An unmanned aerial vehicle forced landing control system and method

Abstract

The application relates to the technical field of unmanned aerial vehicle control, and discloses an unmanned aerial vehicle forced landing control system and method, which comprises an unmanned aerial vehicle state recognition module, an unmanned aerial vehicle forced landing estimation module, a forced landing processing module and a forced landing evaluation module; the unmanned aerial vehicle forced landing estimation module comprises a landing point recognition unit, a landing risk evaluation unit and a pre-warning execution unit; and the forced landing evaluation module comprises a landing self-check unit and a dispatching recovery unit. The system analyzes the current forced landing state through the unmanned aerial vehicle state recognition module, evaluates the flight distance of the unmanned aerial vehicle in real time, obtains a preliminary landing point after preliminary screening of landing points, judges whether the risk coefficient of the preliminary landing point exceeds a preset warning value after the preliminary landing point is screened out, issues a risk pre-warning and implements corresponding strategies in time if the risk coefficient exceeds the preset warning value, rapidly takes countermeasures, prevents the unmanned aerial vehicle from being forced to land at a high-risk point, and thus the safety and reliability are enhanced.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) control technology, specifically to a UAV forced landing control system and method. Background Technology

[0002] With the rapid development of drone technology, drones are increasingly entering people's lives. Because drones can operate day and night, are simple in structure, easy to use, low in cost, and highly efficient, and eliminate concerns about human injury, drone operations are becoming increasingly popular in high-risk environments. They can be used for scene monitoring, weather reconnaissance, highway inspection, surveying and mapping, flood monitoring, aerial photography, traffic management, forest fire fighting, and many other applications, demonstrating extremely broad prospects.

[0003] Currently, traffic management departments often use remotely controlled drones to manage common violations such as illegal parking within their jurisdiction. However, when situations arise, such as a drone's vision module losing connection and the operator being unable to remotely control the drone based on its surroundings, an emergency landing is required. Since the drone is already in an emergency state and needs to execute a landing operation quickly, the most suitable landing point needs to be selected accurately and quickly. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a control system and method for forced landing of unmanned aerial vehicles (UAVs), which has advantages such as autonomous forced landing control of UAVs and solves the aforementioned technical problems.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a drone forced landing control system, comprising:

[0006] The drone status recognition module is used to analyze the current emergency landing status of the drone, construct the actual flight distance of the drone, and construct a safe flight area based on the constructed actual flight distance of the drone.

[0007] The drone forced landing prediction module is used to read the pre-stored terrain environment database and images captured by ground cameras. Based on the safe flight area where the drone is currently located, it constructs several drone landing points and calculates the landing risk coefficient for each drone landing point. Then, it selects the landing point with the lowest landing risk coefficient as the backup landing point. If the landing risk coefficient of the backup landing point is greater than the preset risk warning value, it issues the first risk warning signal and executes the first warning strategy. If the landing risk coefficient of the backup landing point is less than or equal to the preset risk warning value, it performs a forced landing with the backup landing point as the target. After the forced landing is completed, it issues a forced landing completion command.

[0008] The forced landing processing module is called after the first warning strategy is executed. It performs a secondary evaluation of the landing risk coefficient of several drone landing points based on the landing risk coefficient of different drone landing points, obtains the forced landing risk coefficient of several drone landing points, and selects the one with the smallest risk coefficient as the landing point for forced landing. After the forced landing is completed, it issues a forced landing completion command.

[0009] The forced landing assessment module performs a self-check on the current status of the drone after the forced landing completion command is issued, and then dispatches personnel to recover it.

[0010] As a preferred embodiment of the present invention, the drone status recognition module analyzes the current emergency landing status of the drone and constructs a safe flight area based on the current emergency landing status of the drone, and the specific steps are as follows:

[0011] Step A1: Obtain the drone's maximum flight distance based on the drone's real-time battery level;

[0012] Step A2: Construct environmental deviation coefficients based on the current forced landing status of the drone. Specifically, this involves subtracting the standard value corresponding to several drone flight environment impact values ​​from their corresponding drone flight environment impact values, then comparing the difference with the standard value corresponding to the drone flight environment impact value, and summing the results to obtain the environmental deviation coefficients. The standard value corresponding to the drone flight environment impact value is obtained through testing during drone production, and the drone flight environment impact value is acquired through sensors carried by the drone.

[0013] Step A3: Construct the actual flight distance of the UAV based on the environmental deviation coefficient and the maximum flight distance of the UAV. Specifically, the environmental deviation coefficient and the maximum flight distance of the UAV are coupled, and the coupled result is processed a second time through a preset correction constant to obtain the actual flight distance of the UAV.

[0014] As a preferred technical solution of the present invention, the UAV forced landing prediction module includes a landing point identification unit, a landing risk assessment unit, and an early warning execution unit;

[0015] The landing point identification unit constructs several drone landing points based on the safe flight area where the drone is currently located. The specific steps are as follows:

[0016] Step B1: Read the terrain environment database stored on the drone's offline terminal, remove the pre-marked no-landing areas from the safe flight area, obtain the drone's overhead coverage radius, generate the minimum landing circle for the drone, and fill it into the safe flight area according to the map scale to obtain several preliminary landing points.

[0017] Step B2: Evaluate the landing evaluation coefficient for each initial landing point. This includes obtaining the drone landing impact value and the corresponding standard value at the initial landing point, calculating the deviation rate between the standard value and the drone landing impact value, and summing all deviation rates to obtain the landing evaluation coefficient.

[0018] Step B3: Remove all points that do not meet the landing evaluation coefficient < preset landing evaluation threshold from the safe flight area, determine whether any two initial landing points are adjacent, and merge all adjacent initial landing points to obtain several UAV landing points.

[0019] As a preferred embodiment of the present invention, the landing risk assessment unit constructs several drone landing points based on the safe flight area where the drone is currently located, and the specific steps for constructing the landing risk coefficient of the drone landing points are as follows:

[0020] Step C1: Obtain the ratio of the number of people at the current drone landing point to the total number of people at all drone landing points, and at the same time obtain the ratio of the total number of people entering and leaving the current drone landing point to the sum of the total number of people entering and leaving the drone landing points. Then, average the two values ​​to obtain the dynamic risk coefficient of the drone landing point. The total number of people entering and leaving the drone landing point and the number of people at the drone landing point are both obtained by reading the camera at the location of the drone landing point.

[0021] Step C2: Obtain the area ratio of the drone landing point and the landing deviation ratio of the drone landing point. After weighted summation of the two, perform exponential calculation and multiply it by the distance ratio between the current drone position and the current drone landing point to obtain the static risk coefficient of the drone landing point.

[0022] The area ratio of the drone landing point is specifically obtained by comparing the minimum circular area of ​​the drone landing point with the area of ​​the current drone landing point. The landing deviation ratio of the drone landing point is obtained by reading the camera at the location of the current drone landing point. The distance between the current drone position and the current drone landing point is specifically the three-dimensional distance between the current drone position and the center of the current drone landing point.

[0023] Step C3: The landing risk coefficient of the drone landing point is obtained by weighted summation of the dynamic risk coefficient and the static risk coefficient of the drone landing point.

[0024] As a preferred embodiment of the present invention, the early warning execution unit is used to execute a first early warning strategy and issue a forced landing completion command. The specific steps of the early warning execution unit in executing the first early warning strategy are as follows:

[0025] Step D1: Set the drone to hover at 90%~95% of its current altitude, wait for t seconds, then re-execute the drone status recognition module and record the result obtained at this time. The updated actual flight distance of the drone At the same time, the drone forced landing prediction module is re-executed to select a backup landing point and the landing risk coefficient of the backup landing point is recalculated.

[0026] Step D2: If the landing risk coefficient of the prepared landing point is less than or equal to the preset risk warning value, continue to execute step D2, and make an emergency landing with the prepared landing point as the target. After the emergency landing is completed, issue an emergency landing completion command and output a signal that the first warning strategy has not been executed to the emergency landing processing module; otherwise, execute step D3.

[0027] Step D3: If the first The updated actual flight distance of the drone If the value is less than the alarm threshold, step D3 is terminated, and a signal indicating that the first early warning strategy has been completed is output to the forced landing processing module; otherwise, the process jumps back to step D1.

[0028] As a preferred technical solution of the present invention, the landing deviation ratio of the drone landing point is specifically obtained by averaging the area covered by obstacles at the drone landing point and the area of ​​difference between the current image captured at the drone landing point and the preset standard image, and then calculating the ratio with the area captured by the camera at the drone landing point.

[0029] As a preferred technical solution of the present invention, the forced landing processing module performs a secondary evaluation of several drone landing points based on the landing risk coefficient of different drone landing points. Specifically, the forced landing risk coefficient of the drone landing point is obtained by multiplying the landing risk coefficient of the drone landing point with the obtained recovery influence coefficient for attenuation, and the forced landing risk coefficient of several drone landing points is obtained by iterating through them in sequence.

[0030] The recovery impact coefficient is specifically calculated by dividing the area of ​​difference between the external color of the drone and the area captured by the camera at the drone's landing point by the area captured by the camera at the drone's landing point.

[0031] As a preferred technical solution of the present invention, the forced landing assessment module includes a landing self-test unit and a scheduling and recovery unit. The landing self-test unit is used to assess the forced landing process of the UAV, including plane deviation assessment, landing acceleration assessment and attitude angle assessment at landing, and outputs the UAV forced landing deviation coefficient after averaging the three assessment results.

[0032] As a preferred embodiment of the present invention, the scheduling and recovery unit will use the current position of the UAV and the UAV forced landing deviation coefficient. Send to the current scheduler, the specific steps are as follows:

[0033] Obtain the time when the drone operator arrives at the emergency landing location. And obtain a collection of times when all patrol personnel arrived at the emergency landing location. and obtain the time collection. minimum value ,when < At that time, the scheduling time collection minimum value The corresponding patrol personnel shall retrieve the drone; otherwise, the current drone operator shall retrieve it.

[0034] The present invention also provides a method for controlling the forced landing of an unmanned aerial vehicle (UAV), comprising the following steps:

[0035] Step 1: Analyze the current forced landing status of the drone and construct the actual flight distance of the drone. And based on the actual flight distance of the constructed drone To create a safe flight zone;

[0036] Step 2: Based on the current safe flight area of ​​the drone, construct several drone landing points, and construct the first... Landing risk coefficient of each drone landing point And select the landing point with the lowest landing risk coefficient as the backup landing point;

[0037] Step 3: Assess the intended landing point. If the landing risk factor of the intended landing point is... If the preset risk warning value is reached, proceed to step four. If the landing risk coefficient of the prepared landing point is... If the risk level is less than or equal to the preset risk warning value, proceed to step five.

[0038] Step 4: Issue the first risk warning signal and execute the first warning strategy. After the first warning strategy is executed, based on the... Landing risk coefficient of each drone landing point A secondary evaluation is performed on several drone landing sites to obtain the forced landing risk coefficient for the i-th drone landing site. The smallest one is selected as the landing point for forced landing. After the forced landing is completed, a forced landing completion command is issued, and step six is ​​executed.

[0039] Step 5: Perform an emergency landing at the designated landing point. After the emergency landing is completed, issue a landing completion command and proceed to Step 6.

[0040] Step Six: After the emergency landing completion command is issued, perform a self-check on the current status of the drone and dispatch personnel to recover it.

[0041] Compared with the prior art, the present invention provides a control system and method for forced landing of unmanned aerial vehicles (UAVs), which has the following beneficial effects:

[0042] 1. This invention analyzes the current forced landing status through the UAV status recognition module, evaluates the UAV's flight distance in real time, and obtains preliminary landing points after screening the landing points. After screening the preliminary landing points, the system judges whether its risk coefficient exceeds the preset warning value. If it does, the system will issue a risk warning in a timely manner and implement corresponding strategies to quickly take countermeasures to prevent the UAV from forced landing at high-risk points, thereby enhancing safety and reliability.

[0043] 2. After the forced landing command is issued, the system will perform a self-check on the drone to detect any damage or malfunction that may occur during the landing process. This provides a basis for subsequent repair and maintenance. After the status self-check is completed, the system automatically dispatches personnel to retrieve the drone, which can quickly and effectively organize resources, reduce the time that the drone is left unattended in the field, reduce the risk of further damage caused by environmental factors, and thus improve overall management efficiency. Attached Figure Description

[0044] Figure 1 This is a schematic diagram of the system framework of the present invention;

[0045] Figure 2 This is a schematic diagram of the process of the present invention. Detailed Implementation

[0046] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] Please see Figure 1 The application scenarios of this invention include, but are not limited to, situations where the drone's vision module is disconnected, and the operator is unable to remotely control the drone based on the surrounding environment, requiring a forced landing. In the event of a GPS disconnection or a disconnection of the sensor controlling the automatic landing, the drone can be controlled to hover and a warning can be issued to patrol personnel.

[0048] A drone forced landing control system includes:

[0049] The drone status recognition module is used to analyze the drone's current forced landing status and construct the drone's actual flight distance. And based on the actual flight distance of the constructed drone To establish a safe flight zone, the specific steps are as follows:

[0050] Step A1: Obtain the drone's maximum flight distance based on the drone's real-time battery level. ;

[0051] Step A2: Construct environmental deviation coefficients based on the current forced landing status of the drone. The specific expressions are as follows:

[0052]

[0053] in, This represents the standard value corresponding to the k-th drone's flight environment impact value, obtained through testing during drone production. This represents the environmental impact value of the k-th drone flight, typically... To avoid affecting the minimum values ​​of parameters corresponding to the drone's flight environment, when an anomaly occurs... It is usually less than 0. Cases greater than 0 generally have no impact or are not considered; this information is obtained through sensors carried by the drone. This represents a summation, where K represents the total number of environmental factors affecting the drone's flight environment. By analyzing the deviations of multiple parameters affecting the drone's flight environment with standard values ​​from the factory testing process, the impact of the current environment on the drone's flight can be compared, thus allowing for the determination of the maximum flight distance. Perform a deviation estimate to measure it;

[0054] Please refer to Table a below for details:

[0055] Table a

[0056]

[0057] Step A3: Based on the environmental deviation coefficient Maximum flight distance of drones Constructing the actual flight distance of drones The specific expression is as follows:

[0058]

[0059] Step A4: Using the current drone location as the center, and the actual flight distance of the drone... Using a radius, a match is made in the offline map stored within the drone, and the matched area is designated as the safe flight zone. This represents a correction constant greater than 1, preset by the operator before the drone takes flight. It is usually used to avoid the overall range being too large. This calculation is not performed on the drone itself, but is performed at the local traffic management department by collecting local meteorological data or parameters in the external environment obtained through sensors, and then sent to the drone. Subsequent calculations are performed in the same way.

[0060] The drone forced landing prediction module reads a pre-stored terrain environment database and images captured by ground cameras, constructs several drone landing points based on the current safe flight area of ​​the drone, and builds the first... Landing risk coefficient of each drone landing point The system then selects the landing point with the lowest landing risk coefficient as a backup landing point, and assesses this backup landing point based on its landing risk coefficient. If the preset risk warning value is reached, a first risk warning signal will be issued and the first warning strategy will be executed. If the landing risk coefficient of the prepared landing point is... The risk warning value is ≤, and the forced landing is carried out with the prepared landing point as the target. After the forced landing is completed, the forced landing completion command is issued.

[0061] The drone forced landing prediction module includes a landing point identification unit, a landing risk assessment unit, and an early warning execution unit;

[0062] The landing point identification unit constructs several landing points for the drone based on the safe flight area where the drone is currently located. The specific steps are as follows:

[0063] Step B1: Read the terrain environment database stored on the drone's offline terminal, and remove the pre-marked no-landing areas (including common obstacles in the area, such as streetlights, trees, as well as main roads, vehicle traffic roads, and water areas) from the safe flight area. At the same time, obtain the drone's overhead coverage radius, generate the minimum landing circle for the drone, and fill it into the safe flight area according to the map scale to obtain several preliminary landing points (for example, generate the first minimum landing circle for the drone centered on the drone's position on the map, and gradually fill it tangentially, or add minimum landing circles for the drone tangentially to any intersecting boundary of the map as the starting edge).

[0064] Step B2: Evaluate the landing assessment coefficient for each initial landing point using the following expression. :

[0065]

[0066] in, Indicates the landing assessment coefficient. Indicates the first of the initial landing points The standard value corresponding to the impact of a drone landing is here, and it is the same as the one mentioned above. Similarly, Indicates the first of the initial landing points The impact value of a drone landing, To express summation, This represents the total impact value of the drone landing. For specific landing assessment coefficients, please refer to Table b below.

[0067] Table b

[0068]

[0069] The above table is for reference only. Those skilled in the art can expand or delete the table content at their own discretion. Landing physical parameters are roughly covered by the drone's own sensors and urban topographic maps, and when... When the denominator is 0, the default is... ;

[0070] Step B3: Complete all landing assessment coefficients that do not meet the requirements. <Preset landing assessment threshold> The points are removed from the safe flight area, and it is determined whether any two initial landing points are adjacent. All adjacent initial landing points are merged separately. After merging, it is necessary to determine whether there is a camera set up by the traffic management department in the jurisdiction corresponding to the drone operator at the initial landing point. If not, it is removed, thereby ensuring the safety of the area and the availability of data, and obtaining several drone landing points.

[0071] The landing risk assessment unit constructs several drone landing points based on the current safe flight area of ​​the drone, and constructs the first... Landing risk coefficient of each drone landing point The specific steps are as follows:

[0072] Step C1: Construct the first Dynamic risk coefficient of each drone landing site The specific expression is as follows:

[0073]

[0074] in, Indicates the first The number of people at each drone landing point is determined by reading the number of drones landing points. Data acquired from cameras at the location of each drone landing site. To express summation, This indicates the total number of drone landing sites. Represents the first digit within a sampling period. The total number of people entering and exiting each drone landing point is determined by reading the data from the [number of drones]. The camera at the location of the drone landing point can be used to acquire data. The specific acquisition process can be to mark the human body using a human body recognition model within the same time period until the human body disappears. At the same time, the human body recognition model can be an existing model. Those skilled in the art can choose any specific model as long as it can achieve the process of marking the number of people at that location.

[0075] Step C2: Construct the first Static risk coefficient of each drone landing site The specific expression is as follows:

[0076]

[0077] in, Indicates the first The area ratio of each drone landing point is determined by comparing the minimum circular area of ​​the drone landing point with the area of ​​the first drone landing point. The result was obtained by comparing the areas of the drone landing sites. Indicates the first The landing deviation ratio of each drone landing point is obtained by reading the landing deviation ratio of the first drone landing point. The data was acquired from the camera at the location of the first drone landing point, i.e., the first... The camera at the location of the drone landing point (this camera refers to all cameras that can be accessed within the jurisdiction) and the first The distance between the center points of each drone landing site, if there are multiple cameras, is the distance from the center of the common circle of all cameras to the first drone landing site. The distance between the center points of each drone landing site Indicates the current position of the drone and the... The distance between each drone landing point is the distance between the drone's current position and the first drone landing point. The three-dimensional distance between the centers of the drone landing sites and These represent the weight coefficients that sum to 1;

[0078] Step C3: Based on the first Dynamic risk coefficient of each drone landing site and static risk coefficient Comprehensive construction of the first Landing risk coefficient of each drone landing point The specific expression is as follows:

[0079]

[0080] in, and These represent the weight coefficients that sum to 1;

[0081] No. The landing deviation of each drone landing point The specific expression is as follows:

[0082]

[0083] in, Indicates the first The area captured by the camera at the location of each drone landing point. Indicates the first The obstacle coverage area at each drone landing site is used to identify preset obstacle types through image recognition models. Indicates the first The area of ​​difference between the current image taken at the location of the drone landing point and the preset standard image.

[0084] The early warning execution unit is used to execute the first early warning strategy and issue a forced landing completion command. The specific steps of the early warning execution unit in executing the first early warning strategy are as follows:

[0085] Step D1: Set the drone to hover at 90%~95% of its current altitude and wait for t seconds (optionally t=10 seconds), then re-execute the drone state recognition module and record the result at this time. The updated actual flight distance of the drone At the same time, the drone emergency landing prediction module is re-executed to select a backup landing point and the landing risk coefficient of the backup landing point is recalculated. The purpose of this step is to reduce the hovering height of the drone and add voice broadcasts to issue avoidance messages to evacuate pedestrians, which can effectively reduce dynamic risks and recalculate the landing risk coefficient of the prepared landing point after the waiting period.

[0086] Step D2: If the landing risk coefficient of the proposed landing point is... If the risk warning value is less than or equal to the preset risk warning value, continue to execute step D2. At this time, the landing risk coefficient of the prepared landing point is recalculated to meet the judgment condition. The prepared landing point is used as the target for forced landing. After the forced landing is completed, a forced landing completion command is issued, and a signal that the first warning strategy has not been executed is output to the forced landing processing module. Otherwise, execute step D3.

[0087] Step D3: If the first The updated actual flight distance of the drone Less than the alarm value If the first warning strategy is completed, step D3 is terminated, and a signal indicating that the emergency landing processing module has completed its execution is output. This indicates that the drone has experienced a significant loss of endurance. Otherwise, the process jumps back to step D1. Optional... .

[0088] The forced landing handling module is invoked after the first warning strategy is executed, based on the... Landing risk coefficient of each drone landing point A secondary evaluation is performed on several drone landing sites to obtain the forced landing risk coefficient for the i-th drone landing site. The smallest one is selected as the landing point for forced landing, and a forced landing completion command is issued after the forced landing is completed.

[0089] The forced landing processing module is based on the first Landing risk coefficient of each drone landing point A secondary evaluation was conducted on several drone landing sites to obtain the first... Forced landing risk coefficient of each drone landing site The specific expression is as follows:

[0090]

[0091] in, The recycling impact coefficient is expressed as follows:

[0092]

[0093] in, Indicates the external color of the drone and the first The difference area of ​​the shooting area acquired by the camera at each drone landing point is specifically calculated by pre-determining the main color of the drone and matching it with the colors in the environment, outputting all areas that do not match the main color of the drone as the difference area. For example... =15, =50, at this time =0.3, Indicates the first The area captured by the camera at the landing site of the drone was determined during the second forced landing due to battery depletion and the inability to meet certain conditions after a prolonged wait. At this point, regarding the... Landing risk coefficient of each drone landing point By conducting secondary evaluations of several drone landing sites, we can ensure that the landing site with the greatest difference between the drone and the environment is selected, making it easier for staff to identify the drone.

[0094] The forced landing assessment module performs a self-check on the current status of the drone after the forced landing completion command is issued, and dispatches personnel for recovery.

[0095] The forced landing assessment module includes a landing self-test unit and a scheduling and recovery unit. The landing self-test unit is used to evaluate the forced landing process of the UAV and output the forced landing deviation coefficient. The specific expression is as follows:

[0096]

[0097] in, This indicates the preset maximum plane deviation. This indicates the current plane deviation value. Indicates landing acceleration. This indicates the preset maximum landing acceleration. This indicates the preset maximum attitude angle. The attitude angle at landing is indicated. The drone forced landing deviation coefficient is used to provide feedback on the degree of deviation of the drone landing. It is used to select the nearest staff member for dispatch during scheduling, thereby achieving rapid recovery and avoiding loss.

[0098] The scheduling and recovery unit will determine the current location of the drone and the drone's forced landing deviation coefficient. Send to the current scheduler, the specific steps are as follows:

[0099] Obtain the time when the drone operator arrives at the emergency landing location. And obtain a collection of times when all patrol personnel arrived at the emergency landing location. and obtain the time collection. minimum value ,when < hour, This reflects the degree of deviation in parameters during drone landing; when it is negative, it reduces... At the same time, when time collection minimum value Greater than This indicates that current drone operators are more efficient at recovery, and the scheduling time is summarized in the data. minimum value The patrol personnel shall retrieve the drone; otherwise, the current drone operator shall retrieve it.

[0100] in, This represents the drone forced landing deviation coefficient;

[0101] Example:

[0102] The actual flight distance of the constructed UAV described in this embodiment The parameters are shown in Table 1 below:

[0103] Table 1

[0104]

[0105] =196m, meaning a 196m circular area around the drone is selected, and the circular area is divided into 4 drone landing points based on the landing point identification unit. The dynamic risk coefficient of the drone landing point is then constructed, as shown in Table 2 below:

[0106] Table 2

[0107]

[0108] The static risk coefficients for constructing drone landing sites are shown in Table 3 below:

[0109] Table 3

[0110]

[0111] The landing risk factor of the drone landing point is shown in Table 4 below:

[0112] Table 4

[0113]

[0114] At this point, the landing point with the lowest landing risk coefficient is selected as the backup landing point. As a backup landing point, at this time If the risk level is less than the preset risk warning value of 0.45, an emergency landing will be initiated.

[0115] The above embodiments only provide one or more feasible solutions and do not represent the optimal solution. The amount and size of data recorded in the embodiments are only for the convenience of those skilled in the art to understand the technical solution and do not mean that the solution will only use the data recorded in the above embodiments in actual application. The threshold size is set for the convenience of comparison. The size of the threshold depends on the amount of sample data and the number of bases set by those skilled in the art for each group of sample data. As long as it does not affect the ratio between the parameter and the quantified value, it is acceptable. The weight can be determined by those skilled in the art based on each sample data and multiple rounds of experiments. The above formulas are all dimensionless calculations.

[0116] Please see Figure 2 The present invention also provides a method for controlling the forced landing of an unmanned aerial vehicle (UAV), comprising the following steps:

[0117] Step 1: Analyze the current forced landing status of the drone and construct the actual flight distance of the drone. And based on the actual flight distance of the constructed drone To create a safe flight zone;

[0118] Step 2: Based on the current safe flight area of ​​the drone, construct several drone landing points, and construct the first... Landing risk coefficient of each drone landing point And select the landing point with the lowest landing risk coefficient as the backup landing point;

[0119] Step 3: Assess the intended landing point. If the landing risk factor of the intended landing point is... If the preset risk warning value is reached, proceed to step four. If the landing risk coefficient of the prepared landing point is... If the risk level is less than or equal to the preset risk warning value, proceed to step five.

[0120] Step 4: Issue the first risk warning signal and execute the first warning strategy. After the first warning strategy is executed, based on the... Landing risk coefficient of each drone landing point A secondary evaluation is performed on several drone landing sites to obtain the forced landing risk coefficient for the i-th drone landing site. The smallest one is selected as the landing point for forced landing. After the forced landing is completed, a forced landing completion command is issued, and step six is ​​executed.

[0121] Step 5: Perform an emergency landing at the designated landing point. After the emergency landing is completed, issue a landing completion command and proceed to Step 6.

[0122] Step Six: After the emergency landing completion command is issued, perform a self-check on the current status of the drone and dispatch personnel to recover it.

[0123] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A drone forced landing control system, characterized in that: include: The drone status recognition module is used to analyze the current emergency landing status of the drone, construct the actual flight distance of the drone, and construct a safe flight area based on the constructed actual flight distance of the drone. The drone forced landing prediction module is used to read the pre-stored terrain environment database and images captured by ground cameras. Based on the safe flight area where the drone is currently located, it constructs several drone landing points and calculates the landing risk coefficient for each drone landing point. Then, it selects the landing point with the lowest landing risk coefficient as the backup landing point. If the landing risk coefficient of the backup landing point is greater than the preset risk warning value, it issues the first risk warning signal and executes the first warning strategy. If the landing risk coefficient of the backup landing point is less than or equal to the preset risk warning value, it performs a forced landing with the backup landing point as the target. After the forced landing is completed, it issues a forced landing completion command. The drone forced landing prediction module includes a landing risk assessment unit; The landing risk assessment unit constructs several drone landing points based on the current safe flight area of ​​the drone, and the specific steps for constructing the landing risk coefficient of the drone landing points are as follows: Step C1: Obtain the ratio of the number of people at the current drone landing point to the total number of people at all drone landing points, and at the same time obtain the ratio of the total number of people entering and leaving the current drone landing point to the sum of the total number of people entering and leaving the drone landing points. Then, average the two values ​​to obtain the dynamic risk coefficient of the drone landing point. The total number of people entering and leaving the drone landing point and the number of people at the drone landing point are both obtained by reading the camera at the location of the drone landing point. Step C2: Obtain the area ratio of the drone landing point and the landing deviation ratio of the drone landing point. After weighted summation of the two, perform exponential calculation and multiply it by the distance ratio between the current drone position and the current drone landing point to obtain the static risk coefficient of the drone landing point. The area ratio of the drone landing point is specifically obtained by comparing the minimum circular area of ​​the drone landing point with the area of ​​the current drone landing point. The landing deviation ratio of the drone landing point is obtained by reading the camera at the location of the current drone landing point. The distance between the current drone position and the current drone landing point is specifically the three-dimensional distance between the current drone position and the center of the current drone landing point. Step C3: The landing risk coefficient of the drone landing point is obtained by weighted summation of the dynamic risk coefficient and the static risk coefficient of the drone landing point. The forced landing processing module is called after the first warning strategy is executed. It performs a secondary evaluation of the landing risk coefficient of several drone landing points based on the landing risk coefficient of different drone landing points, obtains the forced landing risk coefficient of several drone landing points, and selects the one with the smallest risk coefficient as the landing point for forced landing. After the forced landing is completed, it issues a forced landing completion command. The forced landing assessment module performs a self-check on the current status of the drone after the forced landing completion command is issued, and then dispatches personnel to recover it.

2. The unmanned aerial vehicle (UAV) forced landing control system according to claim 1, characterized in that: The drone status recognition module analyzes the current emergency landing status of the drone and constructs a safe flight area based on the current emergency landing status. The specific steps are as follows: Step A1: Obtain the drone's maximum flight distance based on the drone's real-time battery level; Step A2: Construct environmental deviation coefficients based on the current forced landing status of the drone. Specifically, this involves subtracting the standard value corresponding to several drone flight environment impact values ​​from their corresponding drone flight environment impact values, then comparing the difference with the standard value corresponding to the drone flight environment impact value, and summing the results to obtain the environmental deviation coefficients. The standard value corresponding to the drone flight environment impact value is obtained through testing during drone production, and the drone flight environment impact value is acquired through sensors carried by the drone. Step A3: Construct the actual flight distance of the UAV based on the environmental deviation coefficient and the maximum flight distance of the UAV. Specifically, the environmental deviation coefficient and the maximum flight distance of the UAV are coupled, and the coupled result is processed a second time through a preset correction constant to obtain the actual flight distance of the UAV.

3. The unmanned aerial vehicle (UAV) forced landing control system according to claim 2, characterized in that: The drone forced landing prediction module includes a landing point identification unit. The landing point identification unit constructs several drone landing points based on the safe flight area where the drone is currently located. The specific steps are as follows: Step B1: Read the terrain environment database stored on the drone's offline terminal, remove the pre-marked no-landing areas from the safe flight area, obtain the drone's overhead coverage radius, generate the minimum landing circle for the drone, and fill it into the safe flight area according to the map scale to obtain several preliminary landing points. Step B2: Evaluate the landing evaluation coefficient for each initial landing point. This includes obtaining the drone landing impact value and the corresponding standard value at the initial landing point, calculating the deviation rate between the standard value and the drone landing impact value, and summing all deviation rates to obtain the landing evaluation coefficient. Step B3: Remove all points that do not meet the landing evaluation coefficient < preset landing evaluation threshold from the safe flight area, determine whether any two initial landing points are adjacent, and merge all adjacent initial landing points to obtain several UAV landing points.

4. The unmanned aerial vehicle (UAV) forced landing control system according to claim 3, characterized in that: The UAV forced landing prediction module includes an early warning execution unit, which is used to execute a first early warning strategy and issue a forced landing completion command. The specific steps of the early warning execution unit in executing the first early warning strategy are as follows: Step D1: Set the drone to hover at 90%–95% of its current altitude, and wait for t seconds. Then, re-execute the drone status recognition module and record the actual flight distance R of the drone after the Nth update. * N At the same time, the drone forced landing prediction module is re-executed to select a backup landing point and the landing risk coefficient of the backup landing point is recalculated. Step D2: If the landing risk coefficient of the prepared landing point is less than or equal to the preset risk warning value, continue to execute step D2, and make an emergency landing with the prepared landing point as the target. After the emergency landing is completed, issue an emergency landing completion command and output a signal that the first warning strategy has not been executed to the emergency landing processing module; otherwise, execute step D3. Step D3: If the actual flight distance R of the UAV after the Nth update... * N If the value is less than the alarm threshold, step D3 is terminated, and a signal indicating that the first early warning strategy has been completed is output to the forced landing processing module; otherwise, the process jumps back to step D1.

5. A drone forced landing control system according to claim 1, characterized in that: The landing deviation ratio of the drone landing point is specifically calculated by averaging the area covered by obstacles at the drone landing point and the area of ​​difference between the current image captured at the drone landing point and a preset standard image, and then comparing it with the captured area obtained by the camera at the drone landing point.

6. The unmanned aerial vehicle (UAV) forced landing control system according to claim 1, characterized in that: The forced landing processing module performs a secondary evaluation of several drone landing points based on the landing risk coefficient of different drone landing points. Specifically, it obtains the forced landing risk coefficient of the drone landing point by multiplying the landing risk coefficient of the drone landing point by the obtained recovery influence coefficient used for attenuation, and then iterates through several drone landing points to obtain the forced landing risk coefficient. The recovery impact coefficient is specifically calculated by dividing the area of ​​difference between the external color of the drone and the area captured by the camera at the drone's landing point by the area captured by the camera at the drone's landing point.

7. The unmanned aerial vehicle (UAV) forced landing control system according to claim 1, characterized in that: The forced landing assessment module includes a landing self-test unit and a scheduling and recovery unit. The landing self-test unit is used to assess the forced landing process of the UAV, including plane deviation assessment, landing acceleration assessment and attitude angle assessment at landing, and outputs the UAV forced landing deviation coefficient after averaging the three assessment results.

8. A drone forced landing control system according to claim 7, characterized in that: The scheduling and recovery unit sends the UAV's current location and the UAV forced landing deviation coefficient PJPC to the current scheduler. The specific steps are as follows: Get the time T when the current drone operator arrives at the emergency landing location, and get the set SJ of the arrival times of all patrol personnel at the emergency landing location. Then, get the minimum value SJ in the set SJ. min , when SJ min When <T*(1+PJPC), the minimum value SJ in the set of scheduling times SJ. min The corresponding patrol personnel shall retrieve the drone; otherwise, the current drone operator shall retrieve it.

9. A method for controlling forced landing of an unmanned aerial vehicle (UAV), based on a UAV forced landing control system as described in any one of claims 1-8, characterized in that: Includes the following steps: Step 1: Analyze the current forced landing status of the drone and construct the actual flight distance R of the drone. * And based on the actual flight distance R of the constructed drone * To create a safe flight zone; Step 2: Based on the current safe flight area of ​​the drone, construct several drone landing points, and construct the landing risk coefficient JLFXXS for the i-th drone landing point. i And select the landing point with the lowest landing risk coefficient as the backup landing point; Step 3: Determine the intended landing point. If the landing risk factor of the intended landing point is JLFXXS min If the preset risk warning value is reached, proceed to step four. If the landing risk coefficient of the prepared landing point is JLFXXS min If the risk level is less than or equal to the preset risk warning value, proceed to step five. Step 4: Issue the first risk warning signal and execute the first warning strategy. After the first warning strategy is executed, calculate the landing risk coefficient JLFXXS based on the i-th UAV landing point. i A secondary evaluation is performed on several drone landing sites to obtain the forced landing risk coefficient PJFXXS for the i-th drone landing site. i The smallest one is selected as the landing point for forced landing. After the forced landing is completed, a forced landing completion command is issued, and step six is ​​executed. Step 5: Perform an emergency landing at the designated landing point. After the emergency landing is completed, issue a landing completion command and proceed to Step 6. Step Six: After the emergency landing completion command is issued, perform a self-check on the current status of the drone and dispatch personnel to recover it.

Images