Cooperative recovery method and system based on road parking unmanned inspection robot

Abstract

The application discloses a cooperative recovery method and system based on a road parking unmanned inspection robot, relates to the technical field of intelligent transportation, and comprises the following steps: performing inspection and identification through the road parking unmanned inspection robot, obtaining a plurality of vehicle license plate information for response query, and generating a fee arrears query result; when there is a historical fee arrears order, performing ETC payment qualification determination on the vehicle; screening the historical fee arrears order according to the ETC qualification determination result; performing automatic deduction on an ETC deductible order set to generate order payment data; analyzing the ETC non-deductible order set to trigger the road parking unmanned inspection robot to perform a cooperative recovery task. The technical problems that the road parking fee arrears recovery efficiency is low, the coverage is limited, and it is difficult to automatically recover different payment conditions of vehicles in the prior art are solved, and the technical effects of improving the fee arrears order processing efficiency, the recovery success rate and the intelligent level of parking management are achieved.

Description

Technical Field

[0001] This invention relates to the field of intelligent transportation technology, specifically to a collaborative tracking and recovery method and system based on unmanned road parking inspection robots. Background Technology

[0002] On-street parking, as a crucial component of the urban parking system, plays a key role in alleviating the parking supply-demand imbalance. However, on-street parking management has long faced challenges such as high operating costs, difficulty in supervision, and frequent instances of parking fee evasion. Particularly in the collection of parking fees, traditional manual patrol and collection methods suffer from inefficiency, limited coverage, and difficulty in obtaining evidence, leading to widespread parking fee arrears and severely impacting the standardization of urban parking management. The application of unmanned on-street parking inspection robots offers a new solution for parking management. These robots, through image recognition and vehicle detection technologies, can automatically identify vehicle information and complete parking fee calculations. However, their main focus is on automating the parking process; effective technical means are still lacking for the crucial link of collecting overdue fees. Currently, overdue fee collection mainly relies on manual patrols, posting overdue notices, SMS reminders, and telephone calls, with unsatisfactory collection efficiency and success rates, making it difficult to form an effective closed-loop management system. Furthermore, existing technologies have not yet provided effective solutions for how to effectively combine ETC payment with unmanned roadside parking inspection robots to achieve intelligent screening and automatic deduction of overdue orders, as well as for how to conduct collaborative collection of payments for vehicles that do not meet the conditions for ETC payment.

[0003] Therefore, current technologies suffer from low efficiency in collecting overdue parking fees, limited coverage, and difficulty in automating collection for vehicles with different payment conditions. Summary of the Invention

[0004] This application provides a collaborative collection method and system based on unmanned road parking inspection robots, which solves the technical problems of low efficiency, limited coverage, and difficulty in automating collection of overdue parking fees in the existing technology. It achieves the technical effect of improving the efficiency of overdue order processing, collection success rate, and the level of intelligent parking management.

[0005] This application provides a collaborative collection method based on unmanned road parking inspection robots. The method includes: using the unmanned road parking inspection robot to inspect and identify multiple vehicle license plate information, responding to queries, and generating overdue payment query results; when the overdue payment query results contain historical overdue orders, calling the traffic management system interface to determine the vehicle's ETC payment eligibility, and obtaining the ETC eligibility determination result; filtering the historical overdue orders according to the ETC eligibility determination result to determine the ETC deductible order set and the ETC non-deductible order set; using the unmanned road parking inspection robot to automatically deduct fees from the ETC deductible order set, obtaining the ETC deduction result, updating the order payment status, and generating order payment data; and analyzing the ETC non-deductible order set based on the order payment data to trigger the unmanned road parking inspection robot to execute a collaborative collection task.

[0006] In a possible implementation, historical arrears orders are filtered according to the ETC eligibility determination results to determine the ETC-deductible order set and the ETC-non-deductible order set. The method includes: parsing the ETC eligibility determination results to obtain ETC payment activation parameters and ETC payment non-activation parameters; when the ETC eligibility determination result is the ETC payment activation parameter, extracting the dynamic deduction rules of the ETC system; traversing the historical arrears orders for feature analysis to generate order arrears time features and order arrears amount features; mapping the order arrears time features to the dynamic deduction rules for timeliness comparison to generate a time-series comparison result; mapping the order arrears amount features to the dynamic deduction rules for amount comparison to generate a numerical comparison result; filtering the historical arrears orders based on the time-series comparison result and the numerical comparison result to determine the ETC-deductible order set and the ETC-non-deductible order set.

[0007] In a possible implementation, the historical overdue orders are filtered based on the time-series comparison results and the numerical comparison results to determine the ETC deductible order set and the ETC non-deductible order set. The method includes: extracting order overdue time features that meet the time-sensitivity threshold based on the time-series comparison results to generate a first overdue time feature; extracting order overdue amount features that meet the amount threshold based on the numerical comparison results to generate a first overdue amount feature; filtering historical overdue orders based on the first overdue time feature and the first overdue amount feature to determine the ETC deductible order set; extracting order overdue time features that do not meet the time-sensitivity threshold based on the time-series comparison results to generate a second overdue time feature; extracting order overdue amount features that do not meet the amount threshold based on the numerical comparison results to generate a second overdue amount feature; and filtering historical overdue orders based on the second overdue time feature and the second overdue amount feature to determine the ETC non-deductible order set.

[0008] In a possible implementation, the unmanned road parking inspection robot performs automatic deductions on the ETC-deductible order set, obtains the ETC deduction result, updates the order payment status, and generates order payment data. The method includes: encoding the ETC-deductible order set to construct a deduction instruction sequence; sending this sequence to the unmanned road parking inspection robot via a vehicle-road cooperative communication link, driving the robot to establish a near-field communication data field with the target vehicle; performing batch deductions based on the near-field communication data field according to the deduction instruction sequence, generating a deduction receipt result, which includes a successful deduction order identifier and a failed deduction order identifier; updating the order payment status according to the successful deduction order identifier, generating order payment status information; updating the order payment status according to the failed deduction order identifier, generating pending collaborative collection status information; and adding the paid order status information and the pending collaborative collection status information to the order payment data.

[0009] In a possible implementation, the ETC non-deductible order set is analyzed based on the order payment data to trigger the unmanned road parking inspection robot to execute a collaborative collection task. The method includes: acquiring the ETC non-deductible order set and merging it with the pending collaborative collection status to generate a pending collection order feature set; sending the pending collection order feature set to the unmanned road parking inspection robot as a task parameter for executing collaborative collection; extracting order quantity parameters and outstanding amount parameters based on the pending collection order feature set, dynamically activating and combining the first collaborative collection channel and the second collaborative collection channel to trigger the unmanned road parking inspection robot to execute the collaborative collection task.

[0010] In a possible implementation, the first and second collaborative tracing channels are dynamically activated in combination to trigger the unmanned road parking inspection robot to perform a collaborative tracing task. The method includes: the unmanned road parking inspection robot performing multi-dimensional perception of the vehicle's surroundings to obtain a multimodal perception dataset, which includes image perception data and audio perception data; performing real-time inference based on the image perception data, verifying the image inference results according to the audio perception data, and generating vehicle presence confidence and vehicle location information; when the vehicle presence confidence exceeds a preset confidence threshold, generating a first activation signal to activate the first collaborative tracing channel; and generating a second activation signal based on the vehicle location information to activate the second collaborative tracing channel.

[0011] In a possible implementation, after the unmanned road parking inspection robot is triggered to perform a collaborative collection task, the method includes: triggering a second query of the real-time overdue status to obtain the updated data of the number of overdue orders for the target vehicle; if the updated data of the number of overdue orders is zero, generating a collection termination signal to end the collection process; if the updated data of the number of overdue orders is greater than zero, generating a jump signal, and analyzing and verifying the remaining overdue orders based on the jump signal and the order payment data.

[0012] This application also provides a collaborative collection system based on a road parking unmanned inspection robot. The system includes: an inspection and identification module, used to inspect and identify multiple vehicle license plate information through the road parking unmanned inspection robot, respond to queries, and generate overdue payment query results; an eligibility determination module, used to call the traffic management system interface to determine the vehicle's ETC payment eligibility when the overdue payment query results contain historical overdue orders, and obtain ETC eligibility determination results; an order filtering module, used to filter historical overdue orders according to the ETC eligibility determination results, and determine the ETC deductible order set and the ETC non-deductible order set; an order update module, used to automatically deduct fees from the ETC deductible order set through the road parking unmanned inspection robot, obtain ETC deduction results, update the order payment status, and generate order payment data; and a collaborative collection module, used to analyze the ETC non-deductible order set based on the order payment data, and trigger the road parking unmanned inspection robot to execute a collaborative collection task.

[0013] This application proposes a collaborative collection method and system based on unmanned road parking inspection robots. The unmanned road parking inspection robot performs inspections and identifications, obtaining license plate information for multiple vehicles and generating overdue payment query results. When historical overdue orders exist, the system determines the vehicle's eligibility for ETC payment. Based on the ETC eligibility determination results, historical overdue orders are filtered. Automatic deduction is performed on the ETC-deductible order set, generating order payment data. The system analyzes the ETC-non-deductible order set, triggering the unmanned road parking inspection robot to perform collaborative collection tasks. This solves the technical problems of low efficiency, limited coverage, and difficulty in automating collection for vehicles with different payment conditions in existing road parking overdue payment collection technologies. It achieves the technical effects of improving the efficiency of overdue order processing, collection success rate, and the level of intelligent parking management. Attached Figure Description

[0014] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments of this disclosure will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0015] Figure 1 This is a schematic diagram of the collaborative recovery method based on an unmanned road parking inspection robot provided in an embodiment of this application.

[0016] Figure 2 This is a schematic diagram of the collaborative recovery system based on an unmanned road parking inspection robot provided in an embodiment of this application.

[0017] Explanation of reference numerals in the attached diagram: Inspection and identification module 10, qualification determination module 20, order screening module 30, order update module 40, collaborative recovery module 50. Detailed Implementation

[0018] To further illustrate the technical means and effects adopted by the present invention in order to achieve the intended purpose, the following detailed description is provided in conjunction with the accompanying drawings and preferred embodiments, based on the specific implementation methods, structures, features and effects of the present invention.

[0019] This application provides a collaborative tracking method based on unmanned road parking inspection robots, such as... Figure 1 As shown, the method includes:

[0020] Step S100: The unmanned road parking inspection robot performs inspection and identification, obtains license plate information of multiple vehicles, responds to queries, and generates overdue payment query results.

[0021] Preferably, the unmanned roadside parking inspection robot autonomously travels along a pre-set patrol route. During its journey, it acquires real-time images of parked vehicles on both sides of the road using its onboard image acquisition equipment. It then analyzes and processes these images using image recognition to locate and extract the license plate area of ​​each vehicle. Optical character recognition (OCR) technology converts the license plate image into readable text data, thus obtaining the vehicle's license plate information. The robot then uses its built-in wireless communication unit to send the license plate information as a query keyword to the server of the backend parking management system in real time. Upon receiving the query request, the backend server searches the parking management database for historical parking records associated with the license plate and verifies whether the vehicle has any outstanding parking fees. After completing the search, the robot transmits the query results, such as whether the vehicle has no outstanding fees, has one outstanding fee, or has multiple outstanding fees, back to the unmanned inspection robot via the communication network. Upon receiving this data, the robot completes the outstanding fee query for the vehicle and generates the final outstanding fee query result.

[0022] Step S200: When the overdue payment query result contains historical overdue payment orders, the traffic management system interface is called to determine the vehicle's eligibility for ETC payment and obtain the ETC eligibility determination result.

[0023] Preferably, the unmanned roadside parking inspection robot receives and parses the overdue payment query results returned by the backend server. If the results show that the vehicle has at least one unpaid parking order, the system automatically triggers the subsequent ETC payment eligibility determination process. Through a dedicated data link, it initiates a communication request to the external service interface (i.e., the traffic management system interface) provided by the traffic management department or ETC operator. This request carries the target vehicle's license plate information to query the vehicle's ETC function status. Upon receiving the request, the traffic management system interface searches the ETC vehicle registration database to verify whether the vehicle meets the relevant conditions for ETC payment. Verification includes, but is not limited to, whether the vehicle has installed and activated an ETC on-board unit, whether the ETC account status is normal (e.g., not reported lost, not cancelled, not frozen), whether the account balance or credit limit is sufficient, and whether the vehicle is on the ETC payment blacklist. This generates standardized judgment result data, clearly indicating the vehicle's ETC payment eligibility status in the form of parameters, such as "eligible for ETC payment" or "not eligible for ETC payment," which is then transmitted back through the communication network to obtain the ETC eligibility determination result.

[0024] Step S300: Filter historical overdue orders according to the ETC eligibility determination results to determine the ETC deductible order set and the ETC non-deductible order set.

[0025] Step S300 further includes parsing the ETC eligibility determination result to obtain ETC payment activation parameters and ETC payment non-activation parameters; when the ETC eligibility determination result is the ETC payment activation parameters, extracting the dynamic deduction rules of the ETC system; traversing the historical overdue orders for feature analysis to generate order overdue time features and order overdue amount features; mapping the order overdue time features to the dynamic deduction rules for timeliness comparison to generate time series comparison results; mapping the order overdue amount features to the dynamic deduction rules for amount comparison to generate numerical comparison results; and filtering the historical overdue orders based on the time series comparison results and the numerical comparison results to determine the ETC deductible order set and the ETC non-deductible order set.

[0026] Preferably, the ETC eligibility determination result is parsed to obtain ETC payment activation parameters, indicating that the vehicle is eligible for ETC payment, and ETC payment not activated parameters, indicating that the vehicle is not eligible for ETC payment. The vehicle's eligibility category is identified based on the parsing result. When the parsing result shows that the ETC eligibility determination result is ETC payment activation parameters, a rule retrieval request is sent to the ETC settlement platform to extract the currently effective dynamic deduction rules. These rules contain restrictions on the ETC system's automatic deduction of parking orders, such as allowing deductions only for orders with outstanding fees within the last 30 days, a maximum deduction amount per order not exceeding 500 yuan, and a daily cumulative deduction limit of 1000 yuan, among other time-sensitive and monetary restrictions. All historical outstanding order data for the vehicle is traversed, and feature extraction and analysis are performed on each order to generate order outstanding fee time features, recording the number of days between the order's creation date and the current date, as well as the order's outstanding fee amount features, and recording the unpaid amount of the order. The system takes two steps: First, it takes the arrears time of each historical overdue order as input, maps it to the time limits in the ETC dynamic deduction rules for comparison and calculation, and generates a time-series comparison result. This result indicates whether the arrears time of each order is within the time limit allowed by ETC. Second, it takes the arrears amount of each historical overdue order as input, maps it to the amount limits in the ETC dynamic deduction rules for comparison and calculation, and generates a numerical comparison result. This result indicates whether the arrears amount of each order is within the amount allowed by ETC. Finally, it combines the time-series comparison result and the numerical comparison result for each order for filtering and judgment. Orders with a time-series comparison result of "within the time limit" and a numerical comparison result of "within the amount limit" are classified as ETC deductible orders. Orders with a time-series comparison result of "outside the time limit" or a numerical comparison result of "outside the amount limit" are classified as ETC non-deductible orders.

[0027] Furthermore, step S300 also includes: extracting order arrears time features that meet the timeliness threshold based on the time-series comparison results to generate a first arrears time feature; extracting order arrears amount features that meet the amount threshold based on the numerical comparison results to generate a first arrears amount feature; filtering historical arrears orders according to the first arrears time feature and the first arrears amount feature to determine the ETC deductible order set; extracting order arrears time features that do not meet the timeliness threshold based on the time-series comparison results to generate a second arrears time feature; extracting order arrears amount features that do not meet the amount threshold based on the numerical comparison results to generate a second arrears amount feature; and filtering historical arrears orders according to the second arrears time feature and the second arrears amount feature to determine the ETC non-deductible order set.

[0028] Preferably, the generated time-series comparison results are traversed and retrieved, and orders marked as "meeting the time-sensitive threshold" are selected from the time-series comparison results. The order arrears time feature data corresponding to these orders is extracted and summarized as the first arrears time feature. Similarly, the generated numerical comparison results are traversed and retrieved, and orders marked as "meeting the amount threshold" are selected from the numerical comparison results. The order arrears amount feature data corresponding to these orders is extracted and summarized as the first arrears amount feature. The first arrears time feature and the first arrears amount feature are intersected and the orders that simultaneously possess the first arrears time feature and the first arrears amount feature, that is, the orders that meet both the time-sensitive threshold and the amount threshold, are integrated and determined as the ETC deductible order set. The generated time-series comparison results are traversed and retrieved. Orders marked as "not meeting the timeliness threshold" are selected from the time-series comparison results. The order arrears time feature data corresponding to these orders is extracted and summarized as the second arrears time feature. Similarly, the generated numerical comparison results are traversed and retrieved. Orders marked as "not meeting the amount threshold" are selected from the numerical comparison results. The order arrears amount feature data corresponding to these orders is extracted and summarized as the second arrears amount feature. The second arrears time feature and the second arrears amount feature are merged and processed to select orders that have the second arrears time feature or the second arrears amount feature, that is, orders that do not meet the timeliness threshold or do not meet the amount threshold, and are integrated to determine the ETC non-deductible order set.

[0029] Step S400: The unmanned road parking inspection robot performs automatic deduction on the ETC deductible order set, obtains the ETC deduction result, updates the order payment status, and generates order payment data.

[0030] Step S400 further includes: encoding the ETC deductible order set to construct a deduction instruction sequence; sending the sequence to the unmanned road parking inspection robot via the vehicle-road cooperative communication link to drive the unmanned road parking inspection robot to establish a near-field communication data field with the target vehicle; performing batch deductions based on the near-field communication data field according to the deduction instruction sequence to generate a deduction receipt result, the deduction receipt result including a successful deduction order identifier and a failed deduction order identifier; updating the order payment status according to the successful deduction order identifier to generate order payment status information; updating the order payment status according to the failed deduction order identifier to generate pending collaborative collection status information; and adding the order payment status information and the pending collaborative collection status information to the order payment data.

[0031] Preferably, the determined set of ETC deductible orders is used as the processing object. Each order in the set is formatted and encoded. The order number, license plate number, deduction amount, and deduction timestamp of each order are encapsulated according to the protocol specifications of the ETC deduction interface to generate a standard deduction instruction. All generated deduction instructions are arranged in a preset order to form a complete deduction instruction sequence. The constructed deduction instruction sequence is sent to the unmanned road parking inspection robot through the vehicle-road cooperative communication link. The vehicle-road cooperative communication link refers to the communication channel established based on dedicated short-range communication technology or cellular vehicle networking technology for data exchange between the back-end system and the inspection robot. After receiving the deduction instruction sequence, the inspection robot drives to the location of the target vehicle. The robot transmits communication signals near the vehicle through the on-board dedicated short-range communication device to establish a point-to-point near-field wireless communication connection with the ETC on-board unit installed in the target vehicle. The dedicated communication area established between the robot and the vehicle is the near-field communication data field.

[0032] Preferably, the inspection robot sends deduction instructions to the target vehicle's ETC on-board unit one by one according to the order of the instructions in the deduction instruction sequence. After receiving the instructions, the ETC on-board unit completes the fund deduction operation through interaction with the back-end ETC settlement system. After each instruction deduction operation is completed, the ETC settlement system returns the corresponding deduction result. The inspection robot receives and records the execution result of each instruction. After all instructions have been executed, the robot summarizes all results to generate deduction receipt result data, including a successful deduction order identifier, recording orders with successful deduction operations, and a failed deduction order identifier, recording orders that failed to deduct due to insufficient account balance, communication failure, etc.; receiving The inspection robot returns the deduction receipt data and parses it. For each order marked as a successful deduction in the deduction receipt, the payment status field of the order is modified to "paid" in the parking management database, generating the corresponding order payment status information. For each order marked as a failed deduction in the deduction receipt, the payment status field of the order is modified to "pending collaborative collection", generating the corresponding pending collaborative collection status information. Finally, all order payment status information and all pending collaborative collection status information are associated according to the order number to generate complete order payment data and record all processing results of this ETC automatic deduction.

[0033] Step S500: Analyze the ETC non-deductible order set based on the order payment data, and trigger the unmanned road parking inspection robot to perform a collaborative collection task.

[0034] Step S500 further includes: acquiring the set of ETC non-deductible orders and merging them with the pending collaborative collection status to generate a pending collection order feature set; sending the pending collection order feature set to the road parking unmanned inspection robot as a task parameter for performing collaborative collection; extracting order quantity parameters and arrears amount parameters based on the pending collection order feature set, dynamically activating the first collaborative collection channel and the second collaborative collection channel and combining them to trigger the road parking unmanned inspection robot to perform the collaborative collection task.

[0035] Preferably, the ETC non-deductible order set includes orders that cannot be automatically deducted via ETC due to exceeding the time limit or amount limit. The pending collection status includes orders that failed to be deducted via ETC automatic deduction due to insufficient balance, communication failure, or other reasons. These are merged to form a complete pending collection order set. Features are extracted from each order, including license plate number, overdue order number, overdue time, overdue amount, vehicle brand and color, etc., generating a pending collection order feature set. This feature set is then sent to the unmanned road parking inspection robot via wireless communication network as task parameters for the robot to perform subsequent collection tasks. After receiving the data, the robot stores it in its local memory and then proceeds with the collection process. The system performs statistical analysis on the order feature set to extract the order quantity parameter (total number of orders to be recovered) and the outstanding amount parameter (total outstanding amount of all orders to be recovered). Based on the values ​​of these two parameters, it dynamically selects and activates different collaborative recovery channels. The first and second collaborative recovery channels represent two different types of recovery execution methods. For example, the first channel may correspond to an automatic voice reminder function, while the second channel may correspond to video evidence collection and human intervention. Finally, based on the order quantity and outstanding amount, it decides whether to activate one channel alone or both channels simultaneously. After the channel is activated, a trigger signal is sent to the inspection robot. Upon receiving the trigger signal, the robot begins to execute the specific collaborative recovery task.

[0036] Furthermore, step S500 also includes: performing multi-dimensional perception of the vehicle's surroundings based on the road parking unmanned inspection robot to obtain a multi-modal perception dataset, the multi-modal perception dataset including image perception data and audio perception data; performing real-time inference based on the image perception data, verifying the image inference results according to the audio perception data, and generating vehicle presence confidence and vehicle location information; when the vehicle presence confidence exceeds a preset confidence threshold, generating a first activation signal to activate the first collaborative tracking channel; and generating a second activation signal based on the vehicle location information to activate the second collaborative tracking channel.

[0037] Preferably, after the unmanned roadside parking inspection robot moves to the vicinity of the target vehicle, it activates its onboard multi-dimensional perception sensors to collect data. The robot uses its onboard high-definition camera to collect real-time video streams or image frames around the vehicle, generating image perception data. Simultaneously, the robot uses its onboard microphone array or microphone to collect ambient sounds around the vehicle, including vehicle engine sounds, door opening and closing sounds, and conversations, generating audio perception data. Together, these constitute a multimodal perception dataset. Then, real-time inference calculations are performed on the collected image perception data, including using object detection to identify and locate vehicles in the images, confirming the presence of the target vehicle within the image frame. The image is analyzed to determine the vehicle's location and its pixel coordinates within the image. The image reasoning results are then cross-validated with the audio perception data. Specifically, the robot analyzes the audio perception data to determine if there are characteristic sounds related to the target vehicle, such as engine idling to confirm the vehicle is running, or door unlocking sounds to confirm a person is approaching the vehicle. The audio data is used to corroborate or correct the image reasoning results, outputting the vehicle's existence confidence score—the probability that the target vehicle is indeed present at that location at the current time, calculated by combining image reasoning and audio verification (ranging from 0% to 100%)—as well as the vehicle's location information, i.e., the precise coordinates of the target vehicle in the real-world coordinate system.

[0038] Preferably, the robot compares the generated vehicle presence confidence level with a system-preset confidence threshold. The preset confidence threshold is a fixed probability value, such as 80%. When the vehicle presence confidence level exceeds this preset threshold, the robot determines that the target vehicle does indeed exist at the current location and meets the conditions for executing the recovery task. At this time, the robot generates a first activation signal, which triggers the activation of the first collaborative recovery channel, such as activating the voice broadcast module to play an overdue payment reminder to the car owner. The robot uses the generated vehicle location information as input to independently generate a second activation signal. The vehicle location information includes data such as the distance and azimuth angle of the target vehicle relative to the robot. The second activation signal triggers the activation of the second collaborative recovery channel, such as adjusting the shooting angle of the robot's pan-tilt camera to shoot high-definition video recording evidence of the vehicle license plate, or driving the robot to move to a more suitable interception position. The first and second collaborative recovery channels are activated independently based on different triggering conditions and can be activated simultaneously to execute the recovery task.

[0039] Furthermore, step S500 also includes triggering a secondary query of the real-time overdue status to obtain updated data on the number of overdue orders for the target vehicle; if the updated data on the number of overdue orders is zero, a collection termination signal is generated to end the collection process; if the updated data on the number of overdue orders is greater than zero, a jump signal is generated, and the remaining overdue orders are analyzed and verified based on the jump signal and the order payment data.

[0040] Preferably, after the collaborative collection task is completed, the unmanned road parking inspection robot or the back-end management system automatically triggers a real-time overdue payment status query operation for the target vehicle. This query request is sent to the parking management back-end server's database via wireless communication, carrying the target vehicle's license plate information. Upon receiving the request, the back-end server retrieves all unpaid orders for that license plate from the parking management database at the current moment, calculates the total number of overdue orders, encapsulates the calculated number of overdue orders into updated overdue order data, and then transmits it back via the communication network. The system receives and parses the updated overdue order data, representing the total number of unpaid orders for the target vehicle at the current moment, and judges this value. If the number of overdue orders is more... If the new data value is zero, it means that the vehicle has no outstanding orders, and all outstanding fees have been resolved. A collection termination signal is generated to stop all ongoing collection-related operations and end the collection process for that vehicle. If the updated data for the number of outstanding orders is greater than zero, it means that the vehicle still has outstanding orders. A jump signal is generated to instruct the system to return to a specific stage in the collection process and re-analyze and verify the remaining outstanding orders based on the order payment data. This analysis and verification includes re-extracting order features and re-determining whether the conditions for deduction are met. The collection process is cyclically executed through the jump signal until the number of outstanding orders for the target vehicle is updated to zero or other termination conditions are met.

[0041] In the above text, refer to Figure 1 This paper describes in detail a collaborative recovery method based on an unmanned road parking inspection robot according to an embodiment of the present invention. Next, we will refer to... Figure 2 This invention describes a collaborative tracking and recovery system based on an unmanned road parking inspection robot according to an embodiment of the present invention.

[0042] The collaborative collection system based on an unmanned road parking patrol robot according to embodiments of the present invention addresses the technical problems of low efficiency, limited coverage, and difficulty in automating collection of overdue parking fees in existing technologies. It achieves the technical effects of improving the efficiency of overdue order processing, collection success rate, and the level of intelligent parking management. Figure 2 As shown, the collaborative tracking and collection system based on the unmanned road parking inspection robot includes: inspection and identification module 10, qualification determination module 20, order screening module 30, order update module 40, and collaborative tracking and collection module 50.

[0043] The inspection and identification module 10 is used to inspect and identify multiple vehicle license plate information through the unmanned road parking inspection robot, respond to queries, and generate overdue payment query results; the qualification determination module 20 is used to call the traffic management system interface to determine the vehicle's ETC payment qualification when the overdue payment query results contain historical overdue orders, and obtain the ETC qualification determination result; the order filtering module 30 is used to filter historical overdue orders according to the ETC qualification determination result, and determine the ETC deductible order set and the ETC non-deductible order set; the order update module 40 is used to automatically deduct the ETC deductible order set through the unmanned road parking inspection robot, obtain the ETC deduction result, update the order payment status, and generate order payment data; the collaborative collection module 50 is used to analyze the ETC non-deductible order set based on the order payment data, and trigger the unmanned road parking inspection robot to execute a collaborative collection task.

[0044] The specific configuration of the order filtering module 30 will be described in detail below. The order filtering module 30 further includes: parsing the ETC eligibility determination result to obtain ETC payment activation parameters and ETC payment non-activation parameters; when the ETC eligibility determination result is the ETC payment activation parameter, extracting the dynamic deduction rules of the ETC system; traversing the historical overdue orders for feature analysis to generate order overdue time features and order overdue amount features; mapping the order overdue time features to the dynamic deduction rules for timeliness comparison to generate a time-series comparison result; mapping the order overdue amount features to the dynamic deduction rules for amount comparison to generate a numerical comparison result; filtering the historical overdue orders based on the time-series comparison result and the numerical comparison result to determine the ETC deductible order set and the ETC non-deductible order set.

[0045] The specific configuration of the order filtering module 30 will be described in detail below. The order filtering module 30 further includes: extracting order arrears time features that meet the timeliness threshold based on the time-series comparison results, generating a first arrears time feature; extracting order arrears amount features that meet the amount threshold based on the numerical comparison results, generating a first arrears amount feature; filtering historical arrears orders according to the first arrears time feature and the first arrears amount feature to determine the ETC deductible order set; extracting order arrears time features that do not meet the timeliness threshold based on the time-series comparison results, generating a second arrears time feature; extracting order arrears amount features that do not meet the amount threshold based on the numerical comparison results, generating a second arrears amount feature; and filtering historical arrears orders according to the second arrears time feature and the second arrears amount feature to determine the ETC non-deductible order set.

[0046] The specific configuration of the order update module 40 will be described in detail below. The order update module 40 further includes: encoding the ETC deductible order set to construct a deduction instruction sequence; sending the sequence to the unmanned road parking inspection robot via the vehicle-road cooperative communication link, driving the unmanned road parking inspection robot to establish a near-field communication data field with the target vehicle; performing batch deductions based on the near-field communication data field according to the deduction instruction sequence, generating a deduction receipt result, the deduction receipt result including a successful deduction order identifier and a failed deduction order identifier; updating the order payment status according to the successful deduction order identifier, generating order payment status information; updating the order payment status according to the failed deduction order identifier, generating pending collaborative collection status information; and adding the order payment status information and the pending collaborative collection status information to the order payment data.

[0047] The specific configuration of the collaborative collection module 50 will be described in detail below. The collaborative collection module 50 further includes: acquiring the set of ETC non-deductible orders and merging them with the pending collaborative collection status to generate a pending collection order feature set; sending the pending collection order feature set to the unmanned road parking inspection robot as a task parameter for executing collaborative collection; extracting order quantity parameters and outstanding amount parameters based on the pending collection order feature set, dynamically activating the first collaborative collection channel and the second collaborative collection channel, and triggering the unmanned road parking inspection robot to execute the collaborative collection task.

[0048] The specific configuration of the collaborative recovery module 50 will be described in detail below. The collaborative recovery module 50 further includes: performing multi-dimensional perception of the vehicle's surroundings based on the roadside unmanned inspection robot to obtain a multi-modal perception dataset, which includes image perception data and audio perception data; performing real-time inference based on the image perception data, verifying the image inference results according to the audio perception data, and generating vehicle presence confidence and vehicle location information; generating a first activation signal to activate the first collaborative recovery channel when the vehicle presence confidence exceeds a preset confidence threshold; and generating a second activation signal to activate the second collaborative recovery channel based on the vehicle location information.

[0049] The following section will continue to describe the specific configuration of the collaborative collection module 50 in detail. The collaborative collection module 50 further includes: triggering a secondary query of the real-time overdue payment status to obtain updated data on the number of overdue orders for the target vehicle; if the updated data on the number of overdue orders is zero, generating a collection termination signal to end the collection process; if the updated data on the number of overdue orders is greater than zero, generating a jump signal, and analyzing and verifying the remaining overdue orders based on the jump signal and order payment data.

[0050] The collaborative tracking and recovery system based on the unmanned road parking inspection robot provided in this embodiment of the invention can execute the collaborative tracking and recovery method based on the unmanned road parking inspection robot provided in this embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0051] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A collaborative tracking and recovery method based on unmanned road parking inspection robots, characterized in that, The method includes: By using unmanned road parking inspection robots to inspect and identify multiple vehicle license plate information, respond to queries, and generate overdue payment query results; When the overdue payment query results show historical overdue orders, the traffic management system interface is called to determine the vehicle's eligibility for ETC payment and obtain the ETC eligibility determination result. Based on the ETC eligibility assessment results, historical overdue orders are filtered to determine the ETC deductible order set and the ETC non-deductible order set; The unmanned road parking inspection robot automatically deducts fees from the ETC deductible order set, obtains the ETC deduction result, updates the order payment status, and generates order payment data. Based on the order payment data, the ETC non-deductible order set is analyzed, triggering the unmanned road parking inspection robot to perform a collaborative collection task.

2. The collaborative tracking and recovery method based on an unmanned road parking inspection robot as described in claim 1, characterized in that, Based on the ETC eligibility assessment results, historical overdue orders are filtered to determine the ETC deductible order set and the ETC non-deductible order set. The method includes: Based on the ETC eligibility determination result, the ETC payment activation parameters and ETC payment non-activation parameters are obtained; When the ETC eligibility determination result is the ETC payment activation parameter, the dynamic deduction rules of the ETC system are extracted. The historical overdue orders are traversed and feature analysis is performed to generate features of overdue order time and overdue order amount. The order arrears time characteristics are mapped to dynamic deduction rules for timeliness comparison, and time series comparison results are generated. The order arrears amount features are mapped to dynamic deduction rules for amount comparison, and numerical comparison results are generated. Based on the time-series comparison results and the numerical comparison results, the historical overdue orders are filtered to determine the ETC deductible order set and the ETC non-deductible order set.

3. The collaborative tracking and recovery method based on an unmanned road parking inspection robot as described in claim 2, characterized in that, Based on the time-series comparison results and the numerical comparison results, the historical overdue orders are filtered to determine the ETC deductible order set and the ETC non-deductible order set. The method includes: Based on the time series comparison results, extract the order arrears time features that meet the timeliness threshold, and generate the first arrears time feature; Based on the numerical comparison results, extract the features of the outstanding amount of orders that meet the amount threshold, and generate the first outstanding amount feature. Based on the first arrears time characteristic and the first arrears amount characteristic, historical arrears orders are filtered to determine the ETC deductible order set; Based on the time series comparison results, extract the overdue payment time features of orders that do not meet the timeliness threshold, and generate a second overdue payment time feature; Based on the numerical comparison results, extract the arrears amount features of orders that do not meet the amount threshold, and generate a second arrears amount feature. Based on the second arrears time characteristic and the second arrears amount characteristic, historical arrears orders are filtered to determine the ETC non-deductible order set.

4. The collaborative tracking and recovery method based on an unmanned road parking inspection robot as described in claim 1, characterized in that, The method involves using the unmanned road parking inspection robot to automatically deduct fees from the ETC-deductible order set, obtaining the ETC deduction result, updating the order payment status, and generating order payment data. The set of ETC deductible orders is encoded to construct a deduction instruction sequence; The data is transmitted to the unmanned road parking inspection robot via the vehicle-road cooperative communication link, driving the unmanned road parking inspection robot to establish a near-field communication data field with the target vehicle; Based on the near-field communication data field, batch deductions are executed according to the deduction instruction sequence to generate a deduction receipt result. The deduction receipt result includes a successful deduction order identifier and a failed deduction order identifier. Update the order payment status based on the successful deduction order identifier, and generate order payment status information; Update the order payment status based on the failed deduction order identifier and generate pending collaborative collection status information; Add the order payment status information and the pending recovery status information to the order payment data.

5. The collaborative tracking and recovery method based on an unmanned road parking inspection robot as described in claim 1, characterized in that, Based on the order payment data, the ETC non-deductible order set is analyzed, triggering the unmanned road parking inspection robot to execute a collaborative collection task. The method includes: The set of ETC non-deductible orders is obtained and combined with the pending collection status to generate a feature set of pending collection orders. The feature set of orders to be recovered is sent to the unmanned road parking inspection robot as a task parameter for performing collaborative recovery. Based on the feature set of the orders to be recovered, the order quantity parameters and the amount of arrears parameters are extracted. The first collaborative recovery channel and the second collaborative recovery channel are dynamically activated and combined to trigger the unmanned road parking inspection robot to perform the collaborative recovery task.

6. The collaborative tracking and recovery method based on an unmanned road parking inspection robot as described in claim 5, characterized in that, The method involves dynamically activating and combining the first and second collaborative recovery channels to trigger the unmanned road parking inspection robot to perform a collaborative recovery task. Based on the road parking unmanned inspection robot, the vehicle's surroundings are perceived in multiple dimensions to obtain a multimodal perception dataset, which includes image perception data and audio perception data. Real-time reasoning is performed based on the image perception data, and the image reasoning results are verified according to the audio perception data to generate vehicle presence confidence and vehicle location information. When the confidence level of the vehicle exceeds a preset confidence threshold, a first activation signal is generated to activate the first collaborative tracking channel. A second activation signal is generated based on the vehicle location information to activate the second collaborative tracking channel.

7. The collaborative tracking and recovery method based on an unmanned road parking inspection robot as described in claim 6, characterized in that, After triggering the unmanned road parking inspection robot to perform a collaborative recovery task, the method includes: Trigger a second query of the real-time overdue payment status to obtain updated data on the number of overdue orders for the target vehicle; If the number of overdue orders is updated to zero, a collection termination signal is generated, and the collection process ends. If the updated data for the number of overdue orders is greater than zero, a jump signal is generated, and the remaining overdue orders are analyzed and verified based on the jump signal and the order payment data.

8. A collaborative tracking and recovery system based on an unmanned road parking inspection robot, characterized in that, The system is used to implement the collaborative tracking and recovery method based on an unmanned road parking inspection robot as described in any one of claims 1 to 7, and the system includes: The inspection and identification module is used to inspect and identify vehicles using unmanned road parking inspection robots, obtain license plate information of multiple vehicles, respond to queries, and generate overdue payment query results. The eligibility determination module is used to call the traffic management system interface to determine the vehicle's eligibility for ETC payment when the overdue payment query results show historical overdue orders, and to obtain the ETC eligibility determination result. The order filtering module is used to filter historical overdue orders according to the ETC eligibility determination results, and determine the ETC deductible order set and the ETC non-deductible order set; The order update module is used to automatically deduct fees from the ETC deductible order set through the unmanned road parking inspection robot, obtain the ETC deduction result, update the order payment status, and generate order payment data. The collaborative collection module is used to analyze the ETC non-deductible order set based on the order payment data and trigger the unmanned road parking inspection robot to perform a collaborative collection task.

Images