Unmanned online car-hailing scheduling method and device and storage medium

Abstract

The invention discloses an unmanned online car-hailing scheduling method and device and a storage medium, which are used for solving the technical problem that the service quality and the operation efficiency of unmanned online car-hailing cannot be guaranteed. The method comprises the following steps: when an order receiving system receives a user order, determining a departure place and a destination according to the user order; inputting the departure place and the destination into a pre-constructed route planning model to generate a plurality of routes; calculating a feeling degree index value of a preset feeling degree index of each route; obtaining a user demand; calculating a user evaluation index value of a preset user evaluation index according to the user demand and the user historical data; determining a target vehicle and a target route according to the user evaluation index value and the feeling index value; historical vehicle internal use state information of the target vehicle is obtained, and the vehicle state is adjusted according to the historical vehicle internal use state information; acquiring real-time road condition data of the target route; and route optimization is carried out according to the real-time road condition data and the user evaluation indexes.

Description

Technical Field

[0001] The present invention relates to the technical field of online car-hailing dispatching, and in particular to a method, device and storage medium for dispatching an unmanned online car-hailing vehicle. Background Art

[0003] Although online ride-hailing services provide many conveniences and innovations, there are still some shortcomings in actual operation. Due to differences in driver skills and service attitudes, the service quality of online ride-hailing services is inconsistent, making it difficult to ensure the standardization of each travel experience. The maintenance and upkeep of different vehicles vary, and there are differences in the comfort and safety of passengers.

[0004] Unmanned driving technology, also known as autonomous driving technology, refers to the ability of vehicles to drive autonomously without human intervention through the use of sensors, cameras, radars, lidar, global positioning systems (GPS), computer vision and artificial intelligence. The goal of this technology is to improve traffic safety, reduce traffic accidents, improve travel efficiency and reduce carbon emissions through automation. The introduction of unmanned driving technology is expected to solve many of the current shortcomings of the online car-hailing industry and improve service quality and operational efficiency. The application of unmanned driving technology in online car-hailing can effectively improve safety and reliability and provide people with more efficient, safe and environmentally friendly travel services. However, there is currently no complete technical solution to effectively combine unmanned driving with online car-hailing. The service quality and operational efficiency of unmanned online car-hailing cannot be guaranteed. Summary of the invention

[0005] The present invention provides a method, device and storage medium for dispatching driverless online-hailing vehicles, which are used to solve the technical problem that the service quality and operation efficiency of driverless online-hailing vehicles cannot be guaranteed.

[0006] The present invention provides a method for dispatching an unmanned online car-hailing vehicle, comprising:

[0007] When the order-taking system receives a user order, it determines the departure place and destination according to the user order;

[0008] Inputting the departure place and the destination into a pre-built route planning model to generate multiple routes;

[0009] Calculate the sensitivity index value of the preset sensitivity index of each route;

[0010] Obtain user needs;

[0011] Calculate the user evaluation index value of the preset user evaluation index according to the user demand and user historical data;

[0012] Determine a target vehicle and a target route according to the user evaluation index value and the sensitivity index value;

[0013] Acquiring historical vehicle internal usage status information of the target vehicle, and using the historical vehicle internal usage status information to adjust the vehicle status;

[0014] Acquiring real-time traffic data of the target route;

[0015] Route optimization is performed according to the real-time traffic data and the user evaluation index.

[0016] Optionally, before the step of inputting the departure place and the destination into a pre-built route planning model to generate a plurality of routes, the step further includes:

[0017] Collect online ride-hailing operation data, vehicle type rules and urban road rules;

[0018] Acquire a city road network, and generate a graph model using the city road network;

[0019] A route planning model is constructed using the online car-hailing operation data, the vehicle type rules, the urban road rules and the urban road network.

[0020] Optionally, the step of obtaining user needs includes:

[0021] Collect users’ historical order data, user evaluation data, and user personal information;

[0022] Cleaning the historical order data, the user evaluation data, and the user personal information to obtain cleaned data;

[0023] Performing standardization processing on the cleaned data to obtain standardized data;

[0024] Conducting factor analysis on the standardized data to extract key factors;

[0025] Conduct causal analysis on the key factors and determine user needs.

[0026] Optionally, the step of determining a target vehicle and a target route according to the user evaluation index value and the sensitivity index value includes:

[0027] Calculate the weight of each sensitivity index;

[0028] Calculate the comprehensive score of each route using the weight and the sensitivity index value;

[0029] determining a target route according to the comprehensive score;

[0030] Obtaining vehicle characteristics of each vehicle;

[0031] Calculating the degree of suitability of each of the vehicles using the user evaluation index value and the vehicle characteristics;

[0032] The target vehicle is determined using the fitness degree.

[0033] Optionally, the step of performing route optimization according to the real-time traffic data and the user evaluation index value includes:

[0034] The compliance of the user evaluation index is used as the reward function, and the route optimization is performed using the reinforcement learning algorithm and the real-time traffic data.

[0035] Optionally, the perception index includes: a comfort index, a safety index, an efficiency index, a scenery index and a body perception index.

[0036] Optionally, the vehicle status includes seat position, air-conditioning temperature, in-vehicle air humidity and music parameters.

[0037] Optionally, the step of acquiring real-time traffic data of the target route includes:

[0038] The local news and traffic information are read through the large model as the real-time traffic data of the target route.

[0039] The present invention also provides an unmanned driving online car-hailing dispatching device, comprising:

[0040] A departure and destination determination module, used to determine the departure and destination according to the user order when the order receiving system receives the user order;

[0041] A route generation module, used for inputting the departure place and the destination into a pre-built route planning model to generate multiple routes;

[0042] A sensitivity index value calculation module, used to calculate the sensitivity index value of the preset sensitivity index of each route;

[0043] User demand acquisition module, used to obtain user needs;

[0044] A user evaluation index value calculation module, used to calculate the user evaluation index value of a preset user evaluation index according to the user needs and user historical data;

[0045] A target vehicle and target route determination module, used to determine a target vehicle and a target route according to the user evaluation index value and the perception index value;

[0046] A vehicle state adjustment module, used to obtain historical vehicle internal use state information of the target vehicle and adjust the vehicle state using the historical vehicle internal use state information;

[0047] A real-time traffic data acquisition module, used to acquire the real-time traffic data of the target route;

[0048] The route optimization module is used to optimize the route according to the real-time traffic data and the user evaluation index.

[0049] The present invention also provides a computer-readable storage medium, which is used to store program code, and the program code is used to execute the unmanned driving online car-hailing scheduling method as described in any of the above items.

[0050] It can be seen from the above technical scheme that the present invention has the following advantages: The present invention provides a method for dispatching an unmanned online car-hailing vehicle, comprising: when an order-taking system receives a user order, determining a departure place and a destination according to the user order; inputting the departure place and the destination into a pre-built route planning model to generate multiple routes; calculating the sensitivity index value of a preset sensitivity index for each route; obtaining user demand; calculating the user evaluation index value of a preset user evaluation index according to the user demand and user historical data; determining a target vehicle and a target route according to the user evaluation index value and the sensitivity index value; obtaining historical vehicle internal usage status information of the target vehicle, and using the historical vehicle internal usage status information to adjust the vehicle status; obtaining real-time road condition data of the target route; and optimizing the route according to the real-time road condition data and the user evaluation index.

[0051] After receiving the user's order, the order-taking system of the present invention can quickly and efficiently plan the route according to the departure place and destination, generate multiple routes, ensure that the unmanned online car-hailing vehicle effectively learns the operation route of the manned online car-hailing vehicle in the past, ensure the safety and reliability of the route, and solve the problem of the singleness of the current online car-hailing vehicle route; provide multi-dimensional and multi-angle analysis of the route to improve the user experience according to the perception index; based on the user evaluation index generated by the user's needs, select several vehicles and several routes with the highest adaptability for the user; through personalized route selection, meet the user's needs, thereby improving the user experience. And then improve the service quality and operation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0053] Figure 1 A flowchart of the steps of a method for dispatching an unmanned online car-hailing vehicle provided by an embodiment of the present invention;

[0054] Figure 2 It is the logic diagram of the Agent execution graph algorithm;

[0055] Figure 3 A technical roadmap for autonomous ride-hailing dispatch methods;

[0056] Figure 4 A structural block diagram of an unmanned online car-hailing dispatching device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0057] The embodiments of the present invention provide a method, device and storage medium for dispatching driverless online-hailing vehicles, which are used to solve the technical problem that the service quality and operation efficiency of driverless online-hailing vehicles cannot be guaranteed.

[0058] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0059] See also Figure 1 , Figure 1 A flowchart of the steps of a method for dispatching an unmanned online car-hailing vehicle provided in an embodiment of the present invention.

[0060] The present invention provides a method for dispatching an unmanned online car-hailing vehicle, which may specifically include the following steps:

[0061] Step 101, when the order receiving system receives a user order, the departure place and the destination place are determined according to the user order;

[0062] Step 102, inputting the departure place and the destination into a pre-built route planning model to generate multiple routes;

[0063] In an embodiment of the present invention, when the order-taking system receives a user order, the departure place and the destination can be determined according to the user order, and then the departure place and the destination are input into a pre-built route planning model to generate multiple routes.

[0064] In one example, the step of building a route planning model includes the following sub-steps:

[0065] S11, collects online car-hailing operation data, vehicle form rules and urban road rules;

[0066] S12, obtaining an urban road network, and generating a graph model using the urban road network;

[0067] S13, constructs a route planning model using online ride-hailing operation data, vehicle form rules, urban road rules and urban road network.

[0068] In specific implementations, online car-hailing operation data may include operating routes, GPS trajectories, driving speeds, passenger boarding and alighting locations, etc.

[0069] Urban road rules may include: road speed limit information, number of lanes, road width, traffic light location, prohibited time periods, one-way street information, traffic sign information, etc.

[0070] Vehicle form rules may include: characteristics of different vehicle types, such as vehicle length, vehicle width, turning radius, maximum speed, acceleration performance, braking performance, etc.

[0071] The data set of the collected online car-hailing data is: , where (x i ,y i ) is the location information, t i The timestamp is obtained by cleaning, denoising and standardizing the collected data set. .

[0072] The urban road network can be represented as a graph model G = (V, E), where V is the node set (intersections or key locations) and E is the edge set (roads). The weight of the edge is , where d is the distance, s is the travel time, and c is the road condition.

[0073] Then, the vehicle form rules and urban road rules are integrated into the graph model, and the weight function is updated. , where r is the vehicle form rule and urban road rule.

[0074] Then, the updated graph model and online car-hailing operation data are combined to build a route planning model.

[0075] Since vehicle form rules and urban road rules are added to the route planning model, the route planning is more in line with actual traffic conditions and regulatory requirements. The route planning model that combines vehicle form rules and urban road rules can generate more reasonable and legal driving routes, reduce violation risks, and improve driving safety.

[0076] Next, the order information received by the order system Input route planning model and propose multiple routes. s is the departure place, O d For the destination.

[0077] Specifically, based on the graph model G and combined with the online car-hailing operation data, a route planning algorithm, such as Dijkstra or A* algorithm, can be used to generate multiple candidate routes.

[0078] The routes collection is as follows:

[0079]

[0080] Each route R i =

[0081] In one example, the route planning model can perform vehicle selection and route generation through the Agent execution graph model. The Agent execution graph model is composed of a vehicle as a single agent, and multiple vehicles collaborate to achieve multi-agent execution; each vehicle, as an independent agent, can make independent decisions on route selection based on real-time traffic conditions, vehicle status, and user needs. Through the collaborative work of multiple agents, the overall optimization of vehicle scheduling and route selection can be achieved to improve system efficiency. In a multi-agent system, each agent will consider the routes of other agents when calculating the route, avoiding route conflicts and resource waste, and achieving global optimal traffic scheduling.

[0082] Agent executes the graph algorithm logic such as Figure 2 As shown, specifically:

[0083] In the agent execution attribute graph algorithm, this module is responsible for collecting the attribute graph of each agent, including the characteristics, status and capabilities of the vehicle. The attribute graph is used to describe the performance and characteristics of the vehicle in different situations. The basic attributes and status information of the vehicle are used as input, and the agent attribute graph is used as output to serve as the decision basis for the subsequent action module.

[0084] In the Agent Execution Process Graph algorithm, this module is responsible for collecting the execution process graph of each agent, which specifically describes the entire process from order acceptance to order completion. The execution process graph is used to guide the specific operations of the agent at different stages. The order information and the current status of the vehicle are used as input, and the agent execution process graph is used as output to guide the agent's operation steps.

[0085] In the matching action algorithm module, the module matches the appropriate action algorithm according to the agent's attribute graph and execution process graph. The matched action algorithm will guide the agent on how to perform operations in specific scenarios. The agent's attribute graph and execution process graph are used as input. The matched action algorithm is used as output to guide specific operations. In the command action algorithm module, this module is responsible for commanding and controlling the agent's action execution, and specifically performs operations according to the matched action algorithm. The command action algorithm module ensures that the execution of actions complies with predetermined strategies and rules. The matching action algorithm is used as input. The specific control command is used as output to perform specific operations. In the execution action algorithm module, this module actually performs specific actions and performs specific operations such as driving operations, route adjustments, etc. according to the control commands generated by the command action algorithm module. The control command is used as input and the actual operation is used as the behavior output.

[0086] In the Agent Execution Behavior Code Intelligent Generation Algorithm, this module generates behavior codes in multi-agent systems and intelligently coordinates the behaviors of multiple agents. This algorithm ensures that multiple agents can collaborate efficiently and avoid conflicts. The attribute graphs and execution process graphs of multiple agents are used as input, and the behavior codes of multiple agents are output to guide collaboration as behavior. In the Agent Execution Behavior Interaction Protocol Algorithm, this module designs the interaction protocol between multiple agents to ensure that multiple agents can communicate and coordinate smoothly during the collaboration process, avoiding route conflicts and waste of resources. The multi-agent behavior code is used as input, and the multi-agent interaction protocol is used as output to guide the interaction between multiple agents.

[0087] In the multi-agent interaction action algorithm module, this module executes the interaction actions of multi-agents and actually performs collaborative operations according to the interaction protocol. This algorithm ensures that the multi-agent system can achieve efficient collaboration in actual operation. The multi-agent interaction protocol is used as input and the collaborative actions of multi-agents are used as output.

[0088] In the specific implementation, taking the A* algorithm as an example, the route selection is as follows:

[0089] The starting node is set to the current location of the vehicle, and the target node is set to the user's destination. Calculate the heuristic estimation function h(n) from the starting point to the target node, and select the node with the smallest f(n)=g(n)+h(n) in the open list. g(n) is the cost from the starting point to the current node, and h(n) is the estimated cost from the current node to the target node. Update the neighbor nodes of the current node. If the new route is better, update the cost of the neighbor nodes. Repeat the above steps until the target node is found. Finally, combine the departure node and the destination node to form a complete driving route.

[0090] It should be noted that when planning a route, each agent broadcasts its route plan to other agents. If a route conflict is found, the conflict is resolved by adjusting the route or waiting. Through collaborative planning, the overall travel time of all agents is optimized.

[0091] Through the above route planning model, a candidate route from the departure point to the destination can be generated for each user, and route conflicts between different users can be avoided.

[0092] Step 103, calculating the sensitivity index value of the preset sensitivity index of each route;

[0093] In the embodiment of the present invention, the perception index may include: a comfort index, a safety index, an efficiency index, a scenery index and a body perception index.

[0094] Comfort index: Comfort (R i ), including driving smoothness, road surface smoothness, and noise level;

[0095] Safety index: Safety (R i ), including the road accident rate and the number of large trucks;

[0096] Efficiency index: Efficiency (R i ), including waiting time, driving time, distance, and speed;

[0097] Scenery index: Scenery (R i ), including scenery, sunshine, and air;

[0098] UserExperience (R i ), including in-car facilities (air conditioning, refrigerator, audio, windows).

[0099] After obtaining each perception index, the perception index of each route can be quantified to obtain the corresponding perception index value, and the evaluation vector of each route can be calculated based on the perception index of each route: E i

[0100] Step 104, obtaining user needs;

[0101] In the embodiment of the present invention, demand analysis can be performed based on user habits, and causal analysis can be constructed through factor extraction and analysis to determine user needs.

[0102] In one example, the step of obtaining user requirements may include the following sub-steps:

[0103] S41, collecting user's historical order data, user evaluation data, and user personal information;

[0104] S42, performing data cleaning on historical order data, user evaluation data, and user personal information to obtain cleaned data;

[0105] S43, performing standardization processing on the cleaned data to obtain standardized data;

[0106] S44, factor analysis was performed on the standardized data to extract key factors;

[0107] S45, conduct causal analysis on key factors and determine user needs.

[0108] In the specific implementation, we can first collect user historical order data, user evaluation data and user personal information to obtain the user data set: ,in, For user personal information, For the user's historical order data, Evaluate data for users.

[0109] Then, the collected data is cleaned and standardized to obtain standardized data. The purpose of data cleaning is to remove duplicate data and abnormal data and fill in missing data. The purpose of data standardization is to standardize data of different dimensions to ensure data comparability.

[0110] Factor analysis was then used to extract key factors from the standardized data .

[0111] Factor analysis is specifically used to analyze the relationship between the extracted features and their impact on user needs; principal component analysis (PCA) is used to reduce the dimension, reduce the number of features, and retain the main influencing factors, i.e. key factors. Cluster analysis can also be used to classify users and find user groups with similar needs.

[0112] In the specific implementation, key factors may include but are not limited to: order frequency (the frequency of users using online ride-hailing services), order time (such as rush hour, night travel, etc.), travel distance (short distance travel, long distance travel), evaluation score (user satisfaction with the service), and user preferences (such as favorite car type, driver service attitude, etc.).

[0113] Then, we conduct causal analysis on the key factors collected to determine user needs.

[0114] The causal model is:

[0115] In one example, a Bayesian network can be used for causal analysis. Based on the inference results of the Bayesian network, the user's demand patterns and preferences can be determined. For example, if the model shows that users frequently use online ride-hailing services on weekday mornings and have high ratings, it can be inferred that users may need fast and safe commuting services. The needs of different user groups are summarized and user demand reports are generated to provide a basis for route planning and service optimization.

[0116] A Bayesian network is a probabilistic graphical model used to represent conditional dependencies between variables. The construction process of a Bayesian network is as follows:

[0117] Variable selection: Select the main characteristic variables in the analysis, such as user evaluation, order frequency, travel time, etc.

[0118] Structural learning: Use data-driven methods to learn the dependencies between variables. Greedy algorithms or Monte Carlo methods can be used to build network structures.

[0119] Parameter learning: Based on historical data, calculate the conditional probability distribution between variables in the Bayesian network.

[0120] The following is a specific example to illustrate the user's needs:

[0121] Assume the following user data:

[0122] User A: He uses online taxis from 7 to 8 every morning, travels about 10 kilometers, and has a high evaluation score.

[0123] User B: He uses online taxis every Friday night, with a travel distance of about 15 kilometers and a medium evaluation score.

[0124] User C: uses online taxis at random times, travels about 5 kilometers, and has a low evaluation score.

[0125] Through the above steps, we can find that:

[0126] User A's demand is a fast and safe commuting service.

[0127] User B may need a vehicle that is more comfortable and has a good service attitude.

[0128] User C's needs may be more focused on price.

[0129] Based on these needs, the system can provide personalized service solutions for different user groups and improve user satisfaction.

[0130] Step 105, calculating a user evaluation index value of a preset user evaluation index according to user needs and user historical data;

[0131] After obtaining user needs, we can map user needs into user evaluation indicators and quantify them to obtain:

[0132]

[0133] in, It is the user evaluation index value.

[0134] Step 106, determining a target vehicle and a target route according to the user evaluation index value and the perception index value;

[0135] In this embodiment of the present invention, step 106 includes:

[0136] S61, calculating the weight of each sensitivity index;

[0137] S62, using weights and perception index values ​​to calculate the comprehensive score of each route;

[0138] S63, determining a target route according to the comprehensive score;

[0139] S64, obtaining vehicle characteristics of each vehicle;

[0140] S65, calculating the adaptability of each vehicle using the user evaluation index value and the vehicle characteristics;

[0141] S66, determining the target vehicle using the degree of fitness.

[0142] In the specific implementation, the comprehensive score of each route can be calculated based on the user evaluation index value:

[0143]

[0144] in, is the jth evaluation index of the i-th route.

[0145] Select the best route based on the comprehensive score:

[0146]

[0147] The vehicle's fitness calculation process is as follows:

[0148]

[0149] in, Characteristics of the vehicle.

[0150] Step 107, obtaining historical vehicle internal usage status information of the target vehicle, and using the historical vehicle internal usage status information to adjust the vehicle status;

[0151] In an embodiment of the present invention, the user can obtain the historical vehicle interior usage status information of the target vehicle before getting on the vehicle to adjust the vehicle status, wherein the vehicle status may include seats, temperature, sound, etc.

[0152] Historical vehicle internal usage status information:

[0153]

[0154] Adjustment operation:

[0155]

[0156] Step 108, obtaining real-time traffic data of the target route;

[0157] In the embodiment of the present invention, the real-time traffic data can be obtained from a local news flash platform or a traffic information service provider.

[0158] In one example, the step of obtaining real-time traffic data of a target route includes:

[0159] Local news and traffic information are read through the large model as real-time traffic data for the target route.

[0160] In one example, you can access a local news platform through an API interface to obtain the latest information on traffic events, road construction, and traffic accidents. You can access a traffic information service provider through an API interface to obtain real-time traffic flow, road congestion, vehicle speed, and other data.

[0161] Then the local news data and traffic information data are integrated, the data format and timestamp are unified to ensure the consistency and accuracy of the data, and the data is cleaned to remove redundant and noisy data and retain valid information.

[0162] Then, the large model is connected in real time to read local news and traffic information to obtain real-time traffic data. Specifically, the collected data can be input into a machine learning model (such as LSTM, Transformer, etc.) for real-time processing and analysis of traffic data.

[0163] The traffic data obtained is:

[0164] The training and deployment process of the machine learning model is as follows:

[0165] Training data preparation: Use historical traffic data and road condition information as training data, including information such as traffic flow and vehicle speed in different time periods and weather conditions.

[0166] Model training: Use deep learning frameworks (such as TensorFlow, PyTorch, etc.) to train models and learn patterns of traffic flow and road condition changes.

[0167] Model deployment: Deploy the trained model to the cloud or edge computing devices to receive and process the latest traffic information in real time.

[0168] Step 109, optimize the route according to the real-time traffic data and user evaluation indicators.

[0169] In an embodiment of the present invention, the compliance of the user evaluation index can be used as a reward function, and a reinforcement learning algorithm and real-time traffic data can be used to perform route optimization.

[0170] In one example, the reinforcement learning algorithm may be a DQN algorithm.

[0171] Furthermore, in an embodiment of the present invention, after completing an order, the user can provide evaluation feedback, and the order-taking system performs self-inspection on the evaluation indicators and supplements or adjusts the evaluation indicators based on the degree of overlap.

[0172] User feedback:

[0173] Self-check formula:

[0174] Indicator adjustment:

[0175] For easier understanding, please refer to Figure 3 , Figure 3 A technical roadmap for autonomous ride-hailing dispatch methods.

[0176] After receiving the user's order, the order-taking system of the present invention can quickly and efficiently plan the route according to the departure place and destination, generate multiple routes, ensure that the unmanned online car-hailing vehicle effectively learns the operation route of the manned online car-hailing vehicle in the past, ensure the safety and reliability of the route, and solve the problem of the singleness of the current online car-hailing vehicle route; provide multi-dimensional and multi-angle analysis of the route to improve the user experience according to the perception index; based on the user evaluation index generated by the user's needs, select several vehicles and several routes with the highest adaptability for the user; through personalized route selection, meet the user's needs, thereby improving the user experience. And then improve the service quality and operation efficiency.

[0177] See also Figure 4 , Figure 4 A structural block diagram of an unmanned online car-hailing dispatching device provided in an embodiment of the present invention.

[0178] The embodiment of the present invention provides an unmanned driving online car-hailing dispatching device, comprising:

[0179] The departure and destination determination module 401 is used to determine the departure and destination according to the user order when the order receiving system receives the user order;

[0180] A route generation module 402 is used to input the departure point and the destination into a pre-built route planning model to generate multiple routes;

[0181] The sensitivity index value calculation module 403 is used to calculate the sensitivity index value of the preset sensitivity index of each route;

[0182] User demand acquisition module 404, used to acquire user demand;

[0183] The user evaluation index value calculation module 405 is used to calculate the user evaluation index value of the preset user evaluation index according to the user demand and the user historical data;

[0184] A target vehicle and target route determination module 406 is used to determine a target vehicle and a target route according to a user evaluation index value and a perception index value;

[0185] The vehicle state adjustment module 407 is used to obtain the historical vehicle internal use state information of the target vehicle and adjust the vehicle state using the historical vehicle internal use state information;

[0186] A real-time traffic data acquisition module 408 is used to acquire real-time traffic data of a target route;

[0187] The route optimization module 409 is used to optimize the route according to the real-time traffic data and user evaluation indicators.

[0188] In an embodiment of the present invention, it also includes:

[0189] The module for collecting online car-hailing operation data, vehicle form rules and urban road rules is used to collect online car-hailing operation data, vehicle form rules and urban road rules;

[0190] A graph model generation module, used for acquiring a city road network and generating a graph model using the city road network;

[0191] The route planning model building module is used to build a route planning model using online car-hailing operation data, vehicle form rules, urban road rules and urban road networks.

[0192] In the embodiment of the present invention, the user demand acquisition module 404 includes:

[0193] Historical order data, user evaluation data, and user personal information collection submodule, used to collect users' historical order data, user evaluation data, and user personal information;

[0194] The data cleaning submodule is used to clean the historical order data, user evaluation data and user personal information to obtain cleaned data;

[0195] The standardization processing submodule is used to perform standardization processing on the cleaned data to obtain standardized data;

[0196] The factor analysis submodule is used to perform factor analysis on the standardized data and extract key factors;

[0197] The causal analysis submodule is used to conduct causal analysis on key factors and determine user needs.

[0198] In the embodiment of the present invention, the target vehicle and target route determination module 406 includes:

[0199] The weight calculation submodule is used to calculate the weight of each sensitivity index;

[0200] The comprehensive score calculation submodule is used to calculate the comprehensive score of each route using weights and perception index values;

[0201] A target route determination submodule is used to determine the target route according to the comprehensive score;

[0202] A vehicle feature acquisition submodule is used to acquire vehicle features of each vehicle;

[0203] A fitness calculation submodule is used to calculate the fitness of each vehicle using the user evaluation index value and vehicle characteristics;

[0204] The target vehicle determination submodule is used to determine the target vehicle using the fitness degree.

[0205] In the embodiment of the present invention, the route optimization module 409 includes:

[0206] The route optimization submodule is used to optimize the route using the compliance of user evaluation indicators as the reward function, and adopts reinforcement learning algorithm and real-time traffic data.

[0207] In the embodiment of the present invention, the perception index includes: a comfort index, a safety index, an efficiency index, a scenery index and a body perception index.

[0208] In the embodiment of the present invention, the vehicle status includes seat position, air-conditioning temperature, in-vehicle air humidity and music parameters.

[0209] In the embodiment of the present invention, the real-time traffic data acquisition module 408 includes:

[0210] The real-time traffic data acquisition submodule is used to read local news and traffic information through the large model as the real-time traffic data of the target route.

[0211] An embodiment of the present invention also provides a computer-readable storage medium, characterized in that the computer-readable storage medium is used to store program code, and the program code is used to execute the unmanned driving online car-hailing scheduling method of an embodiment of the present invention.

[0212] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0213] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0214] It will be appreciated by those skilled in the art that the embodiments of the present invention may be provided as methods, devices, or computer program products. Therefore, the embodiments of the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the embodiments of the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0215] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0216] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0217] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0218] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.

[0219] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0220] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for dispatching driverless online car-hailing vehicles, characterized in that: include: When the order-taking system receives a user order, it determines the departure place and destination according to the user order; Inputting the departure place and the destination into a pre-built route planning model to generate multiple routes; Calculate the sensitivity index value of the preset sensitivity index of each route; Obtain user needs; Calculate the user evaluation index value of the preset user evaluation index according to the user demand and user historical data; Determine a target vehicle and a target route according to the user evaluation index value and the sensitivity index value; Acquiring historical vehicle internal usage status information of the target vehicle, and using the historical vehicle internal usage status information to adjust the vehicle status; Acquiring real-time traffic data of the target route; Route optimization is performed according to the real-time traffic data and the user evaluation index.

2. The method according to claim 1, characterized in that Before the step of inputting the departure place and the destination into a pre-built route planning model to generate a plurality of routes, the method further includes: Collect online ride-hailing operation data, vehicle type rules and urban road rules; Acquire a city road network, and generate a graph model using the city road network; A route planning model is constructed using the online car-hailing operation data, the vehicle type rules, the urban road rules and the urban road network.

3. The method according to claim 1, characterized in that The step of obtaining user needs includes: Collect users’ historical order data, user evaluation data, and user personal information; Cleaning the historical order data, the user evaluation data, and the user personal information to obtain cleaned data; Performing standardization processing on the cleaned data to obtain standardized data; Conducting factor analysis on the standardized data to extract key factors; Conduct causal analysis on the key factors and determine user needs.

4. The method according to claim 1, characterized in that: The step of determining the target vehicle and the target route according to the user evaluation index value and the sensitivity index value comprises: Calculate the weight of each sensitivity index; Calculate the comprehensive score of each route using the weight and the sensitivity index value; determining a target route according to the comprehensive score; Obtain vehicle characteristics of each vehicle; Calculating the degree of suitability of each of the vehicles using the user evaluation index value and the vehicle characteristics; The target vehicle is determined using the fitness degree.

5. The method according to claim 1, characterized in that The step of performing route optimization according to the real-time traffic data and the user evaluation index value comprises: The compliance of the user evaluation index is used as the reward function, and the route optimization is performed using the reinforcement learning algorithm and the real-time traffic data.

6. The method according to claim 1, characterized in that The perception index includes: comfort index, safety index, efficiency index, scenery index and body perception index.

7. The method according to claim 1, characterized in that The vehicle status includes seat position, air conditioning temperature, in-vehicle air humidity and music parameters.

8. The method according to claim 1, characterized in that The step of obtaining the real-time traffic data of the target route includes: The local news and traffic information are read through the large model as the real-time traffic data of the target route.

9. An unmanned online car-hailing dispatching device, characterized in that: include: A departure and destination determination module, used to determine the departure and destination according to the user order when the order receiving system receives the user order; A route generation module, used for inputting the departure place and the destination into a pre-built route planning model to generate multiple routes; A sensitivity index value calculation module, used to calculate the sensitivity index value of the preset sensitivity index of each route; User demand acquisition module, used to obtain user needs; A user evaluation index value calculation module, used to calculate the user evaluation index value of a preset user evaluation index according to the user needs and user historical data; A target vehicle and target route determination module, used to determine a target vehicle and a target route according to the user evaluation index value and the perception index value; A vehicle state adjustment module, used to obtain historical vehicle internal use state information of the target vehicle and adjust the vehicle state using the historical vehicle internal use state information; A real-time traffic data acquisition module, used to acquire the real-time traffic data of the target route; The route optimization module is used to optimize the route according to the real-time traffic data and the user evaluation index.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store program code, and the program code is used to execute the unmanned driving online car-hailing dispatching method described in any one of claims 1-7.

Images