted, including: The predicted continued driving time of the target vehicle is divided by the difference between the preset total driving time and the driving time to obtain the travel overtime degree of the target vehicle, where the target vehicle is any vehicle; Determine the subsequent driving path of the target vehicle based on the preset driving route of the target vehicle and the current position of the vehicle; The future congestion impact factor of the target vehicle is determined by using the target vehicle's travel timeout degree, the route length and congestion degree of each route segment in the subsequent driving path.

[0011] Furthermore, the present invention also proposes to determine the current congestion impact factor of each vehicle based on the passenger anxiety level of each passenger in each vehicle and the passenger's travel time in each vehicle, including: The riding time of each passenger in the target vehicle is accumulated to obtain the cumulative riding time value of the target vehicle, where the target vehicle is any vehicle; Divide each passenger's travel time by the cumulative value of travel time to obtain the proportion of each passenger's travel time; The product of each passenger's travel time ratio and the passenger's anxiety level is accumulated to obtain the current congestion impact factor of the target vehicle.

[0012] Furthermore, the present invention also proposes that the operating status information includes the historical location of the vehicle; Based on the operating status information of each vehicle, the driving risk of each vehicle at the current moment is evaluated, including: Determine the driving vector at each moment based on the direction and distance between the target vehicle's historical position at each moment in a reference period and the vehicle's historical position at the next moment, where the target vehicle is any vehicle and the reference period is a preset time period before and closest to the current moment; Arrange each driving vector in chronological order to construct a driving vector sequence; Based on the angle value and the module length difference value formed between two adjacent driving vectors in the driving vector sequence, an angle sequence and a module length difference sequence are constructed respectively; Based on the angle sequence and the module length difference sequence, the driving risk of the target vehicle at the current moment is determined.

[0013] Furthermore, the present invention also proposes determining the driving risk of the target vehicle at the current moment based on the angle sequence and the module length difference sequence, including: Perform mean processing on each angle value in the angle sequence and the corresponding modulus length difference value in the modulus length difference value sequence to obtain the driving change sequence; Performing mean processing on each driving change amount in the driving change sequence to obtain a driving change mean; Marking a driving change amount in the driving change sequence that is greater than a preset change threshold as a dangerous driving change amount; The driving risk of the target vehicle at the current moment is determined by using the mean driving change value and the change amount of each dangerous driving.

[0014] Furthermore, the present invention also proposes that after determining the dispatch priority factor of each vehicle based on the passenger adverse experience factor and driving risk of each vehicle, the method further includes: Based on the road direction information of the target vehicle in the reference time period, the scheduling priority factor of the target vehicle is modified to obtain the scheduling resource allocation degree of the target vehicle. The target vehicle is any vehicle, and the reference time period is a preset time period before and closest to the current time. Based on the scheduling priority factors of each vehicle, resource scheduling is performed on each vehicle to obtain vehicle scheduling results, including: Based on the scheduling resource allocation degree of each vehicle, resource scheduling is performed on each vehicle to obtain the vehicle scheduling result.

[0015] Furthermore, the present invention also proposes to modify the scheduling priority factor of the target vehicle based on the road direction information of the target vehicle in the reference time period to obtain the scheduling resource allocation degree of the target vehicle, including: Obtain the road centerline of the target vehicle in the reference period; Draw perpendicular lines from the historical positions of the target vehicle at each moment in the reference period to the road centerline to construct an intersection point sequence, which includes the intersection points of each perpendicular line with the road centerline; Based on the direction and distance between two adjacent intersections in the intersection sequence, a road extension vector sequence is constructed, wherein the road extension vector sequence includes a road extension vector composed of the direction and distance between the two adjacent intersections; Based on the road extension vector sequence and the driving vector sequence of the target vehicle, the scheduling priority factor of the target vehicle is modified to obtain the scheduling resource allocation degree of the target vehicle.

[0016] The present invention has the following beneficial effects: In the vehicle dispatching method for a road passenger transport management system for a transportation terminal center provided by an embodiment of the present invention, the system first obtains real-time operating status information and road traffic information for each vehicle, comprehensively understanding the real-time dynamics of vehicles and roads. Based on this information, the system then evaluates each vehicle's current passenger experience factor and driving risk. This system considers vehicle conditions from the perspectives of passenger experience and driving safety, accurately identifying factors that may affect dispatch rationality. A dispatch priority factor is then determined for each vehicle based on the passenger experience factor and driving risk, assigning each vehicle a reasonable dispatch priority, fully considering the dispatch needs of different vehicles under different circumstances. Finally, resource dispatch is performed based on each vehicle's dispatch priority factor, resulting in a vehicle dispatch result. This dispatching method, which uses real-time data as a basis and comprehensively considers passenger experience and driving safety to determine dispatch priorities, enables differentiated dispatching based on the actual conditions of each vehicle, enabling timely response to various emergencies and changing traffic conditions. It effectively overcomes the scheduling irrationality caused by traditional dispatching methods that ignore individual vehicle differences, thereby improving the rationality of vehicle dispatch. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 A flow chart of a vehicle dispatching method for a road passenger transport management system in a transportation terminal center provided by one embodiment of the present invention; Figure 2 A first flow chart of S102 provided in one embodiment of the present invention; Figure 3 This is a second flow chart of S102 provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0019] To further illustrate the technical means and effectiveness of the present invention in achieving its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation, structure, features, and effectiveness of a vehicle dispatching method for a road passenger transport management system for a transportation terminal center proposed by the present invention. In the following description, different references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0020] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0021] It should be noted that the acquisition, storage, use, and processing of data in the technical solution of the present invention comply with the relevant provisions of laws and regulations.

[0022] It should be noted that in the embodiments of the present invention, certain software, components, models and other existing solutions in the industry may be mentioned. They should be regarded as exemplary and their purpose is only to illustrate the feasibility of implementing the technical solution of the present invention, but it does not mean that the applicant has or will necessarily use the solution.

[0023] In traditional road passenger transport management systems, dynamic scheduling decisions rely on real-time vehicle operating status and road traffic information, but lack a comprehensive quantitative assessment mechanism for passenger experience and driving risks. Road passenger transport management systems are unable to establish a correlation model between passenger anxiety and vehicle driving volatility, resulting in scheduling priorities relying solely on single-dimensional data and failing to accurately reflect the dynamic characteristics of a vehicle's actual operating status. This directly causes scheduling resource allocation to deviate from actual demand, resulting in both delays in high-priority vehicle scheduling and redundant resources for low-priority vehicles, reducing overall scheduling efficiency and passenger satisfaction.

[0024] When faced with the above problems, the present invention first realized that the existing road passenger transport management system is unable to effectively integrate multi-dimensional dynamic data, which leads to a fundamental contradiction in decision-making bias. Traditional methods rely only on single-dimensional parameters such as location and congestion. They can neither capture the nonlinear characteristics of passenger experience as the ride duration and road conditions change, nor can they identify the cumulative effect of driving behavior patterns on safety risks. To this end, the present invention attempts to establish a two-dimensional evaluation model, in which the passenger adverse experience factor needs to be associated with the real-time riding environment and the evolution of individual anxiety, and the driving risk should reflect the abnormal fluctuation characteristics of the vehicle's driving trajectory. Through analysis, it was found that simply increasing the frequency of data collection or increasing the type of sensors cannot solve the core problem. The key lies in establishing an evaluation index that can simultaneously quantify ride comfort and driving safety, and dynamically integrating the two to form a basis for scheduling decisions. After multiple rounds of scheme verification, it was finally determined that the scheduling priority can be dynamically adjusted by calculating the composite evaluation value of each vehicle in real time, which not only avoids the one-sidedness of a single indicator, but also achieves dynamic optimization of resource allocation.

[0025] In this regard, Figure 1 As shown, a flow chart of a vehicle dispatching method for a road passenger transport management system of a transportation terminal center is provided. The vehicle dispatching method for the road passenger transport management system of a transportation terminal center can be applied to the road passenger transport management system of a transportation terminal center. The vehicle dispatching method for the road passenger transport management system of a transportation terminal center can include the following S101 to S104.

[0026] S101, obtaining the operating status information and road traffic information of each vehicle in real time; S102, based on the operating status information and road traffic information of each vehicle, respectively assessing the passenger adverse experience factor of each vehicle at the current moment, and based on the operating status information of each vehicle, respectively assessing the driving risk of each vehicle at the current moment; S103, determining a dispatch priority factor for each vehicle based on the passenger adverse experience factor and driving risk of each vehicle; S104: Based on the scheduling priority factor of each vehicle, resource scheduling is performed on each vehicle to obtain a vehicle scheduling result.

[0027] In this embodiment, the operating status information refers to real-time data generated by the vehicle during operation. Specifically, it can be achieved by using on-board sensors or Global Positioning System (GPS) equipment to collect information such as vehicle position, vehicle speed, number of passengers, and driving time, which is used to reflect the current actual operating status of the vehicle.

[0028] Road traffic information refers to traffic condition data on a vehicle's route. Specifically, it can be achieved by obtaining information such as the congestion level of the route segment and the predicted duration of continued driving through a traffic monitoring system or a real-time navigation platform, and is used to evaluate road traffic efficiency.

[0029] The passenger adverse experience factor refers to an indicator that quantifies the degree of discomfort caused by congestion, delays and other factors during the ride. It can be calculated by analyzing the passenger's anxiety level, the length of time on the ride, and the parameters of future congestion impact factors. It is used to reflect the current negative impact of the vehicle on the passenger experience.

[0030] Driving risk refers to the level of safety risk that a vehicle faces while driving. It can be assessed by analyzing data such as changes in the vehicle's driving vector, sudden acceleration and deceleration behaviors, and is used to determine the safety of the vehicle's current driving state.

[0031] The dispatch priority factor refers to a comprehensive evaluation value used to determine the priority of vehicle resource allocation. It can be calculated by weighting or integrating the passenger adverse experience factor and driving risk, and is used to dynamically adjust the dispatch order of different vehicles.

[0032] Resource scheduling refers to the operation of planning vehicle routes and adjusting vehicle schedules based on priority. This can be achieved by using intelligent algorithms to match available resources with vehicle demand, thereby optimizing the operational efficiency of the overall passenger transport system.

[0033] The core innovation of this invention lies in the real-time collection of vehicle operating status information and road traffic information, the establishment of a two-dimensional evaluation model of passenger experience and driving safety, the dynamic calculation of scheduling priorities and the implementation of differentiated resource allocation, thereby achieving vehicle scheduling decisions that take into account both service quality and safety efficiency in complex traffic environments.

[0034] As an example, a road passenger transport management system collects real-time vehicle location, speed, passenger boarding and alighting information, and road congestion conditions through onboard terminals and roadside equipment. Vehicle location is obtained through GPS positioning, speed is measured by speed sensors, passenger information is collected through onboard cameras and passenger card swipe data, and road congestion conditions are analyzed using roadside cameras and vehicle driving data.

[0035] The collected data is then transmitted in real time to the data center. There, it undergoes preprocessing: data cleaning to remove duplicate records, correct erroneous data, and fill in missing values ​​to ensure data quality and accuracy. Data consolidation is then performed to combine data from different sources and formats into a unified dataset. Finally, the cleaned and consolidated data is stored in a database or data lake for subsequent analysis.

[0036] The road passenger transport management system then uses machine learning algorithms to process the collected data and calculate a passenger negative experience factor for each vehicle. Specifically, factors such as passenger ride duration, in-vehicle crowding, and road congestion are fed into a trained neural network model to output the passenger negative experience factor. Simultaneously, the driving risk index is calculated by analyzing changes in acceleration and direction along the vehicle's trajectory.

[0037] Next, the passenger experience factor and the driving risk index are multiplied to determine each vehicle's dispatch priority factor. Finally, vehicles are ranked based on the dispatch priority factors, with high-priority vehicles prioritized for dispatch, such as detours or priority passage. The road passenger transport management system recalculates and updates vehicle dispatch results at regular intervals to adapt to the dynamically changing traffic environment.

[0038] This embodiment first obtains real-time operating status information and road traffic information for each vehicle, comprehensively understanding the real-time dynamics of vehicles and roads. Based on this information, the system then evaluates each vehicle's current passenger experience factor and driving risk. This system considers vehicle conditions from the perspectives of passenger experience and driving safety, accurately identifying factors that may impact scheduling rationality. A scheduling priority factor is then determined for each vehicle based on the passenger experience factor and driving risk, assigning each vehicle a reasonable scheduling priority, fully considering the scheduling needs of different vehicles in different situations. Finally, resource scheduling is performed based on each vehicle's scheduling priority factor, resulting in a vehicle scheduling result. This scheduling method, which uses real-time data as a basis and comprehensively considers passenger experience and driving safety to determine scheduling priorities, enables differentiated scheduling tailored to each vehicle's actual situation, enabling timely response to various emergencies and changing traffic conditions. It effectively mitigates the scheduling inconsistencies caused by traditional scheduling methods that ignore individual vehicle differences, thereby improving the rationality of vehicle scheduling.

[0039] Some of the aforementioned solutions of the present invention propose evaluating the passenger experience factor based on vehicle operating status information and road traffic information. However, relying solely on static data at the current moment may result in the evaluation results failing to fully reflect the dynamic changes in the passenger experience during the ride. For example, the impact of congestion that passengers may face in the future is not effectively taken into account.

[0040] In this regard, Figure 2 As shown, the present invention further proposes that S102 includes the following S201 to S203: S201, determining the passenger anxiety level of each passenger in each vehicle based on the operating status information and road traffic information of each vehicle, and predicting the future congestion impact factor of each vehicle based on the operating status information and road traffic information of each vehicle; S202, determining the current congestion impact factor of each vehicle based on the passenger anxiety level of each passenger on each vehicle and the passenger's travel time on each vehicle, where the travel time represents the time from the moment the passenger boarded the vehicle to the current moment; S203 , performing average processing on the future congestion impact factor of each vehicle and the current congestion impact factor to obtain the passenger adverse experience factor of each vehicle at the current moment.

[0041] In this embodiment, passenger anxiety is an indicator used to quantify a passenger's level of anxiety during a ride. It comprehensively considers the impact of the vehicle's operating status and road traffic conditions on the passenger's psychology. For example, frequent braking and prolonged congestion may cause passengers to feel anxious. This indicator uses a specific algorithm to convert these factors into a specific value to measure the degree of anxiety.

[0042] The Future Congestion Impact Factor (FCIF) is a quantitative prediction of the potential congestion levels that a vehicle could cause over a specific period of time due to changes in road traffic conditions and its own operating state. Based on current operating status and road traffic information, it uses a specific prediction model to estimate the impact of a vehicle on congestion over a specific time period.

[0043] Elapsed time represents the length of time from the moment a passenger boarded the bus to the current moment. This time parameter is used to measure the length of time a passenger spends on the bus and, combined with their anxiety level, to assess the impact of current congestion on them.

[0044] The current congestion impact factor is calculated based on each passenger's anxiety level and their travel time. It quantifies the impact of current congestion on passengers. It reflects the direct impact of current congestion on the passenger experience.

[0045] As an example, a passenger anxiety assessment model can be established. Based on extensive experimental data and psychological research, this model can correlate different operating conditions and road traffic factors with passenger anxiety levels. For example, a higher anxiety weight is assigned when a vehicle brakes frequently and is in heavy congestion; a lower anxiety weight is assigned when the vehicle is traveling smoothly on unobstructed roads. For each passenger, the relevant information is input into the assessment model based on the vehicle's operating condition and the road traffic environment in which the passenger is riding, and the passenger anxiety value is calculated.

[0046] At the same time, a congestion prediction model is constructed based on historical and current vehicle operating status information and road traffic information. This congestion prediction model can utilize machine learning algorithms, such as time series analysis and neural networks. The model is trained using historical data to learn the relationship between different operating status and road traffic information and future congestion levels. The currently collected vehicle operating status and road traffic information are then input into the trained congestion prediction model to predict the future congestion impact factor for vehicles over a period of time. This future congestion impact factor can be quantified based on different levels of congestion, for example, with a lower factor value corresponding to mild congestion and a higher factor value corresponding to severe congestion.

[0047] Then, for each vehicle, the passenger anxiety level and ride duration are collected for each passenger. A current congestion impact factor calculation model is established. This current congestion impact factor calculation model can consider the weighted relationship between passenger anxiety level and ride duration on the current congestion. For example, the higher the passenger anxiety level and the longer the ride duration, the greater the contribution to the current congestion impact factor. For each passenger in the vehicle, the passenger's contribution to the current congestion impact factor is calculated according to the current congestion impact factor calculation model based on their passenger anxiety level and ride duration. The contribution values ​​of all passengers in the vehicle are aggregated (e.g., averaged or weighted summed) to obtain the current congestion impact factor for the vehicle.

[0048] Finally, for each vehicle, its future congestion impact factor and current congestion impact factor are obtained. These two factors are averaged, that is, their arithmetic mean is calculated, and the result is the passenger experience factor of the vehicle at the current moment.

[0049] This embodiment enables a comprehensive assessment of the passenger experience across all vehicles, taking into account both current realities and forecasting potential future changes. This allows for a more accurate understanding of passenger experience needs, providing a more reliable basis for subsequent vehicle scheduling decisions. This approach not only improves passenger satisfaction but also optimizes overall operational efficiency, reducing complaints and negative impacts caused by passenger dissatisfaction.

[0050] In some of the above-mentioned solutions of the present invention, methods for determining passenger anxiety based on operating status information and road traffic information are proposed. However, in the specific implementation process, there is still a problem in how to accurately quantify the anxiety level of each passenger caused by traffic congestion and vehicle load changes during the ride. The existing methods fail to effectively combine the congestion change trend and vehicle load dynamic changes during the passenger's ride period, resulting in inaccurate passenger anxiety assessment.

[0051] In this regard, the present invention further proposes that the operating status information includes the time when passengers board the vehicle, the number of passengers, and the vehicle's speed, and the road traffic information includes the congestion level of each route segment; S201 may specifically include: The time period from the boarding time to the current time of each passenger on the target vehicle is determined as the riding time of each passenger, and the target vehicle is any vehicle; The congestion degree of each passenger on the target vehicle during the riding period is constructed in chronological order to obtain a congestion degree sequence for each passenger; The ratio of the number of passengers at each moment to the maximum load is constructed in chronological order to obtain the passenger ratio sequence of the target vehicle; Taking the passenger proportion sequence as the weight sequence, perform linear fitting on each congestion degree sequence to obtain the slope of the fitted line corresponding to each passenger; The passenger anxiety level of each passenger is determined based on the target congestion level mean of each passenger, the slope of the fitted line, and the vehicle speed at each moment. The target congestion level mean is the mean of all congestion levels in the passenger's congestion level sequence.

[0052] In this embodiment, the congestion degree sequence records the congestion degree change trend of each route segment during the passenger's travel time. The passenger proportion sequence reflects the dynamic changes in vehicle load over time. The straight line fitting process uses the linear regression method to calculate the weighted rate of change of congestion degree over time. The slope of the fitted line represents the overall upward or downward trend of congestion degree during the passenger's travel time. The target congestion degree mean is the average value of the congestion degree sequence. The vehicle speed is used to measure the impact of vehicle driving efficiency on passenger anxiety.

[0053] Specifically, for each passenger on the target vehicle, the time period from the moment they boarded the vehicle to the current moment is first extracted. The real-time congestion levels for each route segment within this time period are then chronologically calculated, forming a congestion sequence. Simultaneously, a passenger share sequence is generated based on the ratio of the number of passengers on the vehicle to its maximum load capacity at each moment. A weighted linear fit is then performed on the congestion sequence using the passenger share sequence as a weight. The slope of the resulting fitted line reflects the weighted trend of congestion changes.

[0054] As an example, for each passenger on the target vehicle, first record the passenger's boarding time and current time. The time interval between the passenger's boarding time and the current time is determined as the passenger's boarding period. For example, if passenger A boarded the vehicle at 9:00 AM and the current time is 10:30 AM, then passenger A's boarding period is 9:00-10:30 AM.

[0055] Then, as the target vehicle is traveling, real-time congestion information for the route segment it is on is obtained. In this solution, the congestion level can be set to a value range of 1-10. A higher congestion level indicates a more congested route segment, while a lower value indicates a smoother route.

[0056] Specifically, congestion levels can be obtained through data interfaces with traffic management departments, onboard sensors (such as cameras and radar), and traffic flow analysis algorithms. For each passenger, the congestion levels of the route segments traveled during their travel time are arranged in chronological order to form a congestion sequence. For example, if passenger B's travel time was 8:00-9:00, and the vehicle traveled through route segment 1 (8:00-8:20, congestion level 3), route segment 2 (8:20-8:50, congestion level 5), and route segment 3 (8:50-9:00, congestion level 4) during this time, passenger B's congestion sequence would be [3, 5, 4].

[0057] At the same time, as the target vehicle is in motion, the number of passengers on board is counted in real time at each moment. This can be achieved using an internal passenger counting system (e.g., based on a weight sensor or infrared sensor). The ratio of the number of passengers to the maximum load at each moment is arranged in chronological order to form a passenger ratio sequence. For example, if the target vehicle has a maximum load of 50 people, and there are 30 passengers at time t1, 40 at time t2, and 35 at time t3, the passenger ratio sequence is [30 / 50, 40 / 50, 35 / 50], or [0.6, 0.8, 0.7].

[0058] Using the passenger percentage sequence as the weight sequence, a straight line fit is performed on each congestion sequence to obtain the slope of the fitted line for each passenger. Linear fitting can be performed using mathematical methods such as the least squares method. The target congestion mean is then calculated for each passenger, averaging the congestion levels in their congestion sequence. For example, if passenger C's congestion sequence is [2, 4, 3, 5], the target congestion mean is (2 + 4 + 3 + 5) / 4 = 3.5.

[0059] The more passengers a vehicle has and the more congested it is, the more likely it is to cause anxiety. This is because when there are many passengers, personal space is compressed, making passengers feel crowded and uncomfortable. A large number of passengers also leads to poor air circulation, increased carbon dioxide concentration, and relatively low oxygen levels. Furthermore, multiple passengers can create more noise and distractions, such as conversations and ringing phones. These sounds can be more jarring in a closed space, increasing passengers' annoyance. If the vehicle is congested at this time, this can further increase passenger anxiety. Therefore, a weighted straight line fit is performed, using the ratio of the number of passengers at each moment to the maximum load as the weight for the congestion level at that moment.

[0060] Finally, a passenger anxiety calculation model was established by comprehensively considering the target congestion mean, the slope of the fitted line, and the vehicle speed at each time. Each passenger's target congestion mean, the slope of the fitted line, and the vehicle speed at each time were substituted into the passenger anxiety calculation model to obtain each passenger's passenger anxiety level.

[0061] Through this embodiment, the anxiety level of each passenger can be assessed more accurately. As a result, vehicle scheduling can be carried out more reasonably, improving passenger satisfaction. Specifically, by considering multiple factors such as passenger boarding time, number of passengers, vehicle speed, and route congestion, the actual passenger experience can be fully reflected. Furthermore, by constructing a congestion degree series and a passenger ratio series, the dynamic changes in congestion conditions and the number of passengers can be captured. Finally, through linear fitting and comprehensive consideration of multiple indicators, a more objective and accurate passenger anxiety assessment result can be obtained. This method can provide a more reliable basis for subsequent vehicle scheduling, thereby optimizing the overall passenger service quality.

[0062] Some of the aforementioned solutions proposed methods for determining passenger anxiety by calculating the slope of a fitted line using passenger congestion score sequences and passenger share sequences, and then combining this with the target congestion score mean and vehicle speed. However, this process fails to effectively quantify the volatility of vehicle speed, resulting in an inability to accurately reflect the dynamic impact of speed fluctuations during actual rides, which in turn affects the accuracy of vehicle dispatch priority factors.

[0063] In this regard, the present invention further proposes to determine the passenger anxiety level of each passenger based on the target congestion mean value of each passenger, the slope of the fitted line, and the vehicle speed at each moment, including: The product of the target congestion mean of each passenger and the slope of the fitted straight line is determined as the congestion evaluation value of each passenger; The ratio of the vehicle speed at each moment to the road speed limit is constructed in chronological order to obtain the speed ratio sequence of the target vehicle; The passenger anxiety level of each passenger is determined using the travel distance, congestion evaluation value, travel time and speed proportion sequence of each passenger. The travel distance is used to represent the distance traveled by the vehicle from the moment the passenger gets on the vehicle to the current moment.

[0064] In this embodiment, the congestion evaluation value quantifies the cumulative effect and changing trend of congestion levels during a passenger's ride by multiplying the target congestion mean by the slope of the fitted straight line. The speed share sequence constructs a dynamic indicator of the psychological impact of speed fluctuations on passengers by measuring the ratio of vehicle speed to road speed limit. The distance traveled is combined with the duration of the ride to measure the passenger's comprehensive perception of spatial displacement and time consumption during the vehicle's travel. The coordinated calculation of the speed share sequence and the congestion evaluation value realizes a multi-dimensional dynamic assessment of passenger anxiety.

[0065] Specifically, the mean of each passenger's congestion score series is first multiplied by the slope of the fitted line to generate a congestion rating that reflects the degree of congestion and its rate of change. Next, the ratio of vehicle speed to the road speed limit is arranged chronologically to form a speed percentage series, which characterizes the fluctuation in speed compliance with the road speed limit during driving. Subsequently, the distance traveled is used as a spatial dimension parameter and weighted with the time dimension parameter of the congestion rating. This is then combined with the cumulative effect of travel time, and finally corrected using the dynamic adjustment factor of the speed percentage series to obtain the final assessment of passenger anxiety.

[0066] As an example, the passenger anxiety level of a passenger may be determined by the following formula 1: Formula 1 In formula 1, F is used to represent the passenger's anxiety, L is used to represent the distance the passenger has traveled, and T is used to represent the passenger's travel time. It is used to represent the mean speed share of the speed share sequence of the target vehicle, D is used to represent the passenger's congestion evaluation value, and norm is used to represent the linear normalization processing.

[0067] in, The smaller it is, the longer it takes to travel a short distance, which is more likely to cause anxiety. The closer it is to 1, the closer the vehicle speed is to the road speed limit, which means that the vehicle is following traffic rules and the time spent in a short distance is relatively long, which is normal. The larger the Turn down.

[0068] Furthermore, since the calculated passenger anxiety level may have an unreasonable value range, normalization can be performed to make the calculated passenger anxiety level easier to understand and compare. This normalization method can employ a maximum-minimum normalization method, a well-known technique that will not be elaborated upon here. Through this normalization process, the passenger anxiety level values ​​can be made more standardized and reasonable, better meeting the needs of practical applications.

[0069] This embodiment comprehensively considers multiple factors, including the distance traveled, congestion conditions, ride duration, and vehicle speed, to more accurately assess a passenger's anxiety level. This provides a more reliable basis for subsequent vehicle dispatch decisions, helping to improve the passenger experience and enhance the quality of passenger transport services.

[0070] In some of the above-mentioned solutions of the present invention, it is proposed to predict the impact factors of future congestion based on vehicle operating status information and road traffic information. However, the risk of vehicle trip timeout and the congestion distribution of subsequent driving paths are not combined in the prediction process, resulting in a deviation between the prediction results of future congestion impact and actual traffic conditions.

[0071] In this regard, the present invention further proposes that the operating status information includes a preset total driving time, a preset driving route, an elapsed driving time, and a current vehicle location, and the road traffic information includes the congestion level of each route segment and the predicted continued driving time; S201 may also include: The predicted continued driving time of the target vehicle is divided by the difference between the preset total driving time and the driving time to obtain the travel overtime degree of the target vehicle, where the target vehicle is any vehicle; Determine the subsequent driving path of the target vehicle based on the preset driving route of the target vehicle and the current position of the vehicle; The future congestion impact factor of the target vehicle is determined by using the target vehicle's travel timeout degree, the route length and congestion degree of each route segment in the subsequent driving path.

[0072] In this embodiment, the trip overtime risk is calculated by comparing the predicted travel time to the remaining planned travel time, quantifying the risk of trip delay. The subsequent travel path is determined by the topological relationship between the preset route and the vehicle's current location, ensuring the accuracy of the route prediction.

[0073] Specifically, the target vehicle's trip overtime risk is calculated by dividing the predicted continued driving time by the difference between the preset total driving time and the time already driven. For example, if the vehicle has been driving for 2 hours, the preset total driving time is 5 hours, and the predicted continued driving time is 3.5 hours, the trip overtime risk is 3.5 / (5-2)=1.17, indicating a 17% risk of overtime. The subsequent driving path is determined by the unfinished sections of the preset driving route. For example, if the vehicle is currently at route point A and the preset route is A→B→C→D, the subsequent path is B→C→D.

[0074] As an example, the future congestion impact factor of a vehicle can be determined by the following formula 2: Formula 2 In formula 2, Used to characterize the future congestion impact factor of vehicles, Used to characterize the predicted driving time of the vehicle, Used to characterize the total preset driving time of the vehicle, Used to represent the driving time of the vehicle. It is used to represent the length of the jth route segment in the subsequent driving path, and N is used to represent the total length of the route of the subsequent driving path. It is used to represent the congestion degree of the jth route segment in the subsequent driving path, and J is used to represent the total number of route segments in the subsequent driving path.

[0075] in, The larger it is, the more serious the trip delay is and the worse the passenger experience is. The bigger, The bigger; As The weight of the congested route is longer, that is, the greater the congestion impact is, which leads to The bigger.

[0076] Through this embodiment, it is possible to accurately predict the congestion that a vehicle may encounter in the future, providing an important reference basis for subsequent vehicle scheduling. This prediction method based on multi-dimensional data takes into account factors such as the vehicle's travel progress, route planning, and road congestion, and can more comprehensively evaluate the vehicle's future operating conditions. By introducing the concept of travel timeout, the solution can effectively reflect whether a vehicle is at risk of delay, thereby giving priority to these vehicles in scheduling decisions. At the same time, by combining the specific conditions of the subsequent driving path, the solution can weight the congestion impact of different road sections, thereby improving the accuracy and pertinence of the prediction. This prediction method provides the vehicle scheduling system with a more refined and dynamic decision-making basis, which helps to improve overall scheduling efficiency, reduce passenger waiting time, and optimize road resource utilization.

[0077] In some of the above-mentioned solutions of the present invention, it is proposed to determine the current congestion impact factor based on the passenger anxiety level and the length of time traveled. However, when integrating the passenger anxiety levels of multiple passengers, if only simple accumulation or averaging is performed, it is difficult to accurately reflect the differences in the degree of contribution of different passengers to the overall congestion due to differences in the length of time traveled.

[0078] In this regard, the present invention further proposes that S202 may specifically include: The riding time of each passenger in the target vehicle is accumulated to obtain the cumulative riding time value of the target vehicle, where the target vehicle is any vehicle; Divide each passenger's travel time by the cumulative value of travel time to obtain the proportion of each passenger's travel time; The product of each passenger's travel time ratio and the passenger's anxiety level is accumulated to obtain the current congestion impact factor of the target vehicle.

[0079] In this embodiment, the calculation of accumulated travel time quantifies the overall load of a group of passengers' travel status through the time dimension. Determining the travel time percentage converts individual travel time into a weight coefficient relative to the overall load. The product-accumulation operation quantifies individual impact differences through a linear combination of the weight coefficient and the passenger's anxiety level. For example, if a passenger's travel time accounts for 30% of the total accumulated value and their passenger anxiety level is 0.8, their contribution to the current congestion impact factor is 0.24.

[0080] Specifically, the target vehicle collects each passenger's boarding time in real time via an onboard terminal and calculates the elapsed time from that boarding time to the current time. The elapsed time of all passengers is accumulated to form a cumulative value, which reflects the cumulative time effect of the vehicle's current passenger load. Each passenger's elapsed time contribution represents their relative weight in the overall time load. When a passenger's elapsed time is significantly longer than others, their contribution increases significantly. Each passenger's anxiety level is multiplied by their contribution and then added together. This strengthens the contribution of passengers with long rides and high anxiety levels to the current congestion impact factor. For example, if there are two passengers in the vehicle with elapsed times of 60 minutes and 30 minutes, respectively, the cumulative value is 90 minutes, with their contributions being 66.7% and 33.3%, respectively. If their anxiety levels are 0.9 and 0.6, respectively, the current congestion impact factor is calculated as 0.9 × 0.667 + 0.6 × 0.333 = 0.8, which more accurately reflects the actual impact difference than the 0.75 obtained by the simple average method.

[0081] As an example, the time spent on board the target vehicle by each passenger is accumulated to obtain the cumulative time spent on board the target vehicle. For example, if there are three passengers on the target vehicle and their time spent on board is 20 minutes, 30 minutes, and 40 minutes respectively, the cumulative time spent on board is 90 minutes.

[0082] Then, divide each passenger's travel time by the cumulative travel time to get each passenger's travel time share. Continuing with the above example, the travel time shares of the three passengers are 2 / 9, 1 / 3, and 4 / 9, respectively.

[0083] Finally, the product of each passenger's travel time percentage and their anxiety level is added together to obtain the current congestion impact factor for the target vehicle. Assuming the anxiety levels of the three passengers are 0.6, 0.8, and 0.7, respectively, the current congestion impact factor is (2 / 9 × 0.6 + 1 / 3 × 0.8 + 4 / 9 × 0.7) = 0.7.

[0084] This embodiment can comprehensively consider the ride duration and anxiety level of each passenger on the vehicle, more accurately assessing the impact of current congestion on the vehicle. This can provide a more reasonable basis for subsequent vehicle scheduling decisions, helping to improve the overall passenger service quality and passenger satisfaction.

[0085] In some of the above-mentioned schemes of the present invention, the assessment of driving risk mainly relies on the real-time operating status information of the vehicle, but fails to effectively capture the sudden changes in direction and speed during driving, resulting in insufficient accuracy in identifying the dynamic changes in the risk of driving behavior, affecting the accuracy of the scheduling priority factor.

[0086] In this regard, Figure 3 As shown, the present invention further proposes that the operating status information includes the vehicle's historical location; S102 may further include the following S301 to S304: S301, determining a driving vector at each moment based on the direction and distance between the historical vehicle position of the target vehicle at each moment in a reference time period and the historical vehicle position at the next moment, where the target vehicle is any vehicle and the reference time period is a preset time period before and closest to the current moment; S302, arranging the driving vectors in chronological order to construct a driving vector sequence; S303, constructing an angle sequence and a modulus difference sequence based on the angle values ​​and modulus difference values ​​formed between two adjacent driving vectors in the driving vector sequence; S304: Determine the driving risk of the target vehicle at the current moment based on the angle sequence and the module length difference sequence.

[0087] In this embodiment, travel vectors are generated from a time series of historical vehicle positions, encompassing both direction and distance. The travel vector sequence is constructed using a time window mechanism, selecting historical position data from a recent preset time period. The angle sequence reflects the severity of changes in vehicle direction, while the modulus difference sequence characterizes the suddenness of speed changes.

[0088] Specifically, the onboard positioning device continuously collects the target vehicle's historical position data within a reference period. The position differences between adjacent moments constitute a driving vector. After the driving vector sequence is sorted by time, the direction angle and module length differences between adjacent driving vectors are calculated to form a characteristic sequence that reflects the dynamic changes in driving behavior.

[0089] As an example, the last five minutes can be selected as the reference period. For the target vehicle, the vehicle's historical location is recorded every 10 seconds, resulting in 30 historical location points. The direction angle and distance between two adjacent historical location points are calculated to obtain 29 driving vectors. These driving vectors are then arranged in chronological order to form a driving vector sequence.

[0090] Furthermore, the angle between two adjacent driving vectors is calculated to obtain 28 angle values, forming an angle sequence. Simultaneously, the difference between the modulus lengths of two adjacent driving vectors is calculated to obtain 28 modulus length difference values, forming a modulus length difference sequence.

[0091] Therefore, a machine learning algorithm can be used to train a driving risk assessment model based on the angle sequence and the module length difference sequence. By inputting the current angle sequence and module length difference sequence into the driving risk assessment model, the driving risk score of the target vehicle at the current moment can be obtained.

[0092] This embodiment enables a comprehensive characterization of a vehicle's driving trajectory by constructing a driving vector sequence, an angle sequence, and a module length difference sequence based on historical vehicle location data. This sequence data can then be used to assess driving risk, accurately reflecting the vehicle's actual driving conditions and effectively identifying dangerous driving behavior. This approach avoids the limitations of single-metric assessments and improves the accuracy and reliability of driving risk assessments. Furthermore, using a fixed reference period for assessment allows for timely capture of changes in driving behavior, enabling dynamic monitoring of driving risk. This provides an important basis for subsequent vehicle scheduling decisions and helps improve road traffic safety.

[0093] Some of the aforementioned solutions of the present invention propose constructing an angle sequence and a modulus difference sequence based on the angle and modulus difference between two adjacent driving vectors in a driving vector sequence, respectively, to determine driving risk. However, independent analysis of angle and modulus difference alone cannot effectively distinguish between normal and dangerous driving variations, potentially leading to errors in the driving risk assessment.

[0094] In this regard, the present invention further proposes that S304 may specifically include: Performing mean processing on each angle value in the angle sequence and the corresponding modulus length difference value in the modulus length difference value sequence to obtain multiple driving variation values; Arrange each driving change in chronological order to construct a driving change sequence; Performing mean processing on each driving change amount in the driving change sequence to obtain a driving change mean; Marking a driving change amount in the driving change sequence that is greater than a preset change threshold as a dangerous driving change amount; The continuous dangerous driving change quantities in the driving change sequence are used to construct a dangerous driving change quantum sequence; The driving risk of the target vehicle at the current moment is determined by using the driving change mean and the quantum sequence of each dangerous driving change.

[0095] In this embodiment, averaging is performed using the arithmetic mean of the angle and the difference between the modulus lengths, enabling driving variations to simultaneously reflect the combined effects of directional and speed changes. The driving variation mean is calculated using a sliding window or overall sequence average algorithm to characterize the overall driving stability of the target vehicle within a reference period. A preset variation threshold is set based on historical accident data or experimental data to screen out driving variations that deviate from normal driving patterns. The number of dangerous driving variations and the degree of deviation serve as quantitative indicators, combined with the driving variation mean to determine driving risk.

[0096] Specifically, the angles and modulus differences between adjacent vectors in a driving vector sequence are combined into a single driving variation, eliminating the limitations of single-parameter analysis. The driving variation sequence is averaged to obtain an overall trend, which serves as a benchmark for determining dangerousness. Dangerous driving variation is flagged based on a preset threshold, effectively identifying abnormal driving behaviors such as sharp turns, sudden acceleration, or deceleration. The combined evaluation of the driving variation mean and dangerous driving variation considers overall driving stability while also capturing unexpected and dangerous maneuvers.

[0097] As an example, the driving risk of the vehicle at the current moment can be determined by the following formula 3: Formula 3 In formula 3, W is used to represent the driving risk of the vehicle at the current moment. Used to represent the mean driving variation. e is used to identify the sequence number of the dangerous driving variation quantum sequence constructed from the continuous dangerous driving variation. Specifically, we divide multiple continuous dangerous driving variation quantities into a subsequence, thereby obtaining multiple dangerous driving variation quantum sequences, each of which corresponds to a sequence number. The value of e ranges from 1 to E, and E is used to represent the total number of dangerous driving variation quantum sequences constructed from the continuous dangerous driving variation quantities. It is used to characterize the maximum value of the dangerous driving change in the dangerous driving change quantum sequence constructed by continuous dangerous driving change. It is used to characterize the length of the dangerous driving change quantum sequence constructed by continuous dangerous driving change quantities.

[0098] in, The larger the value, the longer the continuous emergency steering and braking. The larger the value is, the greater the degree of continuous emergency steering and braking during the period of continuous emergency steering and braking. As The weight of .

[0099] This embodiment dynamically identifies sudden changes in vehicle behavior during driving. By quantitatively analyzing sudden changes in direction and speed fluctuations between consecutive driving vectors, it accurately distinguishes normal driving patterns from potentially dangerous behaviors. This solution can determine driving risk levels based on objective data, avoiding the subjective errors of human judgment, thereby providing a precise risk assessment basis for vehicle scheduling decisions.

[0100] In some of the aforementioned solutions, each vehicle's dispatch priority factor is determined based on its passenger adverse experience factor and driving risk, and resources are dispatched directly based on this dispatch priority factor. However, this approach fails to account for resource allocation bias caused by inconsistencies between vehicle trajectories and road directions, resulting in low accuracy in the dispatch priority factor.

[0101] In this regard, the present invention further proposes that after S103, the vehicle dispatching method of the traffic terminal center road passenger transport management system may further include: Based on the road direction information of the target vehicle in the reference time period, the scheduling priority factor of the target vehicle is modified to obtain the scheduling resource allocation degree of the target vehicle. The target vehicle is any vehicle, and the reference time period is a preset time period before and closest to the current time. S104 may specifically include: Based on the scheduling resource allocation degree of each vehicle, resource scheduling is performed on each vehicle to obtain the vehicle scheduling result.

[0102] In this embodiment, the target vehicle refers to any passenger vehicle to be dispatched and managed in the traffic station center road passenger transport management system. In actual operation, each vehicle entering the management range is identified and tracked so that dispatch-related operations can be carried out separately for each vehicle.

[0103] The reference period is set to the preset time period closest to the current time and before the current time. For example, the preset time period can be 15 minutes, 30 minutes, etc. The specific duration can be adjusted based on actual needs and system performance. This time period is selected to obtain information about the vehicle's recent road conditions, which can better reflect the impact of the current road environment on the vehicle's driving.

[0104] As an example, a vehicle-mounted positioning device and sensors (such as a gyroscope and accelerometer) can be used to collect road direction information. The positioning device can accurately obtain the vehicle's position coordinates. A series of position coordinate points can be used to depict the vehicle's trajectory, thereby determining the curvature of the road and the number of curves.

[0105] Sensors such as gyroscopes and accelerometers can measure a vehicle's posture and motion, such as steering angle and acceleration changes. This data helps more accurately analyze the vehicle's driving conditions on the road, particularly dynamic information during turns. For example, frequent and large steering angle changes may indicate the vehicle is traveling on a winding road.

[0106] Based on historically collected road direction information, a model is developed to modify the dispatch priority factor. This model then determines the dispatch resource allocation based on the target vehicle's road direction during the reference period. This model can take into account multiple factors related to road direction, such as the number of curves, curve curvature, and road slope.

[0107] This embodiment effectively addresses the problem of scheduling resource allocation deviations caused by inconsistencies between vehicle driving trajectories and road directions. By accurately identifying the degree of deviation between a vehicle's driving path and the planned road direction, the scheduling priority factor is corrected in real time. This prioritizes vehicles that deviate from the planned road direction during resource allocation, avoiding wasted scheduling resources due to abnormal vehicle driving and improving the accuracy of matching resource allocation results with the actual road direction.

[0108] In some of the above-mentioned schemes of the present invention, it is proposed to correct the scheduling priority factor based on the road direction information of the target vehicle in the reference time period. However, the degree of deviation between the actual driving trajectory of the vehicle and the center line of the road is not fully considered during the correction process, resulting in insufficient accuracy of the corrected scheduling resource allocation.

[0109] In this regard, the present invention further proposes to modify the scheduling priority factor of the target vehicle based on the road direction information of the target vehicle in the reference time period to obtain the scheduling resource allocation degree of the target vehicle, including: Obtain the road centerline of the target vehicle in the reference period; Drawing perpendicular lines from the historical positions of the target vehicle at each moment in the reference period to the center line of the road to construct an intersection point sequence, wherein the intersection point sequence includes the intersection points of each perpendicular line with the center line of the road; Based on the direction and distance between two adjacent intersections in the intersection sequence, a road extension vector sequence is constructed, wherein the road extension vector sequence includes a road extension vector composed of the direction and distance between the two adjacent intersections; Based on the road extension vector sequence and the driving vector sequence of the target vehicle, the scheduling priority factor of the target vehicle is modified to obtain the scheduling resource allocation degree of the target vehicle.

[0110] In this embodiment, the road centerline is extracted using a geographic information system or high-precision map data, representing the standard geometric centerline of the road. The intersection sequence is generated by perpendicularly projecting the vehicle's historical position onto the road centerline. Each intersection corresponds to the projected position of the vehicle at a certain moment within a reference period. The road extension vector sequence consists of the direction and distance between adjacent intersections, with the direction calculated from the azimuth between the two points and the distance calculated from the coordinate difference. The travel vector sequence consists of the direction and distance between the vehicle's historical position and the next position at the next moment.

[0111] Specifically, during a reference period, the target vehicle's actual trajectory is represented by a driving vector sequence, while the road's orientation is represented by a road extension vector sequence. By comparing the differences between these two vector sequences, the degree of deviation of the vehicle's trajectory from the road centerline can be quantified. A high degree of deviation indicates that the vehicle may be frequently changing lanes or deviating from its planned route, necessitating the prioritization of scheduling resources.

[0112] As an example, the scheduling resource allocation degree of a vehicle can be determined by the following formula 4: Formula 4 In formula 4, V is used to represent the scheduling resource allocation degree of the vehicle. It is used to represent the mean value of the angle between the road extension vector in the road extension vector sequence and the corresponding driving vector in the driving vector sequence. Y is used to represent the vehicle's scheduling priority factor.

[0113] in, The smaller the value, the more the driving direction is in line with the road extension direction, and the scheduling priority factor needs to be reduced. The scheduling resource allocation degree is obtained as the adjustment value of the scheduling priority factor of the vehicle at the current moment.

[0114] This embodiment effectively addresses the problem of traditional dispatching methods failing to consider the degree of compatibility between a vehicle's actual driving trajectory and the planned road course. By quantitatively analyzing the spatial relationship between a vehicle's driving path and the road centerline, it accurately identifies abnormal driving behavior that deviates from the planned route. The resulting dynamic correction mechanism optimizes the spatial distribution of station dispatching resources and reduces dispatching resource mismatches caused by vehicles deviating from the planned route.

[0115] It should be understood that the present invention is not limited to the specific configurations and processes described above and illustrated in the figures. For the sake of brevity, a detailed description of known methods is omitted. In the above embodiments, several specific steps are described and illustrated as examples. However, the method of the present invention is not limited to the specific steps described and illustrated. Those skilled in the art may make various changes, modifications, and additions, or change the order of the steps after understanding the spirit of the present invention.

[0116] It should also be noted that the exemplary embodiments described herein describe methods or systems based on a series of steps or devices. However, the present invention is not limited to the order of the steps described above. In other words, the steps may be performed in the order described in the embodiments, or in a different order, or several steps may be performed simultaneously.

[0117] The above description is only a specific embodiment of the present invention. Those skilled in the art will clearly understand that for the convenience and brevity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the protection scope of the present invention is not limited to this. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention.

Claims

1. A vehicle dispatching method for a road passenger transport management system in a transportation station center, characterized in that: The method comprises: Obtain the operating status information and road traffic information of each vehicle in real time; Based on the operating status information and road traffic information of each vehicle, respectively assessing the passenger adverse experience factor of each vehicle at the current moment, and based on the operating status information of each vehicle, respectively assessing the driving risk of each vehicle at the current moment; Determining a dispatch priority factor for each vehicle based on the passenger adverse experience factor and the driving risk of each vehicle; Based on the scheduling priority factors of the vehicles, resource scheduling is performed on the vehicles to obtain a vehicle scheduling result.

2. The vehicle dispatching method for a traffic terminal center road passenger transport management system according to claim 1, characterized in that: The step of evaluating the passenger adverse experience factor of each vehicle at the current moment based on the operating status information and road traffic information of each vehicle includes: determining a passenger anxiety level of each passenger in each of the vehicles based on the operating status information and road traffic information of each of the vehicles, and predicting a future congestion impact factor of each of the vehicles based on the operating status information and road traffic information of each of the vehicles; Determining the current congestion impact factor of each vehicle based on the passenger anxiety level of each passenger in each vehicle and the passenger's travel time in each vehicle, wherein the travel time is used to represent the time from the moment the passenger boarded the vehicle to the current moment; The future congestion impact factor of each vehicle and the current congestion impact factor are averaged to obtain the passenger adverse experience factor of each vehicle at the current moment.

3. The vehicle dispatching method for a traffic terminal center road passenger transport management system according to claim 2, characterized in that: The operation status information includes the time when passengers board the vehicle, the number of passengers, and the vehicle's speed; the road traffic information includes the congestion level of each route segment; Determining the passenger anxiety level of each passenger in each of the vehicles based on the operating status information and road traffic information of each of the vehicles includes: The time period from the boarding time to the current time of each passenger on the target vehicle is determined as the riding time period of each passenger, and the target vehicle is any vehicle; Constructing a congestion degree sequence for each passenger on the target vehicle according to the time sequence of the route segment where each passenger has been traveling during the period of time; The ratio of the number of passengers to the maximum load at each moment is constructed in chronological order to obtain a passenger ratio sequence of the target vehicle; Using the passenger proportion sequence as a weight sequence, performing a straight line fitting on each of the congestion degree sequences to obtain the slope of the fitting line corresponding to each passenger; The passenger anxiety level of each passenger is determined based on the target congestion level mean of each passenger, the slope of the fitting line, and the vehicle speed at each moment. The target congestion level mean is the mean of the congestion levels in the passenger's congestion level sequence.

4. The vehicle dispatching method for a traffic terminal center road passenger transport management system according to claim 3, characterized in that: The step of determining the passenger anxiety level of each passenger based on the target congestion average of each passenger, the slope of the fitted straight line, and the vehicle speed at each moment includes: The product of the target congestion degree mean of each passenger and the slope of the fitted straight line is determined as the congestion evaluation value of each passenger; The ratio of the vehicle's speed at each moment to the road speed limit is constructed in chronological order to obtain a speed ratio sequence of the target vehicle; The passenger anxiety level of each passenger is determined using the distance traveled by each passenger, the congestion evaluation value, the duration of the ride, and the speed proportion sequence of the target vehicle. The distance traveled is used to represent the distance traveled by the vehicle from the time the passenger boarded the vehicle to the current time.

5. The vehicle dispatching method for a traffic terminal center road passenger transport management system according to claim 2, characterized in that: The operating status information includes the preset total driving time, the preset driving route, the driving time and the current location of the vehicle; the road traffic information includes the congestion level of each route segment and the predicted continued driving time; The predicting of the future congestion impact factor of each vehicle based on the operating status information and road traffic information of each vehicle includes: Dividing the predicted continued driving time of the target vehicle by the difference between the preset total driving time and the actual driving time to obtain the travel overtime degree of the target vehicle, where the target vehicle is any one vehicle; Determining a subsequent driving path of the target vehicle based on the preset driving route of the target vehicle and the current position of the vehicle; A future congestion impact factor of the target vehicle is determined by using the travel overtime degree of the target vehicle, the route length and the congestion degree of each route segment in the subsequent driving path.

6. The vehicle dispatching method for a traffic terminal center road passenger transport management system according to claim 2, characterized in that: The determining of the current congestion impact factor of each vehicle based on the passenger anxiety level of each passenger in each vehicle and the riding time of the passengers in each vehicle includes: Accumulating the riding time of each passenger in the target vehicle to obtain a cumulative riding time value of the target vehicle, where the target vehicle is any one vehicle; Divide the travel time of each passenger by the cumulative value of the travel time to obtain the travel time ratio of each passenger; The product of the proportion of the riding time of each passenger and the passenger anxiety level is accumulated to obtain the current congestion impact factor of the target vehicle.

7. The vehicle dispatching method for a traffic terminal center road passenger transport management system according to any one of claims 1 to 6, characterized in that: The operating status information includes the vehicle's historical location; The step of evaluating the driving risk of each vehicle at a current moment based on the operating status information of each vehicle includes: Determining a driving vector at each moment based on the direction and distance between the target vehicle's historical position at each moment in a reference time period and the vehicle's historical position at the next moment, where the target vehicle is any vehicle and the reference time period is a preset time period before and closest to the current moment; Arranging the driving vectors in chronological order to construct a driving vector sequence; Based on the angle values ​​and the module length differences formed between two adjacent driving vectors in the driving vector sequence, respectively constructing an angle sequence and a module length difference sequence; Based on the angle sequence and the module length difference sequence, the driving risk of the target vehicle at the current moment is determined.

8. The vehicle dispatching method for a traffic terminal center road passenger transport management system according to claim 7, characterized in that: The determining, based on the angle sequence and the module length difference sequence, of the driving risk of the target vehicle at the current moment includes: Performing mean processing on each of the included angle values ​​in the included angle sequence and the corresponding module length difference value in the module length difference value sequence to obtain a plurality of driving variation values; Arranging the driving change amounts in chronological order to construct a driving change sequence; performing mean processing on each of the driving change amounts in the driving change sequence to obtain a driving change mean; marking the driving variation in the driving variation sequence that is greater than a preset variation threshold as a dangerous driving variation; constructing a dangerous driving change quantum sequence by using the continuous dangerous driving change quantities in the driving change sequence; The driving risk of the target vehicle at the current moment is determined by using the driving change mean and each of the dangerous driving change quantum sequences.

9. The vehicle dispatching method for a road passenger transport management system in a transportation terminal center according to any one of claims 1 to 6, characterized in that: After determining the dispatch priority factor of each vehicle based on the passenger adverse experience factor and the driving risk of each vehicle, the method further includes: Based on the road direction information of the target vehicle in a reference time period, the scheduling priority factor of the target vehicle is modified to obtain the scheduling resource allocation degree of the target vehicle, wherein the target vehicle is any vehicle and the reference time period is a preset time period before the current moment and closest to the current moment; The resource scheduling is performed on each vehicle based on the scheduling priority factor of each vehicle to obtain a vehicle scheduling result, including: Based on the scheduling resource allocation degree of each vehicle, resource scheduling is performed on each vehicle to obtain a vehicle scheduling result.

10. The vehicle dispatching method for a traffic terminal center road passenger transport management system according to claim 9, characterized in that: The method of modifying the dispatch priority factor of the target vehicle based on the road direction information of the target vehicle in the reference time period to obtain the dispatch resource allocation degree of the target vehicle includes: Obtain the road centerline of the target vehicle in the reference period; Drawing perpendicular lines from the historical positions of the target vehicle at various times during a reference period to the center line of the road to construct an intersection sequence, wherein the intersection sequence includes intersection points of each perpendicular line with the center line of the road; constructing a road extension vector sequence based on the direction and distance between two adjacent intersections in the intersection sequence, wherein the road extension vector sequence includes a road extension vector formed by the direction and distance between two adjacent intersections; Based on the road extension vector sequence of the target vehicle and the driving vector sequence of the target vehicle, the scheduling priority factor of the target vehicle is corrected to obtain the scheduling resource allocation degree of the target vehicle.

Images