Aircraft automatic inspection-based residential area curb parking management method and system

Abstract

The invention relates to the technical field of traffic engineering, in particular to a residential area curb parking management method and system based on aircraft automatic inspection. The method comprises the following steps: 1, formulating a refined parking management method according to parking information conditions of a residential area and current related parking management policies of the residential area; 2, determining an aircraft automatic inspection take-off timetable based on a bilevel programming model; 3, constructing an on-road parking management system, and realizing automatic routing inspection of the aircraft based on a GPS technology and a PID control algorithm; 4, performing information transmission and communication based on a cellular network communication link; 5, recognizing video information of vehicle parking based on a lightweight mode recognition technology; and 6, managing and interacting parking information by using a parking management subsystem. The method is used for solving the technical problem that the cost and the management efficiency cannot be effectively balanced due to the lack of utilization of intelligent equipment and management methods in current residential area on-road parking management, and the parking management system can be migrated to multiple scenes for use.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic engineering, and particularly to a method and system for managing on-street parking in residential areas based on automatic inspection by an aircraft. Background Art

[0002] With the continuous increase in the number of motor vehicles in China, the problem of difficult parking in residential areas of some large and medium-sized cities has become increasingly prominent, and parking resources are generally in short supply. Therefore, on-street parking areas with short construction periods and low costs have become the main measures to alleviate the parking difficulty. In recent years, aircraft have been widely used in the traffic field due to their characteristics such as light weight, low price, and high efficiency, especially in traffic information collection and traffic monitoring. Moreover, with the continuous progress in sensor technology, automatic flight technology, visual recognition technology, etc., the performance and reliability of aircraft automatic inspection technology have been continuously improved. In parking management, aircraft automatic inspection technology is widely used because it can quickly and efficiently cover large areas.

[0003] The existing management of on-street parking in residential areas mainly relies on traffic police patrols. Even though traffic monitoring cameras are installed at most intersections and main traffic roads, there are still many monitoring blind spots. At the same time, time-based charging facilities such as parking meters and fixed cameras are set up in some areas, but their high system hardware facility layout costs and operating costs limit their application scope. On the one hand, the existing on-street parking management system cannot achieve a good balance between cost and management efficiency, and on the other hand, its flexibility is poor and it cannot be migrated to other scenarios for use. Therefore, there is an urgent need to design a smart, efficient, green, energy-saving, and flexible management system for on-street parking areas in residential areas. Therefore, it is particularly important to conduct automatic inspection based on aircraft and establish an on-street parking management system in residential areas. Summary of the Invention

[0004] The present invention provides a method and system for managing on-street parking in residential areas based on automatic inspection by an aircraft, which is used to solve the technical problem that the current management of on-street parking in residential areas cannot effectively balance cost and management efficiency due to the lack of utilization of intelligent devices and management methods. Moreover, the on-street parking management system of the present invention can be migrated to multiple scenarios for use and has flexibility.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] A method for managing on-street parking in residential areas based on automatic inspection by an aircraft is carried out according to the following steps:

[0007] Step 1: Formulate a refined parking management method based on the parking information situation in the residential area and the current relevant parking management policies in the residential area;

[0008] Step 2: Determine the take-off schedule of the aircraft for automatic inspection based on the bilevel programming model;

[0009] Step 3: Construct an on-street parking management system, which includes a computer control subsystem, an aircraft subsystem, and a parking management subsystem; and realize the automatic inspection of the aircraft based on GPS technology and PID control algorithm;

[0010] Step 4: Transmit and communicate information based on the cellular network communication link;

[0011] Step 5: Identify the video information of vehicle parking based on lightweight pattern recognition technology;

[0012] Step 6: Use the parking management subsystem to manage and interact with the parking information.

[0013] Furthermore, in Step 2, the process of determining the take-off schedule of the aircraft for automatic inspection based on the bilevel programming model is as follows:

[0014] The bilevel programming model in the present invention considers from the perspective of the operator on the one hand and from the perspective of user parking on the other hand;

[0015] Considering from the perspective of the operator, the operating cost of the on-street parking management system includes two parts, namely fixed cost and variable cost; the fixed cost C0 includes the operator's salary, the aircraft purchase cost, the aircraft maintenance cost, and the battery hardware cost, which are regarded as fixed values and not considered for optimization. In the present invention, only the variable cost of the operating cost of the on-street parking management system is considered; the variable cost of the operating cost of the on-street parking management system includes the aircraft energy cost C1 and the aircraft depreciation cost C2;

[0016] During the flight of the aircraft, it consumes oil or electrical energy, so the energy cost C1:

[0017] C1 = N × F i × F p (1)

[0018] In the formula: C1 is the aircraft energy cost; N is the total number of flights of the aircraft; F i is the average power or fuel consumption per flight; F p is the unit price of energy;

[0019] The depreciation cost C2 of the aircraft flight:

[0020]

[0021] In the formula: C2 is the aircraft depreciation cost; P is the purchase price of the aircraft; Y is the scrapping life of the aircraft; 5% is the salvage rate of the aircraft;

[0022] Then the upper-level planning model considering the total cost of the operator is as follows:

[0023] C P = C1 + C2 (3)

[0024] Wherein, C P is the variable cost of the operation cost of the on-street parking management system; C1 is the energy cost of the aircraft; C2 is the depreciation cost of the aircraft;

[0025] The lower-level planning model considering the minimum charging error from the perspective of charging users for parking in the bi-level planning model is as follows:

[0026]

[0027] Wherein: C g is the error cost of aircraft charging; S 1i is the actual charging amount required; S 2i is the actual charging amount at the parking duration detected by aircraft flight;

[0028] Weighing the operation cost of the on-street parking management system and the interest relationship of charging users for parking, a bi-level planning model based on the take-off time interval of the aircraft is established to minimize the total system cost. The integrated model is as follows:

[0029]

[0030] Wherein: C is the total system cost; p is the cost weight coefficient, p ∈ [0, 1]; S 1i is the actual charging amount required; S 2i is the actual charging amount at the parking duration detected by aircraft flight; N is the total number of aircraft flights; F i is the average power or fuel consumption per flight; F p is the unit price of fuel; P is the purchase price of the aircraft; Y is the scrapping life of the aircraft; 5% is the residual value rate of the aircraft;

[0031] The relationship between the parking duration t and the parking fee S charged is as follows:

[0032] S = f(t) (6)

[0033] Wherein: S is the parking fee charged; t is the parking duration; f(t) represents the functional relationship between the parking duration t and the fee;

[0034] The aircraft takes pictures of the vehicle during flight and charges fees within the specified charging time, so there is:

[0035] T start ≤ c j ≤ T end (7)

[0036] Where: T start is the charging start time every day; T end is the free parking start time every day, c j represents the take-off time of the aircraft for the j-th time;

[0037] The time interval between each take-off of the aircraft should not be too large or too small, then:

[0038] T min <x j,k ≤T max (8)

[0039] Where: T min ,, T max are the minimum and maximum time intervals for the aircraft to take off; x j,k is the time interval for the aircraft to take off at the j-th time in the k-th period;

[0040] The time for the next flight of the aircraft is the time of the previous flight plus the take-off time interval:

[0041] c j+1 =c j +x j,k (9)

[0042] Where: c j represents the take-off time of the aircraft for the j-th time; c j+1 represents the take-off time of the aircraft for the (j + 1)-th time; x j,k is the time interval for the aircraft to take off at the j-th time in the k-th period;

[0043] The total number of flights N of the aircraft is the sum of the quotients of the durations of each time period and the corresponding flight time intervals:

[0044]

[0045] Where: N is the total number of flights of the aircraft, T bk is the end time of the k-th period; T ak is the start time of the k-th period; x j,k is the time interval for the aircraft to take off at the j-th time in the k-th period;

[0046] After the model is established, use the genetic algorithm to solve. After obtaining the take-off time intervals of the aircraft in different periods, starting from the charging start time T start accumulate one by one until the charging end time T end to obtain the automatic inspection take-off schedule of the aircraft.

[0047] Further, in step 3, the computer control subsystem is used to analyze and process data and send flight instructions to the aircraft, including a calculation module, a control module, a communication module, and a storage module; the calculation module is used to analyze and process data, decode and encode the data, and provide a basis for mission planning; the control module is used to set network and aircraft parameters, plan inspection tasks, and send connection requests and control instructions to the aircraft subsystem; the communication module is used for information transmission and communication between the computer control subsystem and the aircraft subsystem; the storage module is used to store system data and system configuration parameters.

[0048] Further, in step 3, the aircraft subsystem is used to perform automatic inspection tasks in the off-street parking area, including a communication module, a video acquisition module, a flight control module, and a GPS navigation module; the communication module is used for information transmission and communication between the aircraft subsystem and the computer control subsystem, including a communication component supporting the cellular network, and connects to the network through initialization settings to complete data sending and receiving; the video acquisition module is used to collect on-site video data of the off-street parking area and encode the video data; the flight control module is used to decode flight instructions, encode video data, allocate tasks, and control the flight of the aircraft; the GPS navigation module is used to provide positioning and navigation information for the aircraft and decode key information for the flight control module to call.

[0049] Further, the aircraft subsystem is also provided with a dynamic obstacle avoidance module, which is used to ensure the safety of the flight process, preferably a dynamic obstacle avoidance module based on lidar and binocular vision.

[0050] Further, in step 3, the parking management subsystem is used to realize the management of parking information by the administrator and the interaction between the user and the system, including four modules: a user end, a management end, a service end, and a database.

[0051] Further, in step 3, the automatic inspection of the aircraft is realized based on the GPS technology and the PID control algorithm. The specific steps are as follows:

[0052] (1) The computer control subsystem sends a connection request to the aircraft subsystem and establishes a stable link with it;

[0053] (2) The computer control subsystem performs mission planning, encapsulates the data and sends it to the aircraft subsystem;

[0054] (3) After receiving the data packet, the aircraft subsystem decodes it and allocates flight-related tasks accordingly;

[0055] (4) The flight control module of the aircraft subsystem controls the aircraft to take off and fly according to the planned route, and gives real-time feedback. The aircraft subsystem collects data and stores and transmits it back.

[0056] Further, in step 5, the video information for identifying vehicle parking based on the lightweight pattern recognition technology specifically includes two aspects: vehicle self-information recognition and the recognition of the position relationship between the vehicle and the parking line; the vehicle self-information recognition includes license plate information recognition, pass information recognition, and vehicle model information recognition; the recognition of the position relationship between the vehicle and the parking line includes two types: pressing the line and not pressing the line.

[0057] Further, the license plate information recognition includes four steps: data preprocessing, license plate positioning, character segmentation, and character recognition.

[0058] Further, the recognition of the position relationship between the vehicle and the parking line includes four steps: data preprocessing, parking line detection, vehicle detection and positioning, and line pressing judgment.

[0059] Based on the same inventive concept, the present invention also provides a roadside parking management system for residential areas based on automatic inspection by an aircraft, which is used to implement the above-mentioned roadside parking management method for residential areas based on automatic inspection by an aircraft.

[0060] Compared with the prior art, the present invention has the following beneficial technical effects:

[0061] (1) Scientifically quantifies the time interval of aircraft inspection, and combines the PID control algorithm and GPS positioning technology to achieve real-time dynamic collection and management of roadside parking information in urban residential areas automatically, greatly improving the inspection efficiency and management efficiency.

[0062] (2) Compared with traditional methods such as manual management or video monitoring, the present invention saves the computing resources, storage resources, and power resources required by the parking management system on the premise of ensuring the inspection efficiency, and effectively realizes the balance between the cost and management efficiency of roadside parking management.

[0063] (3) Can be migrated to other parking management scenarios, significantly improving the economy, green low-carbon nature, and flexibility of parking management. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 is the technical roadmap of the roadside parking management method for residential areas of the present invention;

[0065] Figure 2 is the schematic diagram of automatic inspection by the aircraft of the present invention;

[0066] Figure 3 is the schematic diagram of real-time recognition of automatic inspection by the aircraft of the present invention;

[0067] Figure 4 is the schematic diagram of the actual mode of automatic inspection by the aircraft of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0068] To make the objectives, technical solutions, and advantages of the present invention more clear, the following describes the embodiments of the present invention in detail with reference to the accompanying drawings.

[0069] The present invention proposes a method for managing on-street parking in a residential area based on an automatic inspection aircraft, which is carried out according to the following steps:

[0070] Step 1: Develop a refined parking management method based on the parking information situation in the residential area and the current relevant parking management policies in the residential area.

[0071] First, collect relevant on-street parking information through methods such as on-site manual collection, collection by fixed detection devices, or invocation of mobile signaling data, then perform data preprocessing, and based on traffic information processing technology, obtain the parking information situation in the residential area for the recent month or a longer period, mainly including license plate numbers, entry and exit times, parking durations, the proportion of parking for foreign vehicles and residential vehicles, etc.

[0072] Secondly, investigate the current situation of parking management in this area through the government information disclosure website, including investigating the current on-street parking management policies and regulations in the residential area, specifically the parking time limit, parking space delineation, charging standards, etc., and develop a refined parking management method based on resident and non-resident vehicles, so as to provide a data basis for the subsequent full-automatic inspection of the aircraft.

[0073] The temporary parking charging standard on the roadway is generally formulated by the government, but it can be further refined according to the parking composition and distribution characteristics of the residential area to formulate corresponding charging management standards. Since most of the on-street parking areas in the residential area are used by the original residents most of the time, and only a small part of the time is used for temporary parking by foreign vehicles, it is possible to charge residents and non-residents separately for easier management. For residents, they are charged regularly, a certain amount of fees are collected monthly, quarterly, or annually, and then no additional parking fees are charged during normal parking, and a Permit is equipped on the resident vehicle for exemption from inspection. The form of the Permit can be diverse, such as RFID license, Bluetooth / wireless communication license, sensor license, license plate number, etc., which can greatly improve the charging efficiency; for foreign vehicles, due to their temporary parking characteristics, they need to be detected and charged for each parking, and no regular charging is required. Its charging rules can be formulated according to the local government's system for on-street parking charging, or reasonably changed according to the need for social resource allocation on this basis, and reported to and negotiated with the relevant departments to obtain charging permission. Therefore, when the aircraft detects a vehicle during flight inspection and detects the signal of the Permit, it does not need to detect and charge it, and only needs to collect and identify the information of foreign non-resident vehicles, which can save a large amount of inspection time and aircraft information storage space.

[0074] When the aircraft checks the vehicle parking, in addition to detecting the normal parked vehicles of residents and non-residents, it may also detect vehicles with illegal parking behaviors such as over the line. For this, it is also necessary to formulate a method for charging illegal parking fees, and obtain the consent of the relevant management departments to recover the illegal parking fees. When the aircraft detects that a vehicle has an illegal parking behavior such as over the line, it will remind and warn the vehicle owner. If the illegal parking behavior is not corrected after exceeding the allowed time threshold, a fee will be deducted and reported to the relevant departments for point deduction. If the illegal parking fee is not paid after exceeding the allowed time threshold, the penalty will be increased.

[0075] At the same time, regarding the charging of the aircraft, in order to scientifically guide parking demand, the parking time type is divided into free parking time and charging time. Different time periods can be divided within the specified charging time, such as peak periods and off-peak periods, and corresponding charging rules are formulated for charging.

[0076] Step 2: Determine the take-off schedule of the aircraft's automatic inspection based on the bilevel programming model.

[0077] Combined with the constraints such as the parking time limit, parking space demarcation, and charging standard in this area obtained through the survey information, as well as the historical parking data information, a bilevel programming model is established to solve the take-off schedule of the aircraft's automatic inspection, so that the aircraft takes off and inspects according to this schedule.

[0078] The bilevel programming model is used to describe the interaction and decision-making process between two decision-makers. The upper-level decision-maker usually considers the impact of the behavior of the lower-level decision-maker on its own goal; the lower-level decision-maker usually considers the impact of the upper-level decision-maker's decision on its own goal and tries to choose the best strategy to respond.

[0079] The bilevel programming model in the present invention considers from the perspective of the operator on the one hand and from the perspective of the user's parking on the other hand.

[0080] From the perspective of the operator, the operating cost of the on-street parking management system includes two parts, namely fixed cost and variable cost. The fixed cost C0 considered in the present invention includes operator salary, aircraft purchase cost, aircraft maintenance cost, battery hardware cost, etc. Its value is relatively fixed during daily operation and will not change in a short time, so it can be regarded as a fixed value and optimization is not considered. In the present invention, only the variable cost of the operating cost of the on-street parking management system is considered. For the variable cost of the operating cost of the on-street parking management system, the present invention mainly considers the aircraft energy cost C1, aircraft depreciation cost C2, etc.

[0081] During the flight of the aircraft, it needs to consume energy such as engine oil or electric energy, so the energy cost C1:

[0082] C1 = N × F i × F p(1)

[0083] Where: C1 is the energy cost of the aircraft; N is the total number of flights of the aircraft; F i is the power or fuel consumption per flight on average; F p is the unit price of the energy.

[0084] The performance of the aircraft will decline over time and has a certain service life. Therefore, it is evenly spread into the depreciation cost C2 of the aircraft's daily flight:

[0085]

[0086] Where: C2 is the depreciation cost of the aircraft; P is the purchase price of the aircraft; Y is the scrapping life of the aircraft; 5% is the salvage rate of the aircraft;

[0087] Then the upper-level planning model considered from the perspective of the total cost of the operator is:

[0088] C P = C1 + C2 (3)

[0089] Where, C P is the variable cost of the operation cost of the on-street parking management system; C1 is the energy cost of the aircraft; C2 is the depreciation cost of the aircraft.

[0090] The bilevel programming model also considers from the perspective of charging users for parking. The purpose of using the aircraft to detect the entry and exit times of vehicles is for charging. It only needs to ensure that the charging amount S 1i required for the actual parking of the vehicle is as close as possible to the actual charging amount S 2i under the parking duration detected by the aircraft flight. Therefore, the lower-level planning model considered from the perspective of minimizing the charging error is as follows:

[0091]

[0092] Where: C g is the error cost of aircraft charging; S 1i is the charging amount required for the actual need; S 2i is the actual charging amount under the parking duration detected by the aircraft flight.

[0093] Weighing the operation cost of the on-street parking management system and the interest relationship of charging users for parking, a bilevel programming model based on the takeoff time interval of the aircraft is established to minimize the total system cost. The integrated model is as follows:

[0094]

[0095] Where: C is the total system cost; p is the cost weight coefficient, p ∈ [0, 1]; S 1iThe toll amount for actual needs; S 2i The actual toll amount during the parking duration for aircraft flight detection; N is the total number of aircraft flights; F i The power or fuel consumption per average flight; F p The unit price of fuel; P is the purchase price of the aircraft; Y is the scrapping life of the aircraft; 5% is the salvage rate of the aircraft.

[0096] There is a relationship between the parking duration t and the collected parking fee S as follows:

[0097] S = f(t) (6)

[0098] In the formula: S is the collected parking fee; t is the parking duration; f(t) represents the functional relationship between the parking duration t and the fee.

[0099] The aircraft captures vehicles during flight and charges fees only within the specified toll collection time, so there is:

[0100] T start ≤ c j ≤ T end (7)

[0101] In the formula: T start Is the start time of toll collection every day; T end Is the start time of free parking every day, c j Represents the takeoff time of the aircraft for the jth time.

[0102] To avoid excessive takeoff frequency causing an increase in the total cost and ensure that the aircraft can capture the real state of vehicles entering, leaving and parking as much as possible during detection, the time interval between each takeoff of the aircraft should not be too large or too small, then:

[0103] T min < x j,k ≤ T max (8)

[0104] In the formula: T min ,, T max Are the minimum and maximum time intervals between aircraft takeoffs; x j,k Is the time interval between the kth period and the jth takeoff of the aircraft;

[0105] The time for the next flight of the aircraft is the time of the previous flight plus the takeoff time interval:

[0106] c j+1 = c j + x j,k (9)

[0107] In the formula: c jIndicates the take-off time of the aircraft for the j-th time; c j+1 Indicates the take-off time of the aircraft for the (j + 1)-th time; x j,k is the time interval for the aircraft to take off for the j-th time in the k-th time period.

[0108] The total number of flights N of the aircraft is the sum of the quotients of the durations of each time period and the corresponding flight time intervals:

[0109]

[0110] In the formula: N is the total number of flights of the aircraft, T bk is the end time of the k-th time period; T ak is the start time of the k-th time period; x j,k is the time interval for the aircraft to take off for the j-th time in the k-th time period.

[0111] After the model is established, the genetic algorithm is used to solve it, and the process of intelligent calculation of the genetic algorithm is realized by using a python program. After obtaining the take-off time intervals of the aircraft in different time periods, starting from the start time T start accumulate one by one until the end time T end of charging by time, the take-off schedule of the aircraft can be obtained, that is, the automatic inspection take-off schedule of the aircraft.

[0112] Step 3: Construct an on-street parking management system. The on-street parking management system includes: a computer control subsystem, an aircraft subsystem, and a parking management subsystem; and the automatic inspection of the aircraft is realized based on GPS technology and PID control algorithm.

[0113] Step 3.1: Construct an on-street parking management system.

[0114] The on-street parking management system includes a computer control subsystem, an aircraft subsystem, and a parking management subsystem.

[0115] The computer control subsystem is used to analyze and process data and send flight instructions to the aircraft, including a calculation module, a control module, a communication module, and a storage module. Among them, the calculation module is used to analyze and process data, decode and encode the data, and provide a basis for task planning. The data includes flight data and parking management data. The control module is used to set network and aircraft parameters, plan inspection tasks (including setting flight speed, altitude, inspection area, take-off time, etc.), and send connection requests and control instructions to the aircraft subsystem; the communication module is used for information transmission and communication between the computer control subsystem and the aircraft subsystem, including transmitting information such as videos and task data packets, negotiating communication parameters to establish a link, and ensuring smooth information transmission; the storage module is used to store system data (including parking and flight data collected by the aircraft subsystem, such as vehicle information and flight trajectories), as well as system configuration parameters, etc.

[0116] The aircraft subsystem is used to perform the automatic inspection task in the off-street parking area, including a communication module, a video acquisition module, a flight control module, and a GPS navigation module. Among them, the communication module is used for information transmission and communication between the aircraft subsystem and the computer control subsystem, including communication components supporting cellular networks, and connecting to the network through initialization settings to complete data sending and receiving; the video acquisition module is used to collect on-site video data in the off-street parking area and encode the video data; the flight control module is used to decode flight instructions, encode video data, allocate tasks, and control the flight of the aircraft. A position controller is set in the flight control module, which combines GPS technology, PID control algorithm, and other sensor data to achieve navigation and flight control; the GPS navigation module is used to provide positioning and navigation information for the aircraft, and decodes key information through the UBX protocol for the flight control module to call.

[0117] To ensure the safety of the flight process, the aircraft subsystem can also be equipped with a dynamic obstacle avoidance module, preferably a dynamic obstacle avoidance module based on lidar and binocular vision. The lidar is mainly used to detect obstacles around the aircraft fuselage, obtain distance data through continuous scanning and reflection of radar waves, and timely feedback to the aircraft for attitude correction to achieve obstacle avoidance; the binocular vision technology is mainly used to detect obstacles in front of the aircraft. The left camera extracts feature points and calibrates internal and external parameters to obtain a calibration model. The right camera extracts feature points and then determines the coordinates of the feature points, matches the features with the left camera, and uses the matched coordinates for three-dimensional reconstruction to obtain the depth and width information of the obstacle. Finally, obstacle avoidance is performed according to the path algorithm of the aircraft subsystem.

[0118] The parking management subsystem is used to realize the management of parking information by the administrator and the interaction between the user and the system, including four modules: the user side, the management side, the service side, and the database. The user side needs to develop functional modules and design the GUI interface according to the functional modules; the management side needs to select a suitable technology stack according to the system deployment environment (such as a web browser or a desktop application), then develop modules for management functions, and build an interface design framework based on the functional modules. The service side needs to design interfaces in the style of RESTful API, implement the business logic of the interfaces, then integrate with the payment system to realize the payment function, and finally deploy the developed service side application to the server. The database stores and manages the data generated by the user side, the management side, and the service side. The functions and technology stacks of the four modules of the parking management subsystem are shown in Table 1.

[0119] Table 1 Functions and Technology Stacks of the Parking Management Subsystem

[0120]

[0121] Step 3.2: Implement automatic inspection of the aircraft based on GPS technology and PID control algorithm.

[0122] When the preset takeoff time is reached, the computer control subsystem sends a flight instruction to the aircraft subsystem. After receiving the flight instruction, the flight control module of the aircraft subsystem decodes and assigns tasks to the flight instruction, and then uses GPS technology and PID control algorithm to enable the aircraft to accurately start and fly along the starting point of the preset path.

[0123] The specific steps are as follows:

[0124] (1) The computer control subsystem sends a connection request to the aircraft subsystem and establishes a stable link with it;

[0125] (2) The computer control subsystem conducts task planning, encapsulates data and sends it to the aircraft subsystem;

[0126] (3) After receiving the data packet, the aircraft subsystem decodes it and assigns flight-related tasks accordingly.

[0127] (4) The flight control module of the aircraft subsystem controls the aircraft to take off and fly according to the planned route, and provides real-time feedback. The aircraft subsystem collects data and stores and transmits it back.

[0128] This inspection method can greatly reduce the problem of being unable to capture license plate information caused by excessive tilting of the shooting angle during aerial photography. Specifically, the on-street parking management system of the present invention embeds a GPS navigation module into the on-board computer of the aircraft, writes corresponding initialization code, uses the UBX protocol to communicate between the on-board computer and the GPS navigation module, and decodes important information such as longitude and latitude, sea level height, lock type, horizontal / vertical accuracy estimate, horizontal / vertical dilution of precision, GPS noise value, NED velocity, number of satellites, etc. for the flight control module to call. After the flight control module decodes the GPS data, it combines it with the accelerometer data, attitude data, and barometer data for integrated navigation to estimate the required inertial navigation data for automatic inspection navigation. When the aircraft subsystem receives GPS data, it uses a position controller and PID algorithm to control the flight. Based on the real-time calculation of the PID algorithm and the real-time control of the position controller, the aircraft then conducts inspections according to the planned route based on the control of the position controller.

[0129] The setting of the automatic inspection route of the aircraft is as Figure 2 shown.

[0130] Step 4: Conduct information transmission and communication based on the cellular network communication link.

[0131] The cellular network (4G / 5G) communication link has the advantages of low cost and good portability. Therefore, the present invention adopts the cellular network for the transmission and communication of parking management information between the computer control subsystem and the aircraft subsystem.

[0132] During the inspection process, after the video acquisition module on the aircraft subsystem acquires the video, the flight control module of the aircraft subsystem first performs video encoding on the video, and then encapsulates the encoded video data into a suitable transmission protocol. Finally, the communication module of the aircraft subsystem sends the encapsulated video data to the cellular network, and the computer control subsystem receives the video data sent by the aircraft subsystem through the corresponding protocol stack. Conversely, the computer control subsystem encodes and sends the information to the aircraft subsystem in the same way, and the flight control module of the aircraft subsystem receives and decodes the information. Thus, the information transmission and communication functions are completed.

[0133] The aircraft subsystem is equipped with a communication module that supports the cellular network. This module includes a card slot for inserting a 4G / 5G communication card and a modem. After the communication card is inserted, the communication module is initialized, including accurately setting information such as the access point name (APN), username, and password, so that the aircraft can successfully connect to the cellular network of the operator. This cellular network consists of macro base stations or micro base stations deployed by telecommunications operators as the basic coverage units. They are widely connected to the Internet through the core network, building a network foundation for the communication between the aircraft subsystem and the outside, so as to realize the efficient processing and transmission preparation of subsequent data.

[0134] During the execution of tasks by the aircraft subsystem, the video acquisition module on it starts to work, and real-time video images of the scene are acquired according to set parameters such as frame rate and resolution. Due to the bandwidth limitation and transmission characteristic requirements of the cellular network, the acquired video data needs to be efficiently encoded. The present invention adopts the H.265 video encoding standard for efficient encoding. Through a more advanced compression algorithm, H.265 encoding can reduce the data volume by about 30%-50% compared with traditional encoding standards (such as H.264) while ensuring the video quality. This is extremely beneficial in an environment where the cellular network has relatively limited bandwidth and is vulnerable to interference during long-distance transmission, and can effectively improve the efficiency and smoothness of video transmission. The encoded video data needs to be encapsulated into a suitable transmission protocol to meet the reliable transmission requirements in a complex network environment. For real-time video transmission scenarios, protocols based on TCP / IP are commonly used, such as HTTP or RTSP (Real-Time Streaming Protocol). During the encapsulation process, source address, destination address, port number, and protocol header information are added to form a complete data packet that can be transmitted in the network.

[0135] Step 5: Identify the video information of vehicle parking based on lightweight pattern recognition technology.

[0136] During the process of full-automatic flight, the aircraft needs to detect the information of parked vehicles in real time. First, the video acquisition module of the aircraft subsystem collects the real-time parking situation, and then the captured video is transmitted back to the computer control subsystem in real time through the communication module. Next, the high-computing-power and long-endurance computing module of the computer control subsystem performs real-time video data processing.

[0137] Based on the lightweight pattern recognition technology, the video information of vehicle parking is recognized, specifically including two aspects: the recognition of vehicle own information and the recognition of the position relationship between the vehicle and the parking line. The computer control subsystem performs pattern recognition on these two types of information simultaneously, so as to provide a data basis for subsequent management and interaction.

[0138] Specifically, the recognition of vehicle own information mainly includes license plate information recognition, pass information recognition, vehicle model information recognition, etc., all of which are the recognition of the characteristics of the vehicle itself. The recognition of the position relationship between the vehicle and the parking line mainly includes two situations: pressing the line and not pressing the line.

[0139] 1. Recognition of vehicle own information

[0140] The computer control subsystem performs real-time recognition on the video data to obtain the relevant information of the vehicle, which is realized through a lightweight recognition model based on deep learning. The present invention uses the LPRNET deep learning model for pattern recognition. LPRNET has good feature extraction and recognition capabilities and can effectively handle recognition tasks in complex environments. The present invention takes license plate recognition as an example for illustration. License plate information recognition includes four steps: data preprocessing, license plate localization, character segmentation, and character recognition:

[0141] (1) Data preprocessing.

[0142] Data preprocessing is the process of converting the original image into a format that the model can effectively process. In this process, the present invention implements adaptive histogram equalization (AHE) to improve local contrast and correct images with uneven illumination. And a bilateral filter is used to denoise the image, which not only retains edge information but also removes random noise. Finally, the Laplacian operator is used to enhance the edge features of the license plate area for subsequent license plate localization. The process also includes necessary operations such as image scaling, padding, normalization, affine transformation, enhancement (smoothing correction), and dimension rearrangement, which are determined according to the local situation of the data.

[0143] (2) License plate localization

[0144] The present invention uses an improved Canny algorithm to detect the window edges of the sampling area, and at the same time uses the threshold in the Hue-Saturation-Value (HSV) space for segmentation to distinguish license plates of different colors.

[0145] (3) Character segmentation

[0146] The present invention uses a trained positive and negative sample classifier to slide on an image, then generates a probability response map, and then performs NMS (non-maximum suppression) on the raw response. After determining the number of character bounding boxes, the optimal segmentation path is obtained.

[0147] (4) Character recognition

[0148] After the license plate characters are segmented, if each segmented character is clearly visible, a predefined template is used to match the possible character set, otherwise a pre-trained convolutional neural network (CNN) is used for recognition.

[0149] 2. Recognition of the position relationship between the vehicle and the stop line

[0150] The recognition of the position relationship between the vehicle and the stop line is mainly divided into two types: on the line and not on the line. This information is realized by a deep learning algorithm for pattern recognition, including four steps: data preprocessing, parking line detection, vehicle detection and positioning, and on-line judgment.

[0151] (1) Data preprocessing

[0152] On the one hand, data preprocessing is grayscale processing, which converts a color image into a grayscale image, which can reduce the amount of data and highlight key features such as lines in the image. For example, for an RGB color image, it is converted into a grayscale image using the weighted average method, and the formula is:

[0153] G ray = 0.299R + 0.587G + 0.114B (11)

[0154] where R, G, and B are the pixel values of the red, green, and blue channels in the color image, respectively.

[0155] On the other hand, it is filtering and denoising processing, and median filtering or Gaussian filtering is used to remove the noise in the image. Median filtering has a good effect on removing salt-and-pepper noise, and it realizes filtering by replacing the value of each pixel point with the median of its neighboring pixel values. Gaussian filtering is to perform weighted averaging on pixels according to the Gaussian function, mainly used to remove Gaussian noise.

[0156] (2) Parking line detection

[0157] First, the parking space line edge detection is performed, and the Canny edge detection algorithm is used to detect the edges in the image. The Canny algorithm determines the edge by finding the local maximum of the image intensity gradient. It has two main thresholds, a high threshold and a low threshold. Pixel gradients greater than the high threshold are considered strong edges, and those less than the low threshold are considered noise and ignored. Pixels between the two are also considered edges if they are connected to strong edges. Based on the geometric features of the parking space line (such as straight line features), the Hough Transform can be used to extract the parking space line. The Hough Transform can convert the straight line detection problem in the image space into the parameter space for processing.

[0158] (3) Vehicle detection and positioning

[0159] First, vehicle feature extraction is performed. Detection is performed based on the appearance features of the vehicle, such as the outline, color, texture, etc. The YOLO (You Only Look Once) algorithm is used to detect the position of the vehicle in the image. Based on the vehicle bounding box information obtained by the target detection algorithm, the position coordinates of the vehicle in the image are determined. These coordinates can be used to determine whether the vehicle crosses the line later.

[0160] (4) Line pressure judgment

[0161] First, convert the coordinates of the vehicle bounding box and the parking line to the same coordinate system. For example, if the parking line is represented in the global coordinate system of the image, then the coordinates of the vehicle bounding box need to be converted to this global coordinate system. Then calculate the number of intersections between the vehicle bounding box and the parking line or determine whether the vertex of the vehicle bounding box is on one side of the parking line and other geometric relationships. If one side of the vehicle bounding box intersects with the parking line, or the vertex of the vehicle bounding box exceeds the area defined by the parking line, the vehicle is determined to have crossed the line. For example, if the vertex of the left front wheel bounding box of the vehicle is on the left side of the parking line (assuming that the parking line defines the right boundary of the parking area), the vehicle is determined to have crossed the line.

[0162] Step 6: Use the parking management subsystem to manage and interact with parking information.

[0163] After the aircraft subsystem and the computer control subsystem complete the information transmission, the parking management subsystem is needed to enable the administrator to manage the parking information obtained through the detection, as well as the interaction function between the user and the parking management subsystem.

[0164] In order to achieve efficient interaction with users and improve the management efficiency of administrators, the design of the parking management subsystem includes the following important parts:

[0165] (1) User interaction and management system functional architecture:

[0166] Complete key account management functions such as user registration, identity verification, and personal profile modification; provide parking history search and parking fee settlement queries to improve user transparency; guide users to bind mobile devices to achieve real-time notification and communication functions.

[0167] (2) Intelligent parking space monitoring and navigation system: Utilize the data collected by the aircraft subsystem to dynamically present the status of parking spaces in the parking lot; provide parking space navigation and intelligent allocation services through high-precision positioning technology and parking space interconnection technology.

[0168] (3) Dynamic parking fee calculation and payment system: Dynamically calculate parking fees based on parking time, parking space type, and time period; implement a differential charging system based on vehicle size and occupied space; support multiple electronic payment methods such as mobile payment and traditional bank payment to improve payment convenience.

[0169] (4) Illegal behavior detection and notification system: Use aircraft evidence collection and image recognition technology to automatically detect illegal parking behaviors; send warnings and corrective guidance for illegal behaviors to users through preset communication channels.

[0170] (5) User opinion and complaint mechanism: Allow users to submit suggestions and complaints about parking services, including handling objections to fees and penalties.

[0171] (6) Additional service access platform: Establish cooperative relationships with automobile-related service providers to provide users with information on additional services such as refueling, charging, maintenance, and finance.

[0172] To implement the above functions, the following technology stack is adopted:

[0173] The user side adopts the latest technology stack of Vite + TypeScript + Vue3 + ElementPlus technology architecture. Among them, Vue3 uses a progressive JavaScript framework, and the application of Element Plus can build a beautiful front-end interface. Use Vite as the front-end build tool to improve development and packaging speed, and apply TypeScript to enhance the type safety of JavaScript.

[0174] In this invention, the SSM framework (Spring, Spring MVC, and MyBatis) constructed for the management side is a commonly used technology stack in enterprise-level application development. The Spring framework provides powerful dependency injection and aspect-oriented programming capabilities, while Spring MVC is responsible for handling HTTP requests and responses, and MyBatis is used for database operations; use the Django framework to develop VSCode to create RESTful APIs, and use the SpringMVC framework to manage the PostgreSQL database.

[0175] The server mainly consists of the following units: (1) Controller unit: responsible for receiving and querying parameters from accounts, management terminals, and external systems, and calling service modules to process business logic, and finally returning data on response or error information; (2) Service unit: responsible for executing business logic, such as calculating parking costs, calculating remaining spaces, and sending notifications, and calling the Mapper module for database operations. (3) Mapper unit: responsible for expanding, deleting, modifying, and searching database operations to ensure data consistency and security.

[0176] Use FineReport to establish a PostgreSQL database to achieve database visualization, creation, and maintenance. The API integration adopts the RESTful protocol, aiming to separate the front and back ends and promote secondary development.

[0177] Combining the above steps, the actual mode of the automatic inspection of the aircraft of the present invention is as Figure 4 shown.

[0178] Based on the same inventive concept, the present invention also provides a roadside parking management system in a residential area based on the automatic inspection of an aircraft, which is used to implement the above-mentioned roadside parking management method in a residential area based on the automatic inspection of an aircraft.

[0179] Embodiment:

[0180] The following further details the implementation manner of the present invention through a specific embodiment.

[0181] Step 1: Develop a refined parking management method based on the parking information situation in the residential area and the current relevant parking management policies in the residential area.

[0182] First, collect relevant parking information of a roadside parking lot in a residential area in Harbin through the method of manual on-site recording, and then based on traffic information processing technology, the parking information situation of this residential area in the recent week can be obtained, mainly including license plate numbers, entry and exit times, parking durations, and parking situations of foreign vehicles and residential area vehicles, as shown in Table 2:

[0183] Table 2 One-week Parking Details in a Certain Area of Harbin

[0184] License Plate Number Entry Time Exit Time Parking Duration (hours) Is it an External Vehicle Black A*** January 1st, 8:00 January 1st, 18:00 10 Yes Black A*** January 1st, 10:00 January 1st, 15:00 5 Yes Black B*** January 1st, 12:00 January 1st, 17:45 5.75 Yes …… …… …… …… ……

[0185] For residents, a regular fee is charged at 1000 yuan per year, and no additional parking fee is charged during normal parking. A sensor Permit is equipped on the resident's vehicle to avoid inspection. For non-resident vehicles, due to their temporary parking characteristics, each parking needs to be detected and charged, and no regular fee is required. The charging rules follow the management method issued by the Harbin Municipal Government and obtain the charging permit.

[0186] Refer to the on-street parking management method issued by the Harbin government to determine the charging rules for a certain section. The charging time is generally divided into free parking time and charging time. The free parking time is generally from 20:00 at night to 8:00 the next day, and the charging time is from 8:00 to 20:00. Different time periods can be divided within the specified charging time, such as peak periods and off-peak periods, and corresponding charging rules are formulated for charging.

[0187] The specific method is as follows: Free parking within 30 minutes; Charge 5 yuan from 30 minutes to 2 hours; When it is more than 2 hours, an additional 1 yuan is charged for each hour exceeding 2 hours; that is:

[0188]

[0189] When the aircraft subsystem checks the vehicle parking and finds illegal parking behaviors such as crossing the line, a fine of 50 yuan per time is charged, and with the consent of the relevant management department, the illegal parking fee is recovered. When the aircraft subsystem detects that the vehicle has illegal parking behaviors such as crossing the line, it will give a reminder and warning to the vehicle owner. If the illegal parking behavior has not been corrected after more than 30 minutes, the fee will be deducted and reported to the relevant department for point deduction. If the illegal parking fee has not been paid after more than 24 hours, the penalty will be increased.

[0190] Step 2: Determine the take-off schedule of the aircraft for automatic inspection based on the bi-level programming model.

[0191] Based on the residential parking information obtained in Step 1 and combined with the on-street parking charging rules in this area, establish a bi-level programming model to solve the take-off time of the aircraft automatically.

[0192] In the bi-level programming model, the calculation of various costs and take-off times refers to Formulas (1) to (10) and Formula (12). In this embodiment, the parameter situation is: F i ×F p = 0.007 yuan per time, P is 5000 yuan; Y is 3 years, p = 0.8; T start = 8:00; T end = 20:00; T min = 6, T max = 30.

[0193] After completing the calibration of the parameters, use a Python program to implement the process of intelligent calculation of the genetic algorithm. The value of the evolutionary algebra of the population individuals is 100, the value of the crossover probability is 0.9, the value of the mutation probability is 0.05, and the value of the constant term of the model is 0.6. After obtaining the takeoff time intervals of the aircraft at different time periods, starting from the start time charging moment T start Accumulate one by one until the end time charging moment T end , the takeoff schedule of the aircraft can be obtained, and the aircraft can perform the inspection task according to the time in this table. The takeoff schedule is shown in Table 3:

[0194] Table 3 Aircraft Automatic Inspection Takeoff Schedule

[0195]

[0196] Step 3: Build an on-street parking management system. The on-street parking management system includes: a computer control subsystem, an aircraft subsystem, and a parking management subsystem; and realize the automatic inspection of the aircraft based on GPS technology and PID control algorithm.

[0197] The on-street parking management system includes a computer control subsystem, an aircraft subsystem, and a parking management subsystem.

[0198] The computer control subsystem is used to analyze and process data and send flight instructions to the aircraft subsystem, including a calculation module, a control module, a communication module, and a storage module. The aircraft subsystem is used to perform the automatic inspection task of the on-street parking area, including a communication module, a video acquisition module, a flight control module, and a GPS navigation module. The parking management subsystem is used to realize the management of parking information by the administrator and the interaction between the user and the system, including four modules: a user end, a management end, a service end, and a database.

[0199] When the aircraft is performing inspections, it presets the waypoint through GPS technology. When the GPS navigation module of the aircraft subsystem receives GPS data, the flight control module of the aircraft subsystem uses a position controller and PID algorithm to preset the flight state of the aircraft (including altitude, shooting angle, flight speed, flight direction, etc.), and then can control the aircraft to perform inspections along the preset path. The preset inspection path of the aircraft is the center line direction of the on-street parking area. At the same time, the starting and ending positions of the flight also need to be preset to ensure that the inspection can be carried out from the starting point to the ending point. Each time the aircraft starts from the starting point and performs inspections according to the preset path until the ending point. After completing a one-way trip, the aircraft returns along the original path to the starting point, thus completing the inspection this time. The real-time automatic inspection status of the aircraft can be seen in Figure 2 and Figure 3 .

[0200] Step 4: Transmit information and communicate based on the cellular network communication link.

[0201] The aircraft is equipped with a communication module that supports the cellular network. This module includes a card slot for inserting a 4G / 5G communication card (ordinary mobile phone card) and a modem. After inserting the communication card, initialize the settings of the communication module, including accurately setting information such as the access point name (APN), username, and password (if required by the operator), so that the aircraft subsystem can successfully connect to the operator's cellular network. At the same time, the aircraft subsystem is equipped with a video acquisition module. In this embodiment, a high-definition camera is used to collect on-site video data and has a corresponding video encoding and processing unit to achieve efficient processing and transmission of subsequent video data.

[0202] Step 5: Identify the video information of vehicle parking based on lightweight pattern recognition technology.

[0203] During the fully automatic flight of the aircraft, it is necessary to detect the information of parked vehicles in real time. First, collect the real-time parking situation through the video data acquisition module of the aircraft subsystem, then transmit the captured video back to the computer control subsystem in real time through the communication module, and then perform real-time video image processing by the high-computing-power and long-endurance computing module of the computer control subsystem. In real-time video image processing, it mainly includes two aspects: identifying the vehicle's own information and the position relationship between the vehicle and the parking line. The computer control subsystem identifies these two types of information simultaneously. Specifically, the vehicle's own information mainly includes license plate information, pass information, vehicle model information, etc. This information is realized through a lightweight license plate recognition model based on deep learning. This method needs to go through four steps: data preprocessing stage, license plate positioning stage, character segmentation stage, and character recognition stage, so as to effectively handle the license plate recognition task in a complex environment. The position relationship between the vehicle and the parking line is mainly divided into two types: pressing the line and not pressing the line. This information is realized through a deep learning algorithm and needs to go through four steps: data preprocessing, parking line detection, vehicle detection and positioning, and line-pressing judgment. The obtained information provides a data basis for subsequent management and interaction. An example of the aircraft's identification of vehicles parked by local residents during automatic patrol is Figure 3 as shown.

[0204] Step 6: Use the parking management subsystem to manage and interact with parking information.

[0205] Upload the vehicle information identified in Step 5 to the parking management subsystem for the management of management personnel, user information management and information query operations by users, the server for data storage and processing, request response and processing, and implementation of business logic, and the database for the construction and storage of various form information.

[0206] The embodiments of the present invention are preferred embodiments rather than limitations thereof. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, modifications may be made to the specific embodiments or equivalent replacements may be made to some technical features, and all of them shall be regarded as falling within the scope of the present invention.

Claims

1. A method for managing on-street parking in a residential area based on automatic inspection by an aircraft, characterized in that it is which is carried out according to the following steps: Step 1: Develop a refined parking management method based on the parking information in the residential area and the current parking management policies in the residential area; Step 2: Determine the automatic inspection take-off schedule of the aircraft based on the bilevel programming model; Step 3: Construct an on-street parking management system, which includes a computer control subsystem, an aircraft subsystem, and a parking management subsystem; and realize the automatic inspection of the aircraft based on GPS technology and PID control algorithm; Step 4: Conduct information transmission and communication based on the cellular network communication link; Step 5: Identify the video information of vehicle parking based on lightweight pattern recognition technology; Step 6: Use the parking management subsystem to manage and interact with the parking information.

2. The method for managing on-street parking in residential areas based on automatic inspection by an aircraft according to claim 1, wherein In Step 2, the specific process of determining the automatic inspection take-off schedule of the aircraft based on the bilevel programming model is as follows: The bilevel programming model in the present invention considers from the perspective of the operator on the one hand and from the perspective of user parking on the other hand; Considering from the perspective of the operator, the operating cost of the on-street parking management system includes two parts, namely fixed cost and variable cost; the fixed cost C0 includes operator salary, aircraft purchase cost, aircraft maintenance cost, and battery hardware cost, which are regarded as fixed values and not considered for optimization. In the present invention, only the variable cost of the operating cost of the on-street parking management system is considered; the variable cost of the operating cost of the on-street parking management system includes aircraft energy cost C1 and aircraft depreciation cost C2; During the flight of the aircraft, it consumes oil or electrical energy, so the energy cost C1: C1 = N × F i × F p (1) Where: C1 is the energy cost of the aircraft; N is the total number of flights of the aircraft; F i is the power or fuel consumption per flight on average; F p is the unit price of the energy; The depreciation cost C2 of the aircraft flight: In the formula: C2 is the aircraft depreciation cost; P is the aircraft purchase price; Y is the aircraft scrap life; 5% is the aircraft salvage rate; Then the upper-level programming model considering from the perspective of the total operator cost is: C P = C1 + C2 (3) where C P is the variable cost of the operation cost of the off-street parking management system; C1 is the energy cost of the aircraft; C2 is the depreciation cost of the aircraft; The bilevel programming model considers from the perspective of charging users for parking. The lower-level programming model considering from the perspective of minimizing the charging error is as follows: Where: C g is the error cost of aircraft charging; S 1i is the charging amount actually required; S 2i is the actual charging amount for the parking duration during the flight inspection of the aircraft; Weigh the operating cost of the on-street parking management system and the interest relationship of charging users for parking, establish a bilevel programming model based on the aircraft take-off time interval to minimize the total system cost, and the integrated model is as follows: Where: C is the total system cost; p is the cost weight coefficient, p ∈ [0, 1]; S 1i is the charging amount actually required; S 2i is the actual charging amount at the parking duration of the aircraft flight detection; N is the total number of aircraft flights; F i is the average power or fuel consumption per flight; F p is the unit price of fuel; P is the purchase price of the aircraft; Y is the scrapping life of the aircraft; 5% is the salvage rate of the aircraft; There is the following relationship between the parking duration t and the parking fee S charged: S = f(t) (6) In the formula: S is the parking fee charged; t is the parking duration; f(t) represents the functional relationship between the parking duration t and the fee; During the flight of the aircraft, it takes pictures of the vehicle and charges fees within the specified charging time, so there is: T start ≤c j ≤T end (7) Where: T start is the charging start time every day; T end is the free parking start time every day, and c j represents the take-off time of the aircraft for the j-th time; The time interval between each take-off of the aircraft should not be too large or too small, then: T min <x j,k ≤T max (8) Where: T min、 and T max are the minimum and maximum time intervals for the aircraft to take off; x j,k is the time interval for the aircraft to take off at the k-th time period and the j-th time; The time of the next flight of the aircraft is the time of the previous flight plus the take-off time interval: c j+1 = c j + x j,k (9) where: c j represents the takeoff time of the aircraft for the j-th time; c j+1 represents the takeoff time of the aircraft for the (j + 1)-th time; x j,k is the time interval of the aircraft taking off for the j-th time in the k-th period; The total number of flights N of the aircraft is the sum of the quotients of the durations of each time period and the corresponding flight time intervals: Where: N is the total number of flights of the aircraft, T bk is the end time of the k-th time period; T ak is the start time of the k-th time period; x j,k is the time interval of the j-th takeoff of the aircraft in the k-th time period; After the completion of the model establishment, the genetic algorithm is used to solve the problem. After obtaining the takeoff time intervals of the aircraft at different time periods, starting from the time T when the charging starts start accumulate one by one until the end time T of charging end , an automatic inspection takeoff schedule of the aircraft is obtained.

3. The method for managing on-street parking in a residential area based on automatic inspection by an aircraft according to claim 1, wherein In step 3, the computer control subsystem is used to analyze and process data and send flight instructions to the aircraft, including a calculation module, a control module, a communication module, and a storage module. The calculation module is used to analyze and process data, decode and encode the data, and provide a basis for mission planning. The control module is used to set network and aircraft parameters, plan inspection tasks, and send connection requests and control instructions to the aircraft subsystem. The communication module is used for information transmission and communication between the computer control subsystem and the aircraft subsystem. The storage module is used to store system data and system configuration parameters.

4. The method for managing on-street parking in a residential area based on automatic inspection of an aircraft as claimed in claim 1, wherein In step 3, the aircraft subsystem is used to perform the automatic inspection task of the on-street parking area, including a communication module, a video acquisition module, a flight control module, and a GPS navigation module. The communication module is used for information transmission and communication between the aircraft subsystem and the computer control subsystem, including a communication component supporting the cellular network, and connecting to the network through initialization settings to complete data sending and receiving. The video acquisition module is used to acquire on-site video data of the on-street parking area and encode the video data. The flight control module is used to decode flight instructions, encode video data, allocate tasks, and control the flight of the aircraft. The GPS navigation module is used to provide positioning and navigation information for the aircraft and decode key information for the flight control module to call.

5. The method for managing on-street parking in a residential area based on automatic inspection of an aircraft according to claim 4, characterized in that, The aircraft subsystem is also provided with a dynamic obstacle avoidance module, which is used to ensure the safety of the flight process, preferably a dynamic obstacle avoidance module based on lidar and binocular vision.

6. The method for managing on-street parking in a residential area based on automatic inspection of an aircraft according to claim 1, wherein In step 3, the parking management subsystem is used to realize the management of parking information by the administrator and the interaction between the user and the system, including four modules: a user end, a management end, a service end, and a database.

7. The method for managing on-street parking in a residential area based on automatic inspection of an aircraft according to claim 1, characterized in that, In step 3, the automatic inspection of the aircraft is realized based on the GPS technology and the PID control algorithm. The specific steps are as follows: (1) The computer control subsystem sends a connection request to the aircraft subsystem and establishes a stable link with it. (2) The computer control subsystem performs mission planning, encapsulates the data, and sends it to the aircraft subsystem. (3) After receiving the data packet, the aircraft subsystem decodes it and allocates flight-related tasks accordingly. (4) The flight control module of the aircraft subsystem controls the aircraft to take off and fly according to the planned route, and provides real-time feedback. The aircraft subsystem collects data, stores it, and transmits it back.

8. The method for managing on-street parking in a residential area based on automatic inspection of an aircraft according to claim 1, wherein In step 5, the video information of vehicle parking is identified based on the lightweight pattern recognition technology, which specifically includes two aspects: the identification of vehicle own information and the identification of the position relationship between the vehicle and the parking line. The identification of vehicle own information includes license plate information identification, pass information identification, and vehicle model information identification. The identification of the position relationship between the vehicle and the parking line includes two cases: pressing the line and not pressing the line.

9. The method for managing on-street parking in residential areas based on automatic inspection of aircraft as claimed in claim 8, wherein The license plate information identification includes four steps: data preprocessing, license plate positioning, character segmentation, and character recognition.

10. The method for managing on-street parking in a residential area based on automatic inspection of an aircraft according to claim 8, wherein, The identification of the position relationship between the vehicle and the parking line includes four steps: data preprocessing, parking line detection, vehicle detection and positioning, and line pressing judgment.

11. A roadside parking management system for residential areas based on automatic inspection by an aircraft, characterized in that, It is used to implement the on-street parking management method in the residential area based on the automatic inspection of the aircraft as described in any one of claims 1 to 10.

Images