A transport management system for freight transport using urban public transport

Abstract

The application discloses a kind of transport management system for freight using urban public transport, it is related to data management technical field, including freight task management module, intelligent scheduling module, goods storage and fixed management module and information interaction module, system passes through intelligent scheduling algorithm, in combination with public transport operation timetable, passenger flow density and dynamic running characteristics, optimize freight route, ensure that passenger and freight coexist without affecting passenger travel;Based on spatial layout and vehicle running characteristics, the application designs a goods storage optimization strategy, improves transport stability, prevents goods loss and potential safety hazard;At the same time, information interaction module monitors goods storage state in real time, and provides abnormal alarm, ensure the safety, reliability and operation efficiency of goods transport, so as to realize the intelligent integration of public transport and urban freight.

Description

Technical Field

[0001] This invention relates to the field of data management technology, and more specifically to a transportation management system for freight transport using urban public transportation. Background Technology

[0002] With rapid urbanization, urban logistics demand is constantly increasing, especially with the rise of e-commerce and express delivery industries, which has gradually increased the pressure on urban freight transport. However, traditional urban freight transport typically relies on dedicated freight vehicles for delivery, which not only exacerbates traffic congestion but also increases environmental pollution and logistics costs. Therefore, how to optimize the urban logistics system, improve transportation efficiency, and reduce traffic burden has become an urgent problem to be solved.

[0003] In recent years, the rise of the sharing economy has provided a new solution for urban freight transport. Based on the concept of "passenger-freight integration," researchers have proposed utilizing urban public transport networks for freight transport. Urban public transport, such as subways and buses, typically has fixed routes and high operating frequencies, and usually has some idle capacity during off-peak hours. Through reasonable scheduling and optimization, public transport networks can be used to carry part of the city's freight demand, achieving passenger-freight co-transport, thereby improving overall transport efficiency, reducing carbon emissions, and lowering logistics costs.

[0004] The existing technology has the following shortcomings:

[0005] Since the primary goal of public transportation systems is to serve passengers, the introduction of freight transport must not disrupt their normal travel. For example, during peak hours, space in bus or subway cars is extremely limited. If freight transport conflicts with passenger travel, it may reduce passenger comfort and even affect operational efficiency. Furthermore, the stability of freight transport is also a critical issue. Public transportation vehicles often experience numerous starts, stops, and turns during operation. If goods are not properly secured, damage or safety accidents may occur. Therefore, designing reasonable cargo storage and securing schemes in a dynamic public transportation environment, and combining this with intelligent dispatching systems to optimize cargo boarding and alighting times and locations to ensure the coexistence of passengers and goods, is a major challenge in this technical field. Summary of the Invention

[0006] The purpose of this invention is to provide a transportation management system for freight transport using urban public transportation, in order to address the shortcomings of the prior art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a transportation management system for freight transport using urban public transportation, comprising a freight task management module, an intelligent scheduling module, a cargo storage and fixed management module, and an information interaction module;

[0008] The freight task management module is used to receive, store, and optimize freight order information, and dynamically allocate freight transportation routes and times based on public transportation timetables, passenger flow data, and route planning.

[0009] The intelligent dispatch module is used to analyze the real-time operating status, passenger flow density and freight demand of public transportation vehicles, determine cargo loading and unloading stations, and coordinate freight tasks with passenger travel needs.

[0010] The cargo storage and fixed management module is used to plan cargo storage areas based on the spatial layout and dynamic operation characteristics of public transportation vehicles.

[0011] The information interaction module is used to connect the transportation system, public transportation operation management platform and cargo owners, and to provide information on the stability of the cargo storage area during vehicle operation and abnormal alarms.

[0012] Preferably, the freight task management module can access real-time data on public transportation operations, including timetable information, passenger flow data, route planning and traffic status, and dynamically adjust transportation routes based on cargo characteristics, including volume, weight and storage requirements.

[0013] Preferably, the intelligent scheduling module adopts reinforcement learning combined with an adaptive heuristic search algorithm to model the public transportation system as a reinforcement learning environment. By training the model, it optimizes the loading and unloading stations and transportation routes of goods to maximize transportation efficiency and minimize the impact on passengers.

[0014] Preferably, the intelligent scheduling module determines the optimal execution plan for freight tasks based on the real-time operating status of vehicles, passenger flow density, and freight demand, and automatically recalculates the route in case of emergencies, providing alternative transportation solutions.

[0015] Preferably, the carriage center of gravity offset rate is generated after analyzing the spatial layout of the public transportation vehicle. The method for obtaining the carriage center of gravity offset rate is as follows:

[0016] Let the length of the carriage be Lcarriage and the width be Wcarriage. Let the original center of gravity of the carriage be Gorig(xorig, yorig). There are N goods, and the parameters of each goods i include: mass... Position coordinates (xi, yi); Volume constraint Vi; Each particle represents a cargo storage scheme, i.e., the coordinate combination of all cargo: Pk={(x1,y1),(x2,y2),…,(xN,yN)}; Let the particle swarm size be M, let the initial position of each particle be randomly distributed within the available area of ​​the carriage, and let the initial velocity of each particle be randomly set. For each particle Pk, calculate the new center of gravity of the carriage. : ;in: and These are the x and y coordinates of the new center of gravity, and the mass of the cargo. The location of the new center of gravity is affected, and then the car's center of gravity offset rate CGD is calculated, with the following expression: .

[0017] Preferably, a dynamic sway index is generated after analyzing the dynamic operating characteristics of the vehicle during operation. The method for obtaining the dynamic sway index is as follows:

[0018] During vehicle operation, an inertial measurement unit (IMU) is installed to collect triaxial acceleration data: forward / backward ax(t), left / right ay(t), and up / down az(t). The sampling frequency is set to fs. Fourier transform is used to convert the time-domain signal to the frequency domain to calculate the vibration intensity of the signal at different frequencies. For the ax(t), ay(t), and az(t) signals, a fast Fourier transform is applied, with the following expression: Where A(f) is the amplitude at frequency f, and N is the number of data points. These are Fourier basis functions; the vibrational energy spectrum P(f) is calculated using the following expression: ; Reflecting the energy magnitude of each frequency component, extracting the principal vibration frequency components, and calculating The sum of vibrational energy across the frequency band; the dynamic sway index (DVI) is defined as... The weighted sum of vibrational energy within the frequency range is calculated using the following expression: ;in: =0.5Hz, =20Hz, It is the total vibrational energy of the entire spectrum, expressed as: ;in .

[0019] Preferably, the car body center of gravity offset rate and dynamic sway index are converted into a comprehensive feature vector. The comprehensive feature vector is used as the input of the machine learning model. The machine learning model uses the prediction of the stability analysis value label of the cargo storage area for each set of comprehensive feature vectors as the prediction objective and minimizes the sum of prediction errors for all cargo storage area stability analysis value labels as the training objective. The machine learning model is trained until the sum of prediction errors converges and the model training stops. The stability analysis value of the cargo storage area is determined based on the model output. The machine learning model is a multinomial regression model.

[0020] Preferably, in the information interaction module, the obtained stability analysis value of the cargo storage area is compared with a preset stability reference threshold. If the stability analysis value of the cargo storage area is greater than or equal to the preset stability reference threshold, it indicates that the cargo can be stably stored in the storage area during vehicle travel, and no warning signal is generated or the storage area is adjusted. If the stability analysis value of the cargo storage area is less than the preset stability reference threshold, it indicates that the cargo cannot be stably stored in the storage area during vehicle travel, and a warning signal is generated, requiring adjustment of the storage area, including adjusting the cargo storage location, optimizing the fixing method, or optimizing the vehicle operation strategy to improve transportation safety.

[0021] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0022] 1. This invention provides a transportation management system for freight transport using urban public transportation, effectively solving key problems such as the impact of freight transport on passenger travel and insufficient stability of freight transport. Firstly, this system uses an intelligent scheduling module, combined with real-time passenger flow data, public transportation timetables, and route planning, to dynamically allocate freight transport routes, avoiding peak hours, ensuring coexistence of passenger and freight transport, and improving public transportation utilization. Simultaneously, it utilizes reinforcement learning and adaptive heuristic search algorithms to optimize loading and unloading stations. In case of emergencies (such as route adjustments or changes in passenger flow), it can automatically recalculate routes and provide alternative transport solutions to ensure the smooth execution of freight tasks. Furthermore, this system employs a multinomial regression model, combined with the center of gravity deviation rate (CGD) and dynamic sway index (DVI), to accurately assess the stability of the freight storage area. When the stability of the freight storage area falls below a threshold, it automatically triggers an early warning signal, optimizing the freight storage plan or adjusting the transport strategy, thereby improving the safety of freight transport.

[0023] 2. The information interaction module of this invention interconnects the transportation system, public transportation operation management platform, and cargo owners via wireless communication, providing real-time monitoring of cargo status, data sharing, and anomaly alarms to ensure all parties can promptly grasp the cargo transportation status. When cargo is unstable (e.g., loosened, tipped over, or violently shaken), the system automatically triggers an alarm, notifying the driver, dispatch center, or cargo owner, ensuring the safety and efficiency of cargo transportation. Furthermore, this invention, based on deep sensor fusion (IMU+FFT) and combined with intelligent algorithms to calculate the dynamic sway index, accurately assesses the stability of cargo during transit, ensuring safe storage and transportation of cargo in various public transportation environments (buses, subways, trams). Ultimately, this invention maximizes the utilization of idle public transportation capacity, reduces urban freight costs, reduces carbon emissions, improves the utilization rate of public transportation resources, and provides a more intelligent, efficient, and safe transportation solution for urban logistics. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0025] Figure 1 This is a system module diagram of the present invention. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] For examples, please refer to Figure 1 As shown in the figure, the transportation management system for freight transportation using urban public transportation described in this embodiment includes a freight task management module, an intelligent scheduling module, a cargo storage and fixed management module, and an information interaction module.

[0028] The freight task management module is used to receive, store, and optimize freight order information, and dynamically allocate freight transportation routes and times based on public transportation timetables, passenger flow data, and route planning.

[0029] The intelligent dispatch module is used to analyze the real-time operating status, passenger flow density and freight demand of public transportation vehicles, determine cargo loading and unloading stations, and coordinate freight tasks with passenger travel needs.

[0030] The cargo storage and fixed management module is used to plan cargo storage areas based on the spatial layout and dynamic operation characteristics of public transportation vehicles.

[0031] The information interaction module is used to connect the transportation system, public transportation operation management platform and cargo owners, and to provide information on the stability of the cargo storage area during vehicle operation and abnormal alarms.

[0032] The freight task management module uses public transportation timetables, passenger flow data, and route planning to dynamically allocate freight transportation routes and times.

[0033] The freight task management module receives freight orders from cargo owners, logistics companies, or public transportation operators, including: cargo information (type, volume, weight, special storage requirements), pick-up and delivery locations, expected delivery time, and acceptable transportation time range. The order information is stored in the database for subsequent dynamic allocation.

[0034] Access real-time public transportation data, including: timetable information: operating time and departure intervals for each route; passenger flow data: passenger load factor for buses and subways at different times; route planning: available bus / subway routes and transfer options; traffic status: whether there are any emergencies or route adjustments.

[0035] Based on the characteristics of the goods (volume, weight, storage requirements) and combined with passenger flow data, assess whether the goods can be safely transported on the target route and during the specified time period to avoid impacting passenger travel during peak hours. If passenger flow is too high during the target time period, adjust to a lower passenger flow period or arrange for alternative routes. If there is no available capacity on the target route, adjust the transportation route or arrange for delayed delivery.

[0036] The optimal transportation route is calculated based on the following optimization objectives: avoiding peak hours to ensure that freight transportation does not affect passenger comfort; minimizing transportation time, reducing transfers, and improving transportation efficiency; and minimizing additional operating costs by making the most of existing public transportation capacity.

[0037] Route planning examples: Option A (preferred): Bus route A (off-peak) → Subway route B → Drop-off station. Option B (alternative): Bus route C → Transfer route D (low passenger flow at night) → Drop-off station. Determine loading and unloading stations for goods, prioritizing stations with lower passenger flow and available storage space.

[0038] The system sends transportation task instructions to the intelligent scheduling module, notifying relevant stations and drivers to perform cargo loading and unloading operations. It also sends transportation arrangements, including estimated delivery time and real-time freight status, to the shipper through the information interaction module.

[0039] The system monitors public transportation operations in real time. In case of unexpected events such as route adjustments or sudden surges in passenger flow, it automatically recalculates routes and adjusts delivery times. If delays occur due to unforeseen circumstances, it notifies cargo owners and relevant operational personnel through the information exchange module and provides alternative transportation solutions (such as transfers or delays).

[0040] The intelligent scheduling module needs to employ reinforcement learning combined with adaptive heuristic search to dynamically optimize freight loading and unloading stations and coordinate freight tasks with passenger travel demands.

[0041] This system employs deep reinforcement learning combined with heuristic search for decision optimization. Its main process is as follows:

[0042] The intelligent scheduling module models the public transportation system as a reinforcement learning environment, defined as follows:

[0043] Status S includes real-time public transportation status (vehicle location, route timetable, capacity); passenger density (peak / off-peak hours, station congestion); freight demand (order quantity, cargo volume, time constraints); and available station space (carriage area for storing goods, station facilities).

[0044] Action A: System-executable scheduling decisions, including selecting a suitable loading station (which station to load goods at); selecting a suitable unloading station (which station to unload goods at); and adjusting the transportation route (selecting different routes or transfer options).

[0045] Reward function R: With the goal of maximizing transportation efficiency and minimizing the impact on passengers: Positive rewards: successful delivery of goods (+10); loading / unloading at low-passenger-flow stations (+5); choosing the shortest path and reducing transfers (+3); Negative penalties: impacting passenger travel (-10); delays in goods transportation (-8); stations being overloaded and unable to store goods (-5).

[0046] Based on reinforcement learning frameworks (such as Deep Q-Networks, DQN, or Proximal Policy Optimization, PPO), the intelligent scheduling module performs the following operations:

[0047] Collect real-time data on public transportation (vehicle status, passenger density, freight demand).

[0048] Calculate the optimal scheduling scheme based on the current environmental state S.

[0049] A deep Q-network (DQN) is used to calculate the Q-value (transport efficiency score) for each available station, and the station with the highest Q-value is selected for loading / unloading decisions.

[0050] During the exploration phase, Adaptive Heuristic Search (AHS) is introduced to avoid local optima and ensure that the optimal solution can be found in different scenarios.

[0051] Decision execution: The intelligent scheduling module sends scheduling instructions to the transportation system (determining the loading and unloading stations and routes for freight tasks).

[0052] Based on the completion status of freight tasks (such as transportation time and impact on passengers), the reinforcement learning model updates the Q value to optimize future decisions.

[0053] To ensure that freight missions do not disrupt passenger travel, the intelligent scheduling module coordinates transportation arrangements using the following strategies:

[0054] Avoid peak-hour scheduling: Predict peak passenger flow periods (such as commuting rush hours) and adjust freight times accordingly. Prioritize low-passenger-flow stations: Select stations near commercial and office areas for loading, avoiding large-scale loading and unloading at core transfer stations. Intelligent dynamic adjustment: If passenger flow suddenly increases, the intelligent scheduling module can automatically adjust unloading stations to subsequent low-passenger-flow stations for freight unloading.

[0055] The cargo storage and securing management module is used to plan cargo storage areas based on the spatial layout and dynamic operation characteristics of public transportation vehicles, and to provide adaptive cargo securing devices.

[0056] Based on the spatial characteristics of different public transportation modes (buses, subways, trams, etc.), rationally plan the goods storage area:

[0057] Buses (limited space, need to avoid affecting passengers): rear seating area (some foldable seats); standing area near the doors (set up foldable shelves); subways (larger space, can utilize specific carriages): set up dedicated freight carriages (such as some low-passenger-flow lines); corners near both ends of the carriage (utilize areas that do not affect passenger passage); trams (some cities have dedicated freight modes): set up hidden freight space under the carriage; expand storage for small goods under seats or in luggage rack areas.

[0058] After analyzing the spatial layout of public transportation vehicles, the center of gravity offset rate of the carriages is generated. The method for obtaining the center of gravity offset rate of the carriages is as follows:

[0059] Let the length of the car be Lcarriage and the width be Wcarriage. Let the original center of gravity of the car (when there is no cargo) be Gorig(xorig, yorig). There are N cargoes, and the parameters of each cargo i include: mass... Position coordinates (xi, yi); Volume constraint Vi; Each particle represents a cargo storage scheme, i.e., the coordinate combination of all cargo: Pk={(x1,y1),(x2,y2),…,(xN,yN)}; Let the particle swarm size be M (i.e., simultaneously optimize M different cargo storage schemes). Let the initial position of each particle be randomly distributed within the available area of ​​the carriage. Let the initial velocity of each particle be randomly set to ensure sufficient coverage of the search space.

[0060] For each particle Pk (i.e., a cargo storage scheme), calculate the new center of gravity of the carriage. : ;in: and These are the x and y coordinates of the new center of gravity, and the mass of the cargo. The location of the new center of gravity is affected, and then the car's center of gravity offset rate CGD is calculated, with the following expression: .

[0061] The vehicle's center of gravity offset rate measures the change in the vehicle's center of gravity relative to its original center of gravity after cargo has been stored. Excessive offset can affect vehicle stability, causing increased swaying of the cargo compartment and raising the risk of cargo tipping over.

[0062] A higher center of gravity offset (CGD) ratio in the cargo compartment leads to decreased stability in the cargo storage area. An excessively high CGD indicates a significant deviation of the cargo's mass distribution from its original center of gravity, potentially causing imbalance during vehicle movement, especially during acceleration, braking, or turning. This can generate substantial inertial forces, exacerbating cargo slippage, tipping, or swaying. Furthermore, an excessively high CGD may affect vehicle handling stability, increase uneven stress on the internal structure of the cargo compartment, and place additional pressure on cargo securing devices in certain areas, thus reducing their reliability and compromising cargo storage safety. In extreme cases, a severe center of gravity offset can affect vehicle driving safety, causing increased cargo swaying and even impacting overall vehicle stability, increasing the risk of accidents.

[0063] A lower center of gravity offset (CGD) ratio in the cargo compartment improves the stability of the cargo storage area. When the CGD is within a reasonable range, it means the cargo mass is evenly distributed within the cargo compartment, allowing the vehicle to maintain balance and stability during operation and preventing violent swaying or tilting caused by excessive center of gravity shift. A lower CGD ensures a more even distribution of inertial forces on the cargo during acceleration, braking, and turning, reducing the likelihood of cargo displacement, tipping, or loosening. Furthermore, a reasonable center of gravity distribution helps optimize cargo securing strategies, reducing additional stress on securing devices, improving the safety of the cargo storage area, and making freight transport more stable and efficient. Therefore, during cargo storage, the CGD should be optimized to remain within a safe range to ensure freight stability and vehicle smoothness.

[0064] After analyzing the dynamic operating characteristics of the vehicle during operation, a dynamic sway index is generated. The method for obtaining the dynamic sway index is as follows:

[0065] During vehicle operation, an inertial measurement unit is installed to collect three-axis acceleration data: forward and backward (X-axis): ax(t) (impact of acceleration and braking), left and right (Y-axis): ay(t) (impact of lateral swaying during turning), and up and down (Z-axis): az(t) (impact of road bumps and track vibration).

[0066] The sampling frequency fs must satisfy the Nyquist sampling theorem (it must be at least twice the highest vibration frequency).

[0067] Typical values ​​are fs = 100~500fHz (the common vibration frequency range of buses, subways, and rail vehicles is 0.5Hz-20Hz).

[0068] The Fourier transform is used to convert a time-domain signal to the frequency domain, calculating the vibration intensity of the signal at different frequencies. For signals ax(t), ay(t), and az(t), the Fast Fourier Transform is applied, and the expression is: Where A(f) is the amplitude at frequency f, and N is the number of data points. These are Fourier basis functions. The vibrational energy spectrum P(f) is calculated using the following expression: This reflects the energy magnitude of each frequency component. Extract the principal vibration frequency components and calculate... The total vibration energy across the frequency band (i.e., the main frequency range of vehicle swaying). Set a frequency threshold to ignore weak vibrations (such as vibration components below a certain energy value).

[0069] Define the dynamic sway index (DVI) as The weighted sum of vibrational energy within the frequency range is calculated using the following expression: ;in: =0.5Hz, =20Hz (the main frequency band of vehicle shaking). It is the total vibrational energy of the entire spectrum, expressed as: ;in (Nyquist frequency). The higher the DVI, the stronger the dynamic shaking of the vehicle and the less stable the cargo storage area. A DVI close to 0 indicates that the vibration energy is concentrated in the low-frequency range, and the cargo is more stable.

[0070] The dynamic sway index measures the impact of dynamic factors such as acceleration, braking, and turning on the stability of stored goods during actual vehicle operation. It comprehensively considers factors such as vehicle speed changes, track / road vibration, and vehicle acceleration to predict whether there is a significant risk of displacement or tipping of goods during transportation.

[0071] The car's center of gravity offset rate and dynamic sway index are converted into a comprehensive feature vector. This comprehensive feature vector is then used as input to a machine learning model. The machine learning model uses the prediction of the cargo storage area stability analysis value label for each set of comprehensive feature vectors as its prediction objective. The training objective is to minimize the sum of prediction errors for all cargo storage area stability analysis value labels. The machine learning model is trained until the sum of prediction errors converges, at which point the model training stops. The cargo storage area stability analysis value is determined based on the model output. The machine learning model is a multinomial regression model.

[0072] The method for obtaining the stability analysis values ​​of the cargo storage area is as follows: Obtain the corresponding function expression from the training data of the comprehensive feature vector of the trained machine learning model. In the formula, It is the output function of the model. The value represents the car's center of gravity offset rate, and DVI represents the dynamic sway index. This represents the stability analysis value for the cargo storage area.

[0073] The information interaction module connects the transportation system, public transportation operation management platform, and cargo owners to achieve real-time monitoring, data sharing, and anomaly alarms for freight transport. This module collects key information such as vehicle driving data, cargo storage stability (CGD, DVI), and the status of securing devices through sensors, and pushes real-time updates to relevant parties via wireless communication (API / WebSocket). When anomalies in cargo storage are detected (such as loosening, tipping, or high vibration), the system automatically triggers an alarm mechanism, notifying the operator, driver, or cargo owner to take appropriate measures to ensure the safe transport of goods on public transportation and optimize the stability and operational efficiency of the passenger-freight co-transport mode.

[0074] The obtained stability analysis value of the cargo storage area is compared with a pre-set stability reference threshold. If the stability analysis value is greater than or equal to the pre-set stability reference threshold, it means that the cargo can be stably stored in the storage area during vehicle travel, and no warning signal is generated or the storage area is adjusted. If the stability analysis value is less than the pre-set stability reference threshold, it means that the cargo cannot be stably stored in the storage area during vehicle travel, and a warning signal is generated, requiring adjustment of the storage area, including adjusting the cargo storage location, optimizing the fixing method, or optimizing the vehicle operation strategy to improve transportation safety.

[0075] It should be noted that the stability reference threshold is a key indicator used to assess the safety of goods stored on public transportation. It is a minimum stability standard set based on historical data, experimental analysis, and machine learning models to determine the stability of goods during transit.

[0076] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0077] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0078] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A transport management system for freight transport using urban public transport, characterized in that: The system comprises a freight task management module, an intelligent scheduling module, a freight storage and fixation management module, and an information interaction module. The freight task management module is configured to receive, store and optimize freight order information, and dynamically allocate the transportation route and time of the freight based on the public transportation operation timetable, passenger flow data and line planning. The intelligent scheduling module is configured to analyze the real-time running state of the public transportation tool, passenger flow density and freight demand, determine the freight loading site and unloading site, and coordinate the freight task and passenger travel demand. The cargo storage and fixing management module is used for planning a cargo storage area based on the spatial layout and dynamic running characteristics of the public transport vehicle; wherein, a dynamic shaking index is generated after analyzing the dynamic running characteristics in the vehicle driving process, and the method for obtaining the dynamic shaking index is as follows: in the vehicle driving process, an inertial measurement unit is installed to collect three-axis acceleration data: front-rear direction ax(t), left-right direction ay(t), and up-down direction az(t); the sampling frequency is set as fs, the Fourier transform is used to convert the time domain signal to the frequency domain, the vibration intensity of the signal at different frequencies is calculated, for the ax(t), ay(t), and az(t) signals, the fast Fourier transform is applied, and the expression is as follows: ; wherein, A(f) is the amplitude at the frequency f, N is the number of data points, is the Fourier base function; the vibration energy spectrum P(f) is calculated, and the expression is as follows: ; the energy size of each frequency component is reflected, the main vibration frequency component is extracted, the vibration energy sum of the frequency band is calculated; the dynamic shaking index DVI is defined as the vibration energy weighted sum in the frequency range, and the calculation expression is as follows: ; wherein: , , is the total vibration energy of the entire frequency spectrum, and the expression is as follows: ; wherein ; The information interaction module is configured to connect the transportation system, the public transportation operation management platform and the freight owner, and provide the stability condition and abnormal alarm of the freight storage area during the running of the vehicle.

2. The transport management system for freight transport using urban public transport according to claim 1, characterized in that: The freight task management module can access real-time data of public transportation operation, including timetable information, passenger flow data, line planning and traffic state, and dynamically adjust the transportation path based on the characteristics of the freight, including volume, weight and storage demand.

3. The transport management system for freight transport using urban public transport according to claim 1, characterized in that: The intelligent scheduling module models the public transportation system as a reinforcement learning environment using reinforcement learning combined with an adaptive heuristic search algorithm, and optimizes the loading site, unloading site and transportation path of the freight through model training to maximize transportation efficiency and minimize the impact on passengers.

4. The transport management system for freight transport using urban public transport according to claim 3, characterized in that: The intelligent scheduling module determines the optimal execution scheme of the freight task based on the real-time running state of the vehicle, passenger flow density and freight demand, and automatically recalculates the path in case of emergency to provide alternative transportation schemes.

5. The transport management system for freight transport using urban public transport according to claim 4, characterized in that: After analyzing the spatial layout of the public transportation tool, the carriage center of gravity offset rate is generated. The method for obtaining the carriage center of gravity offset rate is as follows: Let the carriage length be Lcarriage and the width be Wcarriage. Let the original center of gravity of the carriage be Gorig(xorig, yorig). There are N goods, and the parameters of each good i include: mass ; position coordinates (xi, yi); volume limit Vi; each particle represents a goods storage scheme, that is, the coordinate combination of all goods: Pk = {(x1, y1), (x2, y2), …, (xN, yN)}; let the particle swarm size be M, let the initial position of each particle be randomly distributed in the available area of the carriage, let the initial speed of each particle be randomly set, and for each particle Pk, calculate the new center of gravity of the carriage ; wherein: and are the horizontal and vertical coordinates of the new center of gravity, and the mass of the goods affects the position of the new center of gravity, and then the carriage center of gravity deviation rate CGD is calculated, and the expression is: .

6. The transport management system for freight transport using urban public transport according to claim 5, characterized in that: The carriage center of gravity offset rate and the dynamic shaking index are converted into a comprehensive feature vector, which is used as the input of a machine learning model. The machine learning model predicts the freight storage area stability analysis value label as the prediction target, and minimizes the sum of prediction errors of all freight storage area stability analysis value labels as the training target. The model is trained until the sum of prediction errors converges, and the freight storage area stability analysis value is determined based on the model output result. The machine learning model is a polynomial regression model.

7. The transport management system for freight transport using urban public transport according to claim 6, characterized in that: In the information interaction module, the obtained freight storage area stability analysis value is compared with the pre-set stability reference threshold value. If the freight storage area stability analysis value is greater than or equal to the pre-set stability reference threshold value, it means that the freight can be stably stored in the storage area during the running of the vehicle, and no warning signal is generated and no storage area adjustment is needed. If the freight storage area stability analysis value is less than the pre-set stability reference threshold value, it means that the freight cannot be stably stored in the storage area during the running of the vehicle, and a warning signal is generated and storage area adjustment is needed, including adjusting the freight storage position, optimizing the fixation method or optimizing the vehicle operation strategy to improve transportation safety.

Images