Shared riding system and method for unmanned vehicle

Abstract

The invention relates to the technical field of traffic management, in particular to a shared riding system and method for an unmanned vehicle, and the method comprises the steps: obtaining vehicle price, starting point and target point information inputted by a user, and generating vehicle request information; matching available vehicles closest to the starting point based on the vehicle using request information and generating a scheduling instruction; after receiving the dispatching instruction, the vehicle moves to a starting point, and the user is sent to a target point after the identity of the user is confirmed through biological recognition; after the vehicle reaches the target point, distributing potential vehicle demand parking points for the vehicle based on historical vehicle data, and moving the vehicle to the potential vehicle demand parking points; idle vehicles can be efficiently utilized through the unmanned driving technology, the vehicle utilization rate is increased, and occupied parking spaces are reduced.

Description

Technical Field

[0001] The present invention relates to the field of traffic management technology, and in particular to a ride-sharing system and method for unmanned vehicles. Background Art

[0002] Privately owned motor vehicles still dominate the current urban transportation system. However, this predominantly individual-owned travel model has exposed numerous resource inefficiencies in practice, particularly low seat utilization and excessive parking space occupation.

[0003] In terms of vehicle seat utilization, most private cars are designed to carry four to five passengers. However, in daily use, these vehicles are often used only for single commuting or short trips with a small group of people. Statistics show that during peak commuting hours in cities, the average number of passengers per private car is less than 1.5, meaning that over 60% of seats are vacant.

[0004] Secondly, private cars are often parked when not in use, occupying a significant portion of a city's public parking resources. Whether in residential areas, commercial centers, or office spaces, parking lots are often saturated, and these parked vehicles remain largely unused. Particularly in urban core areas, the imbalance between parking supply and demand is becoming increasingly severe, leading to frequent parking difficulties and illegal parking, severely impacting urban traffic order and the efficient allocation of land resources. Summary of the Invention

[0005] The purpose of the present invention is to provide a shared ride system and method for unmanned vehicles, aiming to efficiently utilize idle vehicles through unmanned driving technology, improve vehicle utilization, and reduce parking space occupancy.

[0006] To achieve the above-mentioned object, in a first aspect, the present invention provides a method for sharing a ride in an unmanned vehicle, comprising obtaining a vehicle price, a starting point, and a destination point information input by a user, and generating vehicle request information;

[0007] Based on the vehicle request information, the nearest available vehicle to the starting point is matched and a dispatch instruction is generated;

[0008] After receiving the dispatch instruction, the vehicle moves to the starting point and sends the user to the destination after confirming the user's identity through biometric recognition;

[0009] After the vehicle reaches the target point, a potential parking spot with vehicle usage demand is allocated to the vehicle based on historical vehicle usage data, and the vehicle is moved to the potential parking spot with vehicle usage demand.

[0010] The specific steps of obtaining the car price, departure point, and destination point information input by the user and generating the car request information include:

[0011] Users can select the corresponding vehicle price range based on the dynamic price matrix interface and the preset price range divided by vehicle type and grade;

[0012] Call the terminal GPS module to obtain the current location as the default starting point;

[0013] After receiving the target point keyword input by the user, the starting point and destination point are adjusted, and the user request information is generated in combination with the car price.

[0014] The specific steps of matching the available vehicle closest to the starting point based on the vehicle request information and generating a dispatch instruction include:

[0015] Filter the corresponding vehicle group in the vehicle database based on the price information in the user's order, and verify the remaining mileage of the vehicles in the vehicle group;

[0016] A dynamic search area centered on the starting point is established, searching the vehicle group for vehicles of the same price within a 500-meter radius of the starting point. If no vehicle is available, the radius is expanded to 1.5 kilometers and the vehicle configuration requirements are lowered. If no matching vehicle is found, cross-regional dispatch is triggered, calculating the vehicle allocation cost of surrounding service stations. Otherwise, the target vehicle is found.

[0017] Issue dispatch instructions to the target vehicle and generate an estimated arrival time.

[0018] The specific steps of screening a corresponding vehicle group in the vehicle database according to the price information in the user order and verifying the remaining cruising range of the vehicles in the vehicle group include:

[0019] Obtaining the availability of each vehicle in the vehicle group, including whether the vehicle is booked and whether it is in maintenance status;

[0020] After the vehicle availability is passed, checking whether its remaining range meets the user's travel requirements, wherein the travel requirements are derived based on the starting point and the destination point;

[0021] The vehicles that pass the travel requirements are grouped together into a verified vehicle group.

[0022] Wherein, after issuing a dispatch instruction to the target vehicle and generating an estimated arrival time, the step further includes: re-matching the vehicle if a vehicle failure occurs or the estimated arrival time exceeds a preset value.

[0023] The specific steps of the vehicle moving to the starting point after receiving the dispatch instruction and sending the user to the destination after confirming the user's identity through biometric recognition include:

[0024] Obtain dispatch instructions and send arrival information to the user after the target vehicle arrives at the designated boarding point according to the dispatch instructions;

[0025] Continuously collect video data around the vehicle and search for passengers. The onboard camera captures the user's facial features and compares them with the order-bound information. If the comparison is successful, the door corresponding to the target seat will be unlocked;

[0026] Dynamically plan the optimal path to the destination based on real-time traffic conditions;

[0027] After arriving at the destination, the in-car sensor scans to confirm that all passengers and belongings have left, and automatically generates an electronic bill and deducts the fee;

[0028] Send a mission completion signal to the dispatch center and put the vehicle into idle state.

[0029] The specific steps of allocating a potential parking spot for the vehicle based on historical vehicle usage data after the vehicle reaches the target point and moving the vehicle to the potential parking spot include:

[0030] When the remaining power of the vehicle is lower than the preset value, the nearest available charging station is selected for charging; when the remaining power of the vehicle is higher than the preset value, the vehicle enters the standby point selection;

[0031] When selecting a standby point, a recommended parking spot is generated based on historical vehicle usage data combined with current parking time and area;

[0032] Based on the recommended parking spots, the vehicle drives along the planned route, communicates with the parking lot gate, and obtains parking spaces with priority; it parks accurately in the parking space through surround-view cameras and ultrasonic sensors, and feeds back the final position to the server.

[0033] The specific steps of generating a recommended parking spot based on historical vehicle usage data and current parking time and area when selecting a standby spot include:

[0034] Retrieve the order origination location and average waiting time for the same period in the target area over the past 30 days from the order database;

[0035] The geographic space is divided into multiple small areas through gridded city maps and temporal features are extracted to generate training and validation sets.

[0036] Train a polynomial regression model using the training set data;

[0037] Evaluate and optimize the polynomial regression model on the test set;

[0038] The current parking time and area are input to generate recommended parking spots based on a polynomial regression model.

[0039] In a second aspect, the present invention also provides a shared ride system for unmanned vehicles, which is applied to the shared ride method for unmanned vehicles.

[0040] The unmanned vehicle shared ride system includes a user terminal module, a cloud dispatch center, an unmanned vehicle control module, a communication network module, and a security management and identity authentication module;

[0041] The user terminal module provides a user input interface for inputting the price of the vehicle, the departure point and the destination point information;

[0042] The cloud dispatch center receives and analyzes the vehicle request information sent by the user terminal, and makes vehicle matching and dispatch decisions based on real-time data;

[0043] The autonomous vehicle control module receives dispatch instructions from the cloud dispatch center and automatically completes the driving task from the current location to the boarding point and then to the destination. After receiving the user identity verification signal, it unlocks the door of the corresponding seat. After arriving at the destination, it determines whether the passenger has disembarked based on the in-vehicle sensors and triggers the bill generation and upload of the task completion signal. After the task is completed, it enters the charging station or goes to a potential high-demand area to wait according to the system recommendation.

[0044] The communication network module is used for communication;

[0045] The security management and identity verification module is used for user identity verification.

[0046] The present invention provides a ride-sharing system and method for autonomous vehicles. Users enter vehicle information, including their desired price range, departure location, and destination, via a mobile terminal or in-vehicle device. A vehicle request packet is automatically generated based on the user's input and uploaded to a cloud-based dispatch platform. Vehicle Scheduling and Matching: Upon receiving the request, the cloud-based dispatch platform uses a multi-objective optimization algorithm to match the most suitable autonomous vehicle based on real-time traffic conditions, vehicle distribution, and historical operating data. Matching criteria include, but are not limited to, the nearest available vehicle to the departure point, the vehicle's current battery / fuel status, the estimated arrival time at the departure point, and whether it meets the user's specific needs. Once the matching is complete, the system generates a dispatch instruction and sends it to the target vehicle. The user pick-up and drop-off process is executed. Upon receiving the dispatch instruction, the autonomous vehicle automatically starts and travels along the optimal route to the user's departure point. Before the user boards the vehicle, the vehicle verifies the user's identity using an integrated biometric recognition system (such as facial recognition, fingerprint recognition, or iris recognition) to ensure safety and exclusive service. After verification, the vehicle takes the user to the destination along a pre-set route, dynamically adjusting the route based on real-time traffic conditions to ensure timeliness. Destination Arrival and Subsequent Scheduling: When the vehicle arrives at the user's destination, the trip is complete. The system records relevant data for the trip, including mileage, travel time, energy consumption, and user ratings, and updates it to a historical database. Subsequently, based on the current time, regional traffic volume, weather conditions, and historical vehicle usage patterns, the system predicts new demand points that may arise in the future and recommends one or more potential parking spots for the vehicle. Based on the recommendations, the vehicle proceeds to the designated area and stands by to quickly respond to the next request, achieving efficient utilization of vehicle and road resources and reducing resource waste. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0048] Figure 1 This is a flow chart of a method for sharing a ride in an unmanned vehicle according to the present invention.

[0049] Figure 2 This is a flow chart of the present invention for obtaining the car price, starting point and destination point information input by the user and generating car request information.

[0050] Figure 3 This is a flowchart of the present invention for matching the available vehicle closest to the starting point based on the vehicle request information and generating a dispatch instruction.

[0051] Figure 4 This is a flowchart of the present invention for screening corresponding vehicle groups in a vehicle database according to price information in a user order, and verifying the remaining cruising range of vehicles in the vehicle group.

[0052] Figure 5 This is a flow chart of the present invention in which a vehicle moves to a starting point after receiving a dispatch instruction, and after confirming the user's identity through biometric recognition, sends the user to the destination point.

[0053] Figure 6 This is a flowchart of the present invention for allocating a potential parking point for the vehicle based on historical vehicle usage data after the vehicle reaches the target point, and moving the vehicle to the potential parking point.

[0054] Figure 7 This is a flow chart of generating recommended parking spots based on historical vehicle usage data combined with current parking time and area when selecting a standby spot according to the present invention. DETAILED DESCRIPTION

[0055] The following describes embodiments of the present invention in detail, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and are not to be construed as limiting the present invention.

[0056] First embodiment

[0057] See also Figures 1 to 7 The present invention provides a method for sharing a ride in an unmanned vehicle, comprising:

[0058] S101 obtains the car price, departure point and destination point information input by the user and generates car request information;

[0059] The specific steps include:

[0060] S201 users can select the corresponding vehicle price range based on the dynamic price matrix interface and the preset price range divided by vehicle type and grade;

[0061] Dynamically adjust vehicle prices for different regions or routes based on real-time supply and demand, traffic flow, peak hours, and other factors. A visual price matrix interface can be displayed on the user app or terminal device, showing various models and their corresponding price ranges. Models can be divided into multiple grades, such as economy, comfort, business, and luxury, each with a different base unit price and floating rules.

[0062] Users can select the desired price range of the car or directly select a specific model on the interface by clicking, sliding, etc.

[0063] After the user makes a selection, the system caches the selected price information locally and passes it to subsequent modules for vehicle screening.

[0064] If the user does not make a clear selection, a default price range can be automatically recommended based on their historical preferences.

[0065] S202 calls the terminal GPS module to obtain the current location as the default starting point;

[0066] When the user starts the car-use function, the system automatically requests permission to use the terminal device's positioning service. Once authorization is successful, the built-in GPS module or third-party map SDK (such as Gaode, Baidu, Google Maps) is immediately called to obtain the current location coordinates. After obtaining the longitude and latitude information, it is converted into a standard address format (such as street name, house number, etc.) through the geocoding service to enhance the user experience. The parsed address information is displayed as the default starting point in the map interface and order confirmation page. Allow users to drag markers on the map or manually enter the address to change the starting point to deal with inaccurate positioning, temporary changes, etc. The modified starting point information will also be recorded and used for subsequent route planning and vehicle matching.

[0067] S203 receives the target point keyword input by the user, adjusts the starting point and the destination point, and generates user request information in combination with the car price.

[0068] In the target point input box, users can enter keywords (such as company names, landmarks, and place names). The system then uses the reverse geocoding interface or the POI (Point of Interest) database for fuzzy matching, presenting a list of candidate locations for the user to choose from. Multiple interaction methods, including voice input and handwriting recognition, are supported to improve input efficiency.

[0069] The selected starting and destination points are checked for plausibility, including whether they are located in no-parking zones or under construction. If any anomalies are detected, a warning prompt will appear and alternative suggestions will be provided. Preliminary estimates of route distance and estimated time can be made based on real-time traffic information, enhancing user experience.

[0070] A structured user request data packet is generated based on the user's selected car price, starting point coordinates and address, destination point coordinates and address, car usage time and other information.

[0071] S102 matches the nearest available vehicle to the starting point based on the vehicle request information and generates a dispatch instruction;

[0072] The specific steps include:

[0073] S301: Filter the corresponding vehicle group in the vehicle database according to the price information in the user order, and verify the remaining mileage of the vehicles in the vehicle group;

[0074] The specific steps include:

[0075] S401 obtains the availability of each vehicle in the vehicle group, including whether it is booked and whether it is in maintenance status;

[0076] The status information of all vehicles is read from the platform's central dispatch database, including whether they are performing tasks (whether they have been booked by other users), whether they are in repair or maintenance status, and whether they are being charged or undergoing system upgrades.

[0077] Only vehicles that are currently untasked, unmaintained, and dispatchable are retained as candidates for the next round of evaluation.

[0078] S402 checks whether the remaining range of the vehicle meets the user's travel requirements after the vehicle availability is passed, where the travel requirements are derived based on the starting point and the destination point;

[0079] The user's estimated driving distance is calculated based on the starting point and destination point; the remaining range of candidate vehicles is queried, and this data can be uploaded to the cloud server in real time through the on-board battery management system (BMS); if the remaining range of a vehicle is less than the mileage required for this trip, it will be excluded to avoid the situation where the service cannot be completed midway.

[0080] S403 gathers the vehicles that meet the travel requirements into a verified vehicle group.

[0081] Vehicles that meet both the "availability" and "endurance" conditions will be formed into a new set of candidate vehicles, and the optimal vehicle will be further matched in this set later.

[0082] S404: Rematch the vehicle if a vehicle failure occurs or the estimated arrival time exceeds a preset value.

[0083] For vehicles that have been preliminarily selected, their real-time status will be verified again before the official dispatch. If a sudden vehicle failure is found (such as perception module abnormality, communication interruption, etc.) or the estimated arrival time (ETA) exceeds the maximum waiting time threshold acceptable to the user (for example, 15 minutes) due to traffic conditions, the vehicle dispatch will be canceled immediately and a new vehicle will be re-matched in the verified vehicle group.

[0084] S302 establishes a dynamic search area centered on the starting point and searches the vehicle group for vehicles of the same price within a 500-meter radius of the starting point. If no vehicle is available, the search area is expanded to a 1.5-kilometer radius and the vehicle configuration requirements are lowered. If no matching vehicle is found, cross-region dispatch is triggered and the cost of vehicle allocation from surrounding service stations is calculated. Otherwise, the target vehicle is found.

[0085] Among the verified vehicle groups, priority will be given to finding vehicles that are within 500 meters of the user's starting point and belong to the grade / price range selected by the user; if a qualified vehicle is found, the next step of the dispatch process will be directly entered.

[0086] If no suitable vehicle is found in the first layer, the search range will be expanded to 1.5 kilometers. At the same time, downgrade matching (such as downgrading from comfort to economy) is allowed to improve the matching success rate. At this time, it is still necessary to ensure that the vehicle has sufficient remaining power and is reachable.

[0087] If the first two steps fail to successfully match a vehicle, the cross-regional dispatch mechanism will be activated; the system will automatically query several surrounding service stations (such as vehicle stops within 3km or 5km) to see whether there are available vehicles to calculate the dispatch cost by comprehensively considering the following factors: dispatch distance and estimated time, vehicle current battery level and round-trip energy consumption, expected user waiting time and satisfaction impact.

[0088] If there is a dispatch plan with a higher comprehensive score, the dispatch preparation stage will be entered; if there is no feasible plan, the user will be given a prompt "No vehicles are currently available" and will be advised to wait for a while or adjust travel parameters.

[0089] S303 issues a dispatch instruction to the target vehicle and generates an estimated arrival time.

[0090] The dispatch instruction is sent to the autonomous driving control center of the target vehicle through the Internet of Vehicles communication protocol (such as MQTT, HTTP long connection, etc.); the instruction content includes the user's starting point coordinates, vehicle navigation path, passenger boarding and disembarking status identification, etc. After receiving the instruction, the vehicle confirms its own status and sends back a dispatch response signal.

[0091] Based on factors such as the vehicle's current location, real-time road conditions, and traffic regulations, a path planning algorithm (such as A*, Dijkstra, etc.) is called to generate the optimal route; the time it takes for the vehicle to arrive at the departure point is calculated and an ETA is generated; the ETA information is synchronously returned to the user terminal and displayed on the App interface, allowing users to reasonably arrange waiting time.

[0092] During the dispatch process, the vehicle operating status and environmental changes are continuously monitored; in case of emergencies (such as temporary road closures, system restarts, etc.), the route is dynamically adjusted or the order is re-dispatched to ensure service continuity.

[0093] After receiving the dispatch instruction, the vehicle moves to the starting point and sends the user to the destination after confirming the user's identity through biometric recognition.

[0094] The specific steps include:

[0095] S501 obtains the dispatch instruction and sends the arrival information to the user after the target vehicle arrives at the designated boarding point according to the dispatch instruction;

[0096] The vehicle control system receives the mission instruction package issued by the dispatch center, which contains information such as the departure coordinates, boarding time window, and priority identification. After the instruction is successfully verified, the autonomous driving navigation system is activated, loading high-precision map data and real-time traffic status.

[0097] The vehicle generates an optimal route based on a path planning module, integrating multiple sensors such as lidar, cameras, and millimeter-wave radar to perceive the environment. When approaching the user's starting point, the system automatically adjusts speed and searches for a designated parking area (such as a roadside stop or designated platform pick-up and drop-off area).

[0098] After the vehicle has accurately stopped at the boarding point, a "vehicle has arrived" notification is sent to the user terminal through the on-board communication module, usually through App push, SMS or voice broadcast to inform the user.

[0099] At the same time, it can trigger the lighting prompts in the car (such as scrolling text on the top display) to help users identify the target vehicle

[0100] S502 continuously collects video data around the vehicle and searches for passengers. The vehicle camera captures the user's facial features and compares them with the order binding information. If the comparison is successful, the door corresponding to the target seat is unlocked;

[0101] This step aims to complete user identity verification through biometric authentication to enhance security and prevent impersonation. After the vehicle stops, it continuously monitors the surrounding environment and uses the on-board camera to identify and track pedestrians near the vehicle. Image recognition algorithms are used to determine whether there are people approaching and staying on the boarding side, and to preliminarily lock potential passengers. When the user approaches the vehicle, the high-definition camera in the specified direction automatically aligns with the user's facial area and collects facial features; the system compares the collected facial data with the user's real-name information bound to the current order; if the match is successful, the next step of the unlocking process is entered; if it fails, a re-identification opportunity is provided or a dynamic verification code is supported as a supplementary verification method. For multi-person sharing scenarios, the system can only unlock the doors corresponding to the user's reserved seats based on the seat allocation information in the order; the unlocking action can be achieved through the electric door lock mechanism, and accompanied by voice prompts or flashing lights to guide the user to get on the right board.

[0102] S503 dynamically plans the optimal path to the destination based on real-time traffic conditions;

[0103] After the user boards the vehicle, the system automatically reads the destination information from the order. It then uses the routing engine to recalculate the optimal route, combining the vehicle's current location, the coordinates of the destination point, historical driving records, and road speed limits and congestion conditions. During driving, the system continuously receives real-time traffic information (such as construction sites, traffic accidents, and traffic light cycles) from the traffic management platform or cloud server. It automatically adjusts the route based on the latest road conditions to avoid delays and detours. Route change information is also synchronously fed back to the user terminal, enhancing transparency and trust.

[0104] After arriving at the destination, S504 scans the vehicle through sensors to confirm that all passengers and belongings have left, and automatically generates an electronic bill and deducts the fee;

[0105] After the vehicle arrives at the destination, the disembarkation process is initiated. The system uses infrared sensors, seat pressure sensors, cameras, etc. in the vehicle to comprehensively determine whether the passengers have completely disembarked. At the same time, it uses image recognition technology to identify whether there are any left-behind items (such as backpacks, umbrellas, etc.). If so, the user is prompted to retrieve them. After the user confirms disembarking, he or she clicks the "Completed" button on the App or confirms completion through voice interaction. The on-board system automatically uploads the departure status to the backend and prepares for settlement. The system uses the billing engine to generate accurate electronic bills based on factors such as actual mileage, vehicle model, time period, and coupon usage. It supports multiple payment methods (such as Alipay, WeChat, bank cards, etc.) to complete automatic deductions. After the deduction is successful, the bill details are pushed to the user end and can be exported as an invoice or travel voucher.

[0106] S505 sends a task completion signal to the dispatch center and puts the vehicle into an idle state.

[0107] After completing a delivery mission, the vehicle uploads a "mission completed" signal to the dispatch center via the IoV communication module. This signal contains: the start and end times of the trip; the actual mileage; the customer satisfaction rating (if any); and the vehicle's status (battery level, fault status, etc.). Once the report is complete, the vehicle automatically switches to an "idle and dispatchable" state, awaiting the next dispatch instruction. It can choose to remain at its current location awaiting new missions, or automatically proceed to the nearest service station or demand hotspot based on optimization strategies. After each mission, the system automatically assesses the vehicle's remaining battery life, tire wear, brake system health, and other factors. If the conditions for the next mission are not met, the system will direct the vehicle to a charging station or maintenance point for processing.

[0108] S104 After the vehicle reaches the target point, a potential parking spot with vehicle usage demand is allocated to the vehicle based on historical vehicle usage data, and the vehicle is moved to the potential parking spot with vehicle usage demand.

[0109] The specific steps include:

[0110] S601: When the remaining power of the vehicle is lower than the preset value, the nearest available charging station is selected for charging; when the remaining power of the vehicle is higher than the preset value, the standby point selection is entered;

[0111] The vehicle control system continuously monitors the current charge percentage provided by the battery management system (BMS). The system sets a reasonable charge threshold (e.g., 20%) to determine whether immediate charging is necessary. If the current charge falls below the threshold, an automatic charging station search is triggered. The system retrieves information from a cloud database about nearby charging stations with sufficient power and in operation. It calculates the optimal route using a shortest path algorithm (e.g., Dijkstra, A*, etc.) and sends a notification to the user that the vehicle is about to proceed to a charging station (if there are subsequent tasks scheduled).

[0112] If the battery is sufficiently charged and the vehicle does not need to be recharged immediately, the system will automatically enter the standby point recommendation and navigation process.

[0113] S602 generates a recommended parking spot based on historical vehicle usage data and current parking time and area when selecting a standby point;

[0114] The specific steps include:

[0115] S701 retrieves the order initiation location and average waiting time for the same period of the past 30 days in the target area from the order database;

[0116] Define a certain urban area (e.g., a 3-kilometer radius) with the vehicle's current location as the center. Filter order records from the order database for the past 30 days within the same or similar time periods (e.g., 6:00 PM to 7:00 PM on weekdays).

[0117] The extracted fields include the geographical coordinates of the order initiation (latitude and longitude) and the average response time (the time from order placement to vehicle arrival).

[0118] S702 divides the geographic space into multiple small areas using a gridded city map and extracts temporal features to generate training and validation sets;

[0119] Use GIS tools to divide the target city map into several regular grids (such as 500m×500m square areas); each small grid represents a candidate standby area and is assigned a unique identifier (Grid ID); statistics are collected on indicators such as order density and average waiting time for each grid within a specific time period.

[0120] In the time dimension, features such as hour, day of the week, whether it is a holiday, and whether it is during the morning and evening rush hour are extracted.

[0121] Divide historical data into a training set (80%) and a validation set (20%) in chronological order;

[0122] S703 uses the training set data to train a polynomial regression model;

[0123] A polynomial regression model is used, suitable for capturing nonlinear trends (such as order fluctuations over different time periods). The input variables are temporal features and spatial grid numbers, and the output variable is the demand intensity (such as the number of orders or waiting time) for orders in the target area. An appropriate order (such as 2nd or 3rd) is set to avoid overfitting or underfitting. A polynomial expansion (such as constructing quadratic terms) is performed on all samples in the training set. The least squares method is used to estimate the model parameters for training.

[0124] S704 evaluates and optimizes the polynomial regression model on the test set;

[0125] The mean square error (MSE), mean absolute error (MAE) and R 2 The coefficient of determination evaluates the performance of the model.

[0126] S705 inputs the current parking time and area according to the polynomial regression model to generate a recommended parking spot.

[0127] The current parking time and location are converted into a feature format consistent with the training data. This is then fed into a trained polynomial regression model, which outputs a "demand popularity score" for each candidate grid. The top N grids are ranked by score and selected as recommended parking spots. These recommended spots include latitude and longitude, estimated time of arrival, and expected order acceptance success rate.

[0128] S603 enables the vehicle to drive along the planned route based on the recommended parking spot, communicate with the parking lot gate, and obtain parking spaces with priority; it parks accurately in the parking space through the surround-view camera and ultrasonic sensor, and feeds back the final position to the server.

[0129] Based on the recommended parking spots, the path planning engine is invoked to generate the optimal driving path from the current location to the destination. This path information is then sent to the autonomous driving controller, and the vehicle begins autonomous driving. Upon arrival at the parking lot entrance, the vehicle interacts with the barrier system via a V2I communication protocol (such as DSRC or C-V2X). The vehicle's identity and permissions are verified, and the barrier is automatically raised for passage. The parking lot management system interface is then called to request a reserved parking space or reserve a fixed space in advance. When the vehicle approaches a designated parking space, automated parking mode is activated. Surround-view cameras, ultrasonic radar, and millimeter-wave radar are used to sense the surrounding environment. The vehicle automatically identifies the boundaries of available parking spaces and controls the steering, throttle, and brakes to complete parallel or perpendicular parking. Once parking is complete, the parking brake automatically engages, and the screen displays "Parking Successful." After parking, the vehicle's final location coordinates are uploaded to the dispatch server, and the vehicle's status is updated to "Idle Standby." The map service also updates the vehicle's dispatchable status.

[0130] If there are no available parking spaces in the recommended area, the system will recommend an alternative location.

[0131] Second embodiment

[0132] The present invention also provides a ride-sharing system for unmanned vehicles, which is applied to the ride-sharing method for unmanned vehicles of the first embodiment.

[0133] In this embodiment, the user terminal module includes a mobile terminal device used by the user (such as a smartphone app, a car-mounted tablet, etc.); provides a user input interface for entering the car price, departure point and destination point information; and supports multiple interaction methods such as map selection, voice input, and historical address memory;

[0134] The cloud dispatch center is the command center of the entire system, deployed on a high-performance server cluster or cloud computing platform. It is responsible for receiving and interpreting vehicle request information sent by user terminals. It makes vehicle matching and dispatch decisions based on real-time data (vehicle location, battery level, order status, historical behavior, etc.). It also implements advanced dispatch logic such as dynamic search area construction, cross-region allocation evaluation, and multi-strategy recommended parking point generation.

[0135] The unmanned vehicle control module is a core control system installed in each unmanned vehicle, including an automatic driving controller, a perception sensor group (lidar, camera, millimeter-wave radar, etc.), and an actuator (steering, braking, and drive systems). It receives dispatch instructions from the cloud dispatch center and automatically completes the driving task from the current location to the boarding point and then to the destination point. After receiving the user identity verification signal, it unlocks the door of the corresponding seat. After arriving at the destination, it determines whether the passenger has got off the vehicle based on the sensors in the vehicle, and triggers the bill generation and upload of the task completion signal. After the task is completed, it enters the charging station or goes to a potential high-demand area to stand by according to the system recommendation.

[0136] The communication network module supports high-speed communication between the vehicle and the cloud server (such as 5G / 4G, Wi-Fi, MQTT, HTTP long connection, etc.);

[0137] The security management and identity authentication module is responsible for user identity verification, including multimodal biometric methods such as face recognition, voice recognition, and fingerprint verification; all user facial feature data is encrypted and compared on the local device to ensure privacy and security; the vehicle startup permission and door unlocking mechanism are controlled, and passengers are allowed to board the vehicle only after successful identity authentication; financial security is achieved during the payment process to prevent malicious swiping, account fraud and other risky behaviors; multiple backup and fault recovery mechanisms are set up to ensure that the system can maintain basic security under abnormal circumstances.

[0138] This enables unmanned operation of the entire process, from order initiation, vehicle dispatch, identity verification, autonomous driving, trip settlement, and resource reallocation. Combining historical data with machine learning models accurately predicts future hotspots for vehicle demand, improving vehicle utilization. The integration of multiple technologies, including facial and voice recognition, enhances system security and user experience.

[0139] The above disclosure is only a preferred embodiment of the present invention, and certainly cannot be used to limit the scope of the rights of the present invention. Ordinary technicians in this field can understand that all or part of the processes of the above embodiment and equivalent changes made in accordance with the claims of the present invention are still within the scope of the invention.

Claims

1. A method for sharing a ride in an unmanned vehicle, characterized in that: include: Obtain the car price, departure point, and destination information entered by the user and generate a car request; Based on the vehicle request information, the nearest available vehicle to the starting point is matched and a dispatch instruction is generated; After receiving the dispatch instruction, the vehicle moves to the starting point and sends the user to the destination after confirming the user's identity through biometric recognition; After the vehicle reaches the target point, a potential parking spot with vehicle usage demand is allocated to the vehicle based on historical vehicle usage data, and the vehicle is moved to the potential parking spot with vehicle usage demand.

2. The method for sharing a ride in an unmanned vehicle according to claim 1, wherein: The specific steps of obtaining the car price, starting point and destination point information input by the user and generating the car request information include: Users can select the corresponding vehicle price range based on the dynamic price matrix interface and the preset price range divided by vehicle type and grade; Call the terminal GPS module to obtain the current location as the default starting point; After receiving the target point keyword input by the user, the starting point and destination point are adjusted, and the user request information is generated in combination with the car price.

3. The method for sharing a ride in an unmanned vehicle according to claim 2, wherein: The specific steps of matching the available vehicle closest to the starting point based on the vehicle request information and generating a dispatch instruction include: Filter the corresponding vehicle group in the vehicle database based on the price information in the user's order, and verify the remaining mileage of the vehicles in the vehicle group; A dynamic search area centered on the starting point is established, searching the vehicle group for vehicles of the same price within a 500-meter radius of the starting point. If no vehicle is available, the radius is expanded to 1.5 kilometers and the vehicle configuration requirements are lowered. If no matching vehicle is found, cross-regional dispatch is triggered, calculating the vehicle allocation cost of surrounding service stations. Otherwise, the target vehicle is found. Issue dispatch instructions to the target vehicle and generate an estimated arrival time.

4. The method for sharing a ride in an unmanned vehicle according to claim 3, wherein: The specific steps of screening the corresponding vehicle group in the vehicle database according to the price information in the user order and verifying the remaining mileage of the vehicles in the vehicle group include: Obtaining the availability of each vehicle in the vehicle group, including whether the vehicle is booked and whether it is in maintenance status; After the vehicle availability is passed, checking whether its remaining range meets the user's travel requirements, wherein the travel requirements are derived based on the starting point and the destination point; The vehicles that pass the travel requirements are grouped together into a verified vehicle group.

5. The method for sharing a ride in an unmanned vehicle according to claim 4, wherein: After issuing the dispatch instruction to the target vehicle and generating the estimated arrival time, the step further includes: re-matching the vehicle if a vehicle failure occurs or the estimated arrival time exceeds a preset value.

6. The method for sharing a ride in an unmanned vehicle according to claim 5, wherein: After receiving the dispatch instruction, the vehicle moves to the starting point and confirms the user's identity through biometric recognition, and then sends the user to the destination point. The specific steps include: Obtain dispatch instructions and send arrival information to the user after the target vehicle arrives at the designated boarding point according to the dispatch instructions; Continuously collect video data around the vehicle and search for passengers. The onboard camera captures the user's facial features and compares them with the order-bound information. If the comparison is successful, the door corresponding to the target seat will be unlocked; Dynamically plan the optimal path to the destination based on real-time traffic conditions; After arriving at the destination, the in-car sensor scans to confirm that all passengers and belongings have left, and automatically generates an electronic bill and deducts the fee; Send a mission completion signal to the dispatch center and put the vehicle into idle state.

7. The method for sharing a ride in an unmanned vehicle according to claim 6, wherein: After the vehicle reaches the target point, the specific steps of allocating a potential parking spot for the vehicle based on historical vehicle usage data and moving the vehicle to the potential parking spot include: When the remaining battery power of the vehicle is lower than the preset value, the nearest available charging station is selected for charging; when the remaining battery power of the vehicle is higher than the preset value, the vehicle enters the standby point selection; When selecting a standby point, a recommended parking spot is generated based on historical vehicle usage data combined with current parking time and area; Based on the recommended parking spots, the vehicle drives along the planned route, communicates with the parking lot gate, and obtains parking spaces with priority; it parks accurately in the parking space through surround-view cameras and ultrasonic sensors, and feeds back the final position to the server.

8. The method for sharing a ride in an unmanned vehicle according to claim 7, wherein: The specific steps of generating a recommended parking spot based on historical vehicle usage data and current parking time and area when selecting a standby point include: Retrieve the order origination location and average waiting time for the same period in the target area over the past 30 days from the order database; The geographic space is divided into multiple small areas through gridded city maps and temporal features are extracted to generate training and validation sets. Train a polynomial regression model using the training set data; Evaluate and optimize the polynomial regression model on the test set; The current parking time and area are input to generate recommended parking spots based on a polynomial regression model.

9. A shared ride system for unmanned vehicles, characterized in that: A method for sharing a ride in an unmanned vehicle as claimed in any one of claims 1 to 8.

10. The unmanned vehicle sharing system according to claim 9, characterized in that: The unmanned vehicle shared ride system includes a user terminal module, a cloud dispatch center, an unmanned vehicle control module, a communication network module, and a security management and identity authentication module; The user terminal module provides a user input interface for inputting the price of the vehicle, the departure point and the destination point information; The cloud dispatch center receives and analyzes the vehicle request information sent by the user terminal, and makes vehicle matching and dispatch decisions based on real-time data; The unmanned vehicle control module receives dispatch instructions from the cloud dispatch center and automatically completes the driving task from the current location to the boarding point and then to the destination point; after receiving the user identity verification signal, it unlocks the door of the corresponding seat; Upon arrival at the destination, the vehicle determines whether the passenger has disembarked based on in-vehicle sensors, triggering bill generation and uploading of a mission completion signal. Upon completion of the mission, the vehicle proceeds to a charging station or to a potential high-demand area based on system recommendations. The communication network module is used for communication; The security management and identity verification module is used for user identity verification.

Images