A method and system for evaluating a commuting network based on artificial intelligence

Abstract

The application provides a kind of commuting network evaluation method and system based on artificial intelligence, through S1 generate the smooth manifold of commuting network on Lie group space, according to S2 construct dynamic traffic manifold, through S3 construct multi-agent forest model, generate commuting score vector through agent forest model, then through S4 generate score relationship equation to integrate commuting score vector, and the three commuting paths of highest score are marked and pushed to the user, considering various commuting factors, so that the final commuting score is more accurate, while adding the historical data of the user, more in line with the needs of the user, through S5 obtain the final commuting path of the user and carry out tracing, while generating traceability report, according to S6 optimize dynamic traffic manifold, and return to S2, find the factors affecting the score by the method of tracing and analyze, while optimizing the dynamic traffic manifold according to the analysis result, to achieve the purpose of optimization evaluation.

Description

An AI-based method and system for evaluating commuter networks Technical Field

[0001] This invention relates to the field of traffic commuting control technology, and more specifically, to an artificial intelligence-based commuting network evaluation method and system. Background Technology

[0002] With the acceleration of urbanization and the popularization of intelligent transportation systems, efficient commuting route planning has become a key link in improving urban operational efficiency. Traditional route evaluation methods mainly rely on static road network topology and fixed weight scoring models, such as shortest path search based on the xx algorithm or time-optimal planning based on the xx algorithm. Their evaluation dimensions are singular and difficult to adapt to dynamic traffic environments. In order to cope with complex road conditions, existing technologies are gradually introducing artificial intelligence learning models to achieve responses to some dynamic factors.

[0003] Most existing technologies focus on evaluating a single indicator, lacking a systematic approach that combines multi-source data for analysis and evaluation. They also fail to incorporate user opinions and feedback. Therefore, there is an urgent need for a commuting assessment method that integrates continuous spatiotemporal modeling and user feedback to improve the accuracy of the assessment. Summary of the Invention

[0004] In view of the aforementioned problems, and in conjunction with the first aspect of the present invention, embodiments of the present invention provide an artificial intelligence-based commuter network assessment method, the method comprising:

[0005] S1: Obtain the commuting network and map it to the Lie group space to generate a smooth manifold of the commuting network in the Lie group space;

[0006] S2: Obtain real-time traffic flow data, traffic light phase timing, and road topology metadata; construct a dynamic traffic manifold based on the real-time traffic flow data, traffic light phase timing, road topology metadata, and the smooth manifold.

[0007] S3: Construct a multi-agent forest model to obtain the user's input commuting start point and commuting destination. The agent forest model generates a commuting score vector based on the commuting start point, commuting destination, and dynamic traffic manifold. The commuting score vector includes a travel time score, an energy consumption score, and an abnormal event impact probability score.

[0008] S4: Generate a rating relationship equation based on the user's historical preferences, integrate the commuting rating vector based on the rating relationship equation, obtain the commuting route rating set, and mark the three commuting routes with the highest rating in the route rating set and push them to the user;

[0009] S5: Obtain the user's final commuting route, trace the final commuting route, and generate a tracing report;

[0010] S6: Optimize the dynamic traffic flow pattern based on user feedback and source tracing reports, and return to step S2.

[0011] As a further aspect of the present invention, the step of obtaining the commuting network and mapping the commuting network to a Lie group space to generate a smooth manifold of the commuting network on the Lie group space includes:

[0012] The commuting network includes commuting nodes and commuting routes. The commuting nodes and commuting routes are mapped to a Lie group space. The commuting nodes correspond to elements in the Lie group space, and the commuting routes correspond to group action operators in the Lie group space. A smooth manifold is generated based on the commuting nodes and commuting routes.

[0013] As a further aspect of the present invention, the step of acquiring real-time traffic flow data, traffic light phase timing, and road topology metadata, and constructing a dynamic traffic manifold based on the real-time traffic flow data, traffic light phase timing, road topology metadata, and the smooth manifold includes:

[0014] Real-time traffic flow data is acquired at a sampling frequency of once per second, and the real-time traffic flow data includes vehicle position and motion status.

[0015] The traffic light phase timing sequence is obtained with a time resolution of 0.1 seconds. The traffic light phase timing sequence includes the change time and state of the traffic signal.

[0016] Obtain road topology metadata, which includes the number of lanes, gradient, and radius of curvature of the road;

[0017] A commuting differential equation is constructed based on real-time traffic flow data, traffic light phase timing, and road topology metadata. A dynamic traffic manifold is then constructed based on the commuting differential equation and the smooth manifold.

[0018] As a further aspect of the present invention, the step of constructing a commuting differential equation based on real-time traffic flow data, traffic light phase sequence, and road topology metadata, and constructing a dynamic traffic manifold based on the commuting differential equation and the smooth manifold, includes:

[0019] The commuting location of a vehicle on a smooth manifold is obtained. A traffic flow tensor is obtained based on real-time traffic flow data and road topology metadata. The traffic flow tensor represents the flow state of the traffic flow. The location and the traffic flow tensor are combined to form a commuting vector.

[0020] Traffic signal changes are obtained based on the phase and timing of traffic lights;

[0021] Differentiate the commuting vector over time to obtain the commuting differential equation, and construct a dynamic traffic manifold with the smooth manifold. The time of change is represented by the time before and after the traffic signal change.

[0022] As a further aspect of the present invention, the construction of a multi-agent forest model obtains the user-input commuting start and commuting destination. The agent forest model generates a commuting score vector based on the commuting start, commuting destination, and dynamic traffic manifold. The commuting score vector includes a travel time score, an energy consumption score, and an abnormal event impact probability score, including:

[0023] Users input their commute origin and destination into the mobile client. The mobile client analyzes the commute origin, destination, and dynamic traffic flow based on a multi-agent forest model to generate a commute score vector.

[0024] The multi-agent forest model includes a time agent, an energy-consuming agent, and a risk agent.

[0025] As a further aspect of the present invention, the method further includes:

[0026] The time agent predicts the travel time based on LSTM, obtains the predicted travel time, and scores the predicted travel time to obtain a travel time score.

[0027] The energy-consuming agent solves the energy problem based on the dynamic equation to obtain the energy consumed during commuting, and scores the consumed energy to obtain an energy consumption score.

[0028] The risk agent analyzes the impact of abnormal events based on Bayesian networks, obtains the probability of the impact of abnormal events, and scores the probability of the impact of abnormal events to obtain an impact probability score.

[0029] As a further aspect of the present invention, the step of generating a rating relationship equation based on user historical preferences, integrating commuting rating vectors based on the rating relationship equation to obtain a commuting route rating set, and marking and pushing the three highest-rated commuting routes in the commuting route rating set to the user includes:

[0030] Based on historical data, user historical preferences are obtained, and the trip time score, energy consumption score, and abnormal event impact probability score are assigned and weighted according to the user historical preferences, and a score relationship equation is generated.

[0031] The travel time score, energy consumption score, and probability score of abnormal events are input into the scoring relationship equation to generate a commuting route score set. The three commuting routes with the highest scores in the commuting route score set are extracted and marked, and the three commuting routes with the highest scores are pushed to the user.

[0032] As a further aspect of the present invention, the step of extracting and marking the three highest-rated commuter routes from the commuter route rating set, and then pushing the three highest-rated commuter routes to the user, includes:

[0033] The markings include time markings, energy consumption markings, and comprehensive markings;

[0034] If the extracted commuting route has the highest travel time score, the extracted commuting route is marked as having the shortest travel time.

[0035] If the extracted commuting route has the highest energy consumption score, the extracted commuting route is marked as having the lowest energy consumption.

[0036] If the extracted commuting route has the highest score, the extracted commuting route is marked as having the highest overall score;

[0037] The three commuting routes and their corresponding ratings will be pushed to the user.

[0038] As a further aspect of the present invention, the step of obtaining the user's final commuting route, tracing the final commuting route, and generating a tracing report includes:

[0039] Based on the user's selection, the final commuting route and the corresponding rating are obtained. The final commuting route is represented by the commuting route finally selected by the user. The commuting rating vector corresponding to the rating is obtained, and the partial derivative of the corresponding commuting rating vector is taken to obtain the sensitivity of the commuting rating vector. The commuting influence matrix is ​​obtained.

[0040] Based on the commuting impact matrix, key features are filtered to obtain a commuting impact list;

[0041] A dynamic decision tree is constructed based on the commuting impact list, and a source tracing report is generated based on the dynamic decision tree.

[0042] Furthermore, embodiments of the present invention also provide an artificial intelligence-based commuter network assessment system, comprising:

[0043] The acquisition module is used to acquire commuter network data, real-time traffic flow data, traffic light phase timing data, and road topology metadata, and to acquire the user's final commuting route.

[0044] The generation module is used to map the commuting network to a Lie group space, generate a smooth manifold of the commuting network on the Lie group space, and generate a commuting score vector based on the start point, end point and dynamic traffic manifold of the agent forest model, and generate a score relationship equation based on the user's historical preferences.

[0045] The construction module constructs a dynamic traffic manifold based on real-time traffic flow data, traffic light phase sequence, road topology metadata, and the smooth manifold, and constructs a multi-agent forest model.

[0046] The integration module integrates commuting score vectors based on the scoring relationship equation, obtains a commuting route score set, and marks the three commuting routes with the highest scores in the route score set and pushes them to the user;

[0047] The tracing module is used to trace the final commuting route and generate a tracing report.

[0048] An optimization module optimizes dynamic traffic flow patterns based on user feedback and source tracing reports.

[0049] Based on the above, this application embodiment achieves the following: Step S1: Obtaining the commuting network and mapping it to a Lie group space to generate a smooth manifold of the commuting network on the Lie group space; Step S2: Obtaining real-time traffic flow data, traffic light phase timing, and road topology metadata; Constructing a dynamic traffic manifold using the real-time traffic flow data, traffic light phase timing, road topology metadata, and the smooth manifold; Step S3: Constructing a multi-agent forest model to obtain the user-inputted commuting start and end points; The agent forest model generates a commuting score vector based on the commuting start, end, and dynamic traffic manifold, whereby the commuting score vector includes a travel time score, an energy consumption score, and an abnormal event impact probability score; Step S4: Generating a score relationship equation; Integrating the commuting score vector based on the score relationship equation to obtain a commuting path rating. The system divides the commuting network into sets and marks the three highest-scoring commuting routes in the set, then pushes them to the user. By mapping the commuting network to a Lie group space, it generates commuting differential equations based on travel time and energy consumption, and constructs a dynamic traffic manifold. A multi-agent forest model is used to score the routes generated by the dynamic traffic manifold, taking into account various commuting factors to make the final commuting score more accurate. At the same time, it incorporates the user's historical data to better meet the user's needs. Step S5 obtains the user's final commuting route and traces the source of the final commuting route, generating a source tracing report. Step S6 optimizes the dynamic traffic manifold and returns to step S2. The source tracing method is used to find the factors affecting the score, and the influencing factors are analyzed to obtain the analysis results. Based on the analysis results, the dynamic traffic manifold is optimized to achieve the purpose of optimization evaluation. Attached Figure Description

[0050] Figure 1 is a schematic diagram of the execution flow of an artificial intelligence-based commuter network assessment method provided in an embodiment of the present invention.

[0051] Figure 2 is a schematic diagram of an artificial intelligence-based commuter network evaluation system provided in an embodiment of the present invention. Detailed Implementation

[0052] The present invention will now be described in detail with reference to the accompanying drawings. Figure 1 is a schematic diagram of the execution flow of an artificial intelligence-based commuter network assessment method according to an embodiment of the present invention. The following is a detailed description of the artificial intelligence-based commuter network assessment method.

[0053] Step S1: Obtain the commuting network and map it to the Lie group space to generate a smooth manifold of the commuting network in the Lie group space.

[0054] The commuting network includes commuting nodes and commuting routes. The commuting nodes and commuting routes are mapped to a Lie group space. The commuting nodes correspond to elements in the Lie group space, and the commuting routes correspond to group action operators in the Lie group space. A smooth manifold is generated based on the commuting nodes and commuting routes.

[0055] It should be noted that a Lie group space is a class of mathematical objects with a special structure that combines the characteristics of a "group" and a "manifold". Specifically, a Lie group is a set that satisfies the operational rules of a group and has a smooth manifold structure, so complex commuting networks can be subjected to calculus operations on Lie groups.

[0056] Specifically, the commuting nodes and routes in the commuting network are mapped to the Lie group space. Intersections are regarded as commuting nodes in the commuting network, and roads are regarded as commuting routes in the commuting network. Each commuting node corresponds to an element in the Lie group, representing a position or transformation in the space. Each commuting route corresponds to a group action operator in the Lie group, describing the transformation relationship between nodes. Through this mapping relationship, the entire commuting network forms a smooth manifold structure in the Lie group space, which provides a continuous and differential mathematical basis for analyzing and studying commuting behavior and its changes.

[0057] Step S2: Obtain real-time traffic flow data, traffic light phase sequence, and road topology metadata; construct a dynamic traffic manifold based on the real-time traffic flow data, traffic light phase sequence, road topology metadata, and the smooth manifold.

[0058] In this embodiment, step S2 includes:

[0059] Step S21: Acquire real-time traffic flow data at a sampling frequency of once per second. The real-time traffic flow data includes vehicle location and motion status.

[0060] Specifically, real-time traffic flow data is acquired through acquisition module 101, which samples once per second. The real-time traffic flow data includes vehicle location and motion status. For example, vehicle xx is traveling at a speed of 60 kilometers per hour on xx Avenue, and its latitude and longitude are expressed as: longitude 116.4074, latitude 39.9042 degrees north.

[0061] The traffic light phase timing sequence is obtained with a time resolution of 0.1 seconds. The traffic light phase timing sequence includes the change time and state of the traffic signal.

[0062] Specifically, the traffic light phase sequence is obtained through the acquisition module 101, with a time resolution of 0.1 seconds. The traffic light phase sequence includes the change time and state of the traffic signal. For example, the red light on xx Avenue is on for 20 seconds, the green light is on for 15 seconds, and the yellow light is on for 3 seconds.

[0063] Obtain road topology metadata, which includes the number of lanes, gradient, and radius of curvature of the road.

[0064] Specifically, road topology metadata is obtained through module 101. The road topology metadata includes the number of lanes, slope, and radius of curvature of the road. For example, xx Avenue has five lanes, a slope of 15 degrees, and a radius of curvature of 150 meters.

[0065] A commuting differential equation is constructed based on real-time traffic flow data, traffic light phase timing, and road topology metadata. A dynamic traffic manifold is then constructed based on the commuting differential equation and the smooth manifold.

[0066] Step S22: Obtain the commuting location of the vehicle on the smooth manifold, obtain the traffic flow tensor based on real-time traffic flow data and road topology metadata, the traffic flow tensor represents the flow state of the traffic flow, and combine the location and the traffic flow tensor to form a commuting vector.

[0067] Traffic signal changes are obtained based on the phase and timing of traffic lights.

[0068] Differentiate the commuting vector over time to obtain the commuting differential equation, and construct a dynamic traffic manifold with the smooth manifold. The time of change is represented by the time before and after the traffic signal change.

[0069] In this embodiment, the commuting position of the vehicle on the smooth manifold is obtained by the acquisition module 101. At the same time, the real-time traffic flow data and road topology metadata are analyzed to obtain the flow state of the real-time traffic flow. The position and the flow state of the real-time traffic flow are combined to form a commuting vector. Meanwhile, traffic signal changes are obtained by the traffic light phase timing.

[0070] Furthermore, the aforementioned locations, real-time traffic flow status, and traffic signal changes are input into the smooth manifold, and a dynamic traffic manifold is constructed through the construction module 103.

[0071] Specifically, the commuting vector is differentiated over time to obtain the commuting differential equation, which is expressed as:

[0072] ; ;

[0073] in, Represented as a commuting vector. This is represented by the coordinates of the vehicle's position on the dynamic traffic flow. This represents the real-time traffic flow status. This is represented as an external stimulus, which is a change in traffic signals.

[0074] Step S3: Construct a multi-agent forest model, obtain the user's input commuting start point and commuting destination, and generate a commuting score vector based on the commuting start point, commuting destination and dynamic traffic manifold. The commuting score vector includes a travel time score, an energy consumption score and an abnormal event impact probability score.

[0075] In this embodiment, step S3 includes:

[0076] In step S31, the user inputs the commuting origin and commuting destination into the mobile user terminal. The mobile user terminal analyzes the commuting origin, commuting destination and dynamic traffic flow based on the multi-agent forest model to generate a commuting score vector.

[0077] Specifically, the user inputs the commuting origin and commuting destination into the mobile client. The multi-agent forest model in the mobile client analyzes the user's commuting origin and commuting destination based on the dynamic traffic manifold, and generates a commuting score vector according to the generation module 102.

[0078] Step S32: The multi-agent forest model includes a time agent, an energy consumption agent, and a risk agent.

[0079] It should be noted that the above multi-agent forest model includes a time agent, an energy consumption agent, and a risk agent. When commuting, users tend to focus more on commuting time, commuting energy consumption, and commuting risks. Commuting scores, which are composed of commuting time, commuting energy consumption, and commuting risks, are more focused on user experience.

[0080] The time agent predicts the travel time based on LSTM, obtains the predicted travel time, and scores the predicted travel time to obtain a travel time score.

[0081] It should be noted that the LSTM model, or Long Short-Term Memory network model, is a type of recurrent neural network in deep learning used to process and predict time series data. By learning from historical data, LSTM can capture the temporal dependencies and periodicity in the data, thereby making predictions about future points in time.

[0082] In this embodiment, LSTM can predict traffic conditions for the next few seconds to minutes based on historical traffic flow and vehicle status, and predict the travel time for each vehicle option. At the same time, it scores the travel time of each option to obtain a travel time score.

[0083] For example, the travel time from point A to point B is 15 minutes for Option 1, 14 minutes for Option 2, and 16 minutes for Option 3. Option 1, Option 2, and Option 3 are scored, with Option 1 receiving a score of 88, Option 2 receiving a score of 90, and Option 3 receiving a score of 85.

[0084] The energy-consuming agent solves the energy problem based on the dynamic equation to obtain the energy consumed during commuting, and scores the consumed energy to obtain an energy consumption score.

[0085] In this embodiment, the acquisition module 101 acquires the vehicle's real-time speed, travel distance, and traffic signal changes, solves the data, obtains the energy consumption for commuting, and scores the energy consumption to obtain an energy consumption score.

[0086] For example, the vehicle in Option 1 traveling from point A to point B consumes less than 1% of the electricity, the vehicle in Option 2 traveling from point A to point B consumes 1% of the electricity, and the vehicle in Option 3 traveling from point A to point B consumes 2% of the electricity. Option 1, Option 2, and Option 3 are scored, with Option 1 receiving a score of 95, Option 2 receiving a score of 90, and Option 3 receiving a score of 85.

[0087] The risk agent analyzes the impact of abnormal events based on Bayesian networks, obtains the probability of the impact of abnormal events, and scores the probability of the impact of abnormal events to obtain an impact probability score.

[0088] It should be noted that a Bayesian network is a probabilistic graphical model used to describe the conditional dependencies between variables. When analyzing the impact of anomalous events, Bayesian networks can help us understand the extent to which anomalous events affect other variables in the system. By inferring the conditional probabilities of relevant nodes in the network, we can assess the changes and magnitudes of various variables in the system after an anomalous event occurs, thereby identifying the causes and potential impacts of the anomaly.

[0089] In this embodiment, the probability of an abnormal event is obtained by the acquisition module 101, and the probability of an abnormal event is scored to obtain the probability score of an abnormal event.

[0090] For example, the probability of an abnormal event affecting Plan 1 from location A to location B is 10%, the probability of an abnormal event affecting Plan 2 from location A to location B is 20%, and the probability of an abnormal event affecting Plan 3 from location A to location B is 50%. Plans 1, 2, and 3 are scored, with Plan 1 scoring 90 points, Plan 2 scoring 85 points, and Plan 3 scoring 50 points.

[0091] Step S4: Generate a rating relationship equation based on the user's historical preferences, integrate the commuting rating vector based on the rating relationship equation, obtain the commuting route rating set, and mark the three commuting routes with the highest ratings in the route rating set and push them to the user.

[0092] In this embodiment, step S4 includes:

[0093] Step S41: Obtain user historical preferences based on historical data, assign scores to trip time, energy consumption, and probability of abnormal events based on user historical preferences, and generate a score relationship equation.

[0094] It should be noted that, based on the historical data acquired by the acquisition module 101, the historical preferences of each user are analyzed through the historical data, and the travel time score, energy consumption score, and abnormal event impact probability score are assigned and weighted according to the user's historical preferences. The scoring relationship equation is generated by the generation module 102.

[0095] For example, if a user's historical data shows that the user's commute tends to be shorter, then the rating ratio for commute time, energy consumption, and probability of abnormal events is 6:3:1. If the user's commute tends to be lower in energy consumption, then the rating ratio for commute time, energy consumption, and probability of abnormal events is 3:6:1.

[0096] Specifically, the scoring relationship equation is expressed as:

[0097] ;

[0098] in, This represents the score distribution ratio for the trip time rating. This is represented by the rating distribution ratio for energy consumption scores. This represents the weighting of the probability score for the impact of abnormal events. This is represented by a score for the trip's duration. This is represented by an energy consumption score. This is represented as a probability score for the impact of abnormal events. This is represented as a commuting score.

[0099] The travel time score, energy consumption score, and probability score of abnormal events are input into the scoring relationship equation to generate a commuting route score set. The three commuting routes with the highest scores in the commuting route score set are extracted and marked, and the three commuting routes with the highest scores are pushed to the user.

[0100] For example, if the obtained travel time score, energy consumption score, and abnormal event impact probability score are 90, 80, and 80 respectively, and historical data shows that the user's commute tends to be shorter, then the score distribution ratio of travel time, energy consumption, and abnormal event impact probability score is 6:3:1, resulting in a calculated commuting score of 86. If the obtained travel time score, energy consumption score, and abnormal event impact probability score are 90, 90, and 90 respectively, and historical data shows that the user's commute tends to be lower in energy consumption, then the score distribution ratio of travel time, energy consumption, and abnormal event impact probability score is 3:6:1, resulting in a calculated commuting score of 90. Similarly, scores for multiple commuting routes are obtained, a commuting route score set is generated, and the three routes with the highest commuting scores are selected from the commuting route score set, marked, and pushed to the user.

[0101] It should be noted that if the probability score of abnormal event impact is below 60 points, it indicates that the probability of an abnormal event impact is extremely high. In this case, the score distribution ratio of travel time, energy consumption, and probability score of abnormal event impact is 2:2:6.

[0102] Step S42, the markers include time markers, energy consumption markers, and comprehensive markers.

[0103] If the extracted commuting route has the highest travel time score, the extracted commuting route is marked as having the shortest travel time.

[0104] If the extracted commuting route has the highest energy consumption score, the extracted commuting route is marked as having the lowest energy consumption.

[0105] If the extracted commuting route has the highest score, it is marked as having the highest overall score.

[0106] The three commuting routes and their corresponding ratings will be pushed to the user.

[0107] For example, if the three obtained routes are route 1, route 2, and route 3, and their commuting scores are 88, 90, and 92 respectively, then route 1 has the shortest commuting time, so it is marked as the shortest time. Route 2 has the lowest commuting energy consumption, so it is marked as the lowest energy consumption. Route 3 has the highest overall score, so it is marked as the highest score.

[0108] Step S5: Obtain the user's final commuting route, trace the final commuting route, and generate a traceability report.

[0109] Based on the user's selection, the final commuting route and corresponding rating are obtained. The final commuting route is represented by the commuting route ultimately selected by the user. The commuting rating vector corresponding to the rating is obtained, and the partial derivative of the corresponding commuting rating vector is taken to obtain the sensitivity of the commuting rating vector, and the commuting influence matrix is ​​obtained.

[0110] For example, if a user selects to obtain the final commuting route and the corresponding score: Path 1, 88 points, the commuting score vector for Path 1 is obtained as (90, 85, 80). The partial derivative of the commuting score vector is then performed to obtain the sensitivity of the commuting score vector, and the commuting influence matrix is ​​obtained [time contribution 72%, energy consumption contribution 20%, abnormal event contribution 10%].

[0111] Based on the commuting impact matrix, key features are filtered to obtain a list of commuting impacts.

[0112] For example, by filtering key features of the above commuting impact matrix, the time contribution is selected as the key feature. The time contribution includes the time spent waiting at red lights and the time spent in traffic jams at intersections. The time spent waiting at red lights and the time spent in traffic jams at intersections are obtained, and a commuting impact list is generated [Red light waiting time at intersection A is extended by 18%, and traffic jam time at intersection xx is 9%].

[0113] A dynamic decision tree is constructed based on the commuting impact list, and a source tracing report is generated based on the dynamic decision tree.

[0114] For example, a dynamic decision tree is constructed using the above list of traffic impacts, and a source tracing report is generated based on the dynamic decision tree [xx intersection green light +24, xxx road section smooth +12].

[0115] Step S6: Optimize the dynamic traffic flow pattern based on user feedback and source tracing report, and return to step S2.

[0116] Specifically, user feedback and source tracing reports are analyzed to optimize the dynamic traffic flow pattern, and the optimized dynamic traffic flow pattern is returned to step S2.

[0117] Furthermore, by tracing the source, we can find the factors that affect the score, analyze the influencing factors, obtain the analysis results, and optimize the dynamic traffic flow based on the analysis results to achieve the purpose of optimization evaluation.

[0118] Figure 2 shows a schematic diagram of an artificial intelligence-based commuter network evaluation system that can realize the ideas of this application, provided by some embodiments of this application.

[0119] Specifically, an artificial intelligence-based commuter network assessment system includes:

[0120] The acquisition module 101 is used to acquire the commuting network, acquire real-time traffic flow data, traffic light phase timing and road topology metadata, and acquire the user's final commuting route.

[0121] The generation module 102 is used to map the commuting network to the Lie group space, generate a smooth manifold of the commuting network on the Lie group space, and generate a commuting score vector based on the starting point, the ending point and the dynamic traffic manifold of the agent forest model, and generate a score relationship equation based on the user's historical preferences.

[0122] Construction module 103 constructs a dynamic traffic manifold based on real-time traffic flow data, traffic light phase sequence, road topology metadata, and the smooth manifold, and constructs a multi-agent forest model.

[0123] The integration module 104 integrates the commuting score vectors based on the scoring relationship equation, obtains the commuting route score set, and marks the three commuting routes with the highest scores in the route score set and pushes them to the user.

[0124] The tracing module 105 is used to trace the final commuting route and generate a tracing report.

[0125] The optimization module 106 optimizes the dynamic traffic flow pattern based on user feedback and source tracing reports.

[0126] The specific usage and function of this embodiment are explained below:

[0127] First, the commuter network is obtained in step S1 and mapped to a Lie group space, generating a smooth manifold of the commuter network on the Lie group space. Next, real-time traffic flow data, traffic light phase sequence data, and road topology metadata are obtained in step S2. A dynamic traffic manifold is constructed using the real-time traffic flow data, traffic light phase sequence data, road topology metadata, and the smooth manifold. Then, a multi-agent forest model is constructed in step S3, obtaining the user-inputted commuter start and end points. The agent forest model generates a commuter score vector based on the commuter start, end, and dynamic traffic manifold. The commuter score vector includes a travel time score and an energy consumption score. The system generates a commuting score based on energy consumption and anomaly probability, and then uses step S4 to generate a scoring relationship equation. This equation is then used to integrate the commuting score vectors, resulting in a commuting path score set. The three highest-scoring commuting paths in the set are then marked and pushed to the user. By mapping the commuting network to a Lie group space, commuting differential equations are generated from travel time and energy consumption, and a dynamic traffic manifold is constructed. A multi-agent forest model is used to score the paths generated by the dynamic traffic manifold, taking into account various commuting factors to make the final commuting score more accurate. Furthermore, historical user data is incorporated to better meet user needs.

[0128] Step S5 obtains the user's final commuting route and traces the source of the final commuting route, generating a source tracing report. Finally, step S6 optimizes the dynamic traffic flow and returns to step S2. The source tracing method is used to find the factors affecting the score, and the influencing factors are analyzed to obtain the analysis results. Based on the analysis results, the dynamic traffic flow is optimized to achieve the purpose of optimization evaluation.

[0129] Furthermore, embodiments of the present invention also provide an electronic device, comprising:

[0130] At least one processor; and at least one memory communicatively connected to the processor; wherein the memory stores instructions executable by at least one processor, the instructions being executed by at least one processor to enable at least one processor to perform the method proposed in Embodiment 1 of the present invention.

[0131] The following is a detailed introduction to the various components of the electronic device:

[0132] In this context, the processor is the control center of the electronic device. It can be a single processor or a collective term for multiple processing elements. For example, a processor can be one or more central processing units (CPUs), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement Embodiment 1 of this invention, such as one or more digital signal processors (DSPs) or one or more field-programmable gate arrays (FPGAs).

[0133] The processor can perform various functions of an electronic device by running or executing software programs stored in memory and by calling data stored in memory.

[0134] The memory is used to store the software program that executes the solution of the present invention, and the execution is controlled by the processor. For specific implementation methods, please refer to the above method embodiments, which will not be repeated here.

[0135] The memory can be a real-only memory (ROM) or other type of static storage device capable of storing static information and instructions, a random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only (CD-ROM), or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media, or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto. The memory can be integrated with the processor or exist independently and coupled to the processor through an interface circuit of an electronic device; this embodiment of the invention does not specifically limit this.

[0136] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), 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 the present invention 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 limited means (e.g., infrared, wireless, microwave, etc.). 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.

[0137] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0138] It should be understood that, in the embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0139] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A commuter network evaluation method based on artificial intelligence, characterized in that, The method includes: S1: acquiring a commuter network and mapping it to a Lie group space to generate a smooth manifold of the commuter network on the Lie group space; the commuter network includes commuter nodes and commuter routes, mapping the commuter nodes and commuter routes to the Lie group space, wherein the commuter nodes correspond to elements in the Lie group space, and the commuter routes correspond to group action operators in the Lie group space, and generating a smooth manifold based on the commuter nodes and commuter routes; S2: acquiring real-time traffic flow data, traffic light phase time series, and road topology metadata, and constructing a dynamic traffic manifold based on the real-time traffic flow data, traffic light phase time series, road topology metadata, and the smooth manifold; acquiring real-time traffic flow data at a sampling frequency of 1 Hz. Next, the real-time traffic flow data includes vehicle position and motion state; traffic light phase timing is acquired with a time resolution of 0.1 seconds, including the change time and state of traffic signals; road topology metadata is acquired, including the number of lanes, slope, and radius of curvature; commuting positions of vehicles on a smooth manifold are acquired, and a traffic flow tensor is obtained based on the real-time traffic flow data and road topology metadata, representing the flow state of traffic flow, and the positions and the traffic flow tensor are combined to form a commuting vector; traffic signal changes are acquired based on the traffic light phase timing; the commuting vector is differentiated over time to obtain... S3: Construct a multi-agent forest model to obtain the user's input commuting start and end points. The agent forest model generates a commuting score vector based on the commuting start, end point, and dynamic traffic manifold. The commuting score vector includes a travel time score, an energy consumption score, and an abnormal event impact probability score. S4: Generate a score relationship equation based on the user's historical preferences. Integrate the commuting score vector based on the score relationship equation to obtain a commuting path score set. Mark the three commuting paths with the highest scores in the commuting path score set and push them to the user. S5: Obtain the user's final commuting route and trace its origin, generating a source tracing report; based on the user's selection, obtain the final commuting route and its corresponding rating, where the final commuting route represents the user's final selected commuting route; obtain the commuting rating vector corresponding to the rating, and perform partial derivative analysis on the corresponding commuting rating vector to obtain the sensitivity of the commuting rating vector, thus obtaining the commuting influence matrix; based on the commuting influence matrix, perform key feature filtering to obtain a commuting influence list; construct a dynamic decision tree based on the commuting influence list, and generate a source tracing report based on the dynamic decision tree; S6: optimize the dynamic traffic flow pattern based on user feedback and the source tracing report, and return to step S2.

2. The commuter network evaluation method based on artificial intelligence according to claim 1, characterized in that, The process involves constructing a multi-agent forest model to obtain the user's input commuting start and end points. The agent forest model generates a commuting score vector based on the commuting start, end point, and dynamic traffic flow. This commuting score vector includes a travel time score, an energy consumption score, and an abnormal event impact probability score. The process involves the user inputting the commuting start and end point into a mobile user terminal. The mobile user terminal analyzes the commuting start, end point, and dynamic traffic flow based on the multi-agent forest model to generate the commuting score vector. The multi-agent forest model includes a time agent, an energy consumption agent, and a risk agent.

3. The commuter network evaluation method based on artificial intelligence according to claim 2, characterized in that, The method further includes: the time agent predicting travel time based on LSTM, obtaining the predicted travel time, and scoring the predicted travel time to obtain a travel time score; the energy consumption agent solving for energy consumption based on dynamic equations, obtaining commuting energy consumption, and scoring the energy consumption to obtain an energy consumption score; and the risk agent analyzing the impact of abnormal events based on Bayesian networks, obtaining the probability of abnormal event impact, and scoring the probability of abnormal event impact to obtain an abnormal event impact probability score.

4. The commuter network evaluation method based on artificial intelligence according to claim 1, characterized in that, The process of generating a rating relationship equation based on user historical preferences, integrating commuting rating vectors based on the rating relationship equation to obtain a commuting route rating set, and marking and pushing the three highest-rated commuting routes in the commuting route rating set to the user includes: obtaining user historical preferences based on historical data; assigning ratings and weights to travel time ratings, energy consumption ratings, and abnormal event impact probability ratings based on user historical preferences, and generating a rating relationship equation; inputting the travel time rating, energy consumption rating, and abnormal event impact probability rating into the rating relationship equation to generate a commuting route rating set; extracting and marking the three highest-rated commuting routes in the commuting route rating set; and pushing the three highest-rated commuting routes to the user.

5. The commuter network evaluation method based on artificial intelligence according to claim 4, characterized in that, The process of extracting and marking the three highest-rated commuting routes from the commuting route rating set, and then pushing these three highest-rated commuting routes to the user, includes: the marking includes a time marker, an energy consumption marker, and a comprehensive marker; if the extracted commuting route has the highest travel time rating, it is marked as having the shortest travel time; if the extracted commuting route has the highest energy consumption rating, it is marked as having the lowest energy consumption; if the extracted commuting route has the highest overall rating, it is marked as having the highest comprehensive rating; and the three commuting routes and their corresponding ratings are then pushed to the user.

6. An artificial intelligence-based commuter network assessment system, used to implement the method of any one of claims 1-5, characterized in that, include: The acquisition module is used to acquire commuter network data, real-time traffic flow data, traffic light phase timing data, and road topology metadata, and to acquire the user's final commuting route. The generation module is used to map the commuting network to a Lie group space, generate a smooth manifold of the commuting network on the Lie group space, and generate a commuting score vector based on the start point, end point and dynamic traffic manifold of the agent forest model, and generate a score relationship equation based on the user's historical preferences. The module constructs a dynamic traffic manifold based on real-time traffic flow data, traffic light phase sequence, road topology metadata, and the smooth manifold, and builds a multi-agent forest model. The module integrates commuting score vectors based on the score relationship equation, obtains a commuting path score set, and marks and pushes the three commuting paths with the highest scores in the commuting path score set to the user. The tracing module is used to trace the final commuting route and generate a tracing report. An optimization module optimizes dynamic traffic flow patterns based on user feedback and source tracing reports.

Images