A method and system for unmanned landing based on scene adaptation

Abstract

The present invention proposes an unmanned landing method based on scene adaptation, comprising the following steps: a sensor unit and a plurality of positioning sources are carried on an unmanned aerial vehicle, the sensor unit collects surrounding environment information, the plurality of positioning sources respectively locate relative positions, and obtain the positioning information of each positioning source; a positioning adjustment model is constructed, and an optimal positioning source is calculated based on the positioning adjustment model according to the surrounding environment information and the positioning information; the unmanned aerial vehicle comprises an unmanned aerial vehicle control module, and the relative position information of a landing target is obtained according to the optimal positioning source, and the relative position information is input into the unmanned aerial vehicle control module, so that the unmanned aerial vehicle lands according to the relative position information, and a comprehensive evaluation is performed according to the positioning accuracy, scene, signal strength and data delay of different positioning sources, and the positioning source with the highest total score is selected as the current optimal positioning source, so that the positioning accuracy of the unmanned aerial vehicle can be improved, which is beneficial for the unmanned aerial vehicle to more accurately determine its own position in key tasks such as landing, and reduce errors; the overall reliability, stability and adaptability of the unmanned aerial vehicle positioning system are enhanced, and accurate and reliable autonomous landing can be achieved in a complex or dynamically changing environment.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicles, and in particular to an unmanned landing method and system based on scene adaptation. Background Art

[0002] As the application of drones in various fields continues to increase, the requirements for drone landing are becoming higher and higher. The safety and stability of drones are crucial to various applications, especially in harsh environments or complex scenarios. A more reliable and stable landing system is needed to ensure the safety of the drone and the surrounding environment.

[0003] The publication number is CN106054903B. A multi-rotor UAV adaptive landing method and system. After the UAV takes off, the approximate range of the area to be landed is selected through a ground control station (such as a tablet computer, a mobile phone, etc.); after receiving the landing command, the landing mode is turned on and the landing action is waited; the landing surface of the landing area is detected to see if it is suitable for the landing of the UAV. If so, the UAV lands and ends the landing mode; otherwise, the UAV posture is changed and the landing continues.

[0004] The current drone landing system has the following problems: the landing error is large. The current GNSS system positioning accuracy is generally above one meter, which cannot meet the high-precision landing requirements; it lacks adaptability and cannot select appropriate peripheral data for landing according to the current environment; the landing stability is insufficient, and there is no corresponding protection system when abnormalities occur during landing in harsh or complex environments, which reduces the accuracy and reliability of the drone landing. Summary of the invention

[0005] In view of this, the present invention proposes an unmanned landing method and system based on scene adaptation, which can achieve accurate and reliable autonomous landing in complex or dynamically changing environments.

[0006] The technical solution of the present invention is implemented as follows: In a first aspect, the present invention provides an unmanned landing method based on scene adaptation, the method comprising the following steps:

[0007] S1, the UAV is equipped with a sensor unit and several positioning sources. The sensor unit collects surrounding environment information, and several positioning sources respectively locate the relative positions to obtain the positioning information of each positioning source;

[0008] S2, constructing a positioning adjustment model, and calculating and obtaining the optimal positioning source based on the positioning adjustment model according to the surrounding environment information and positioning information;

[0009] S3, the drone includes a drone control module, which obtains relative position information of the landing target according to the optimal positioning source, and inputs the relative position information into the drone control module, so that the drone lands according to the relative position information.

[0010] On the basis of the above technical solution, preferably, the drone in step S1 is equipped with a sensor unit and several positioning sources, the sensor unit collects surrounding environment information, and the several positioning sources respectively locate the relative position to obtain the positioning information of each positioning source, wherein the sensor unit includes a photosensitive sensor and a laser radar, and the several positioning sources include a visual positioning module, an ALIGN-GPS positioning module and a UWB positioning module.

[0011] On the basis of the above technical solution, preferably, the photosensitive sensor is used to collect the illumination data of the surrounding environment and determine whether the current scene is indoors or outdoors;

[0012] The laser radar is used to collect and obtain the positions of obstacles in the surrounding environment;

[0013] The visual positioning module is used to capture images and videos of the surrounding environment, and obtain the light intensity, positioning error and video frame rate of the surrounding environment;

[0014] The ALIGN-GPS positioning module is used to collect positioning error, signal strength and data delay of relative positioning position;

[0015] The UWB relative positioning module is used to collect positioning error, signal strength and data delay of the relative positioning position.

[0016] On the basis of the above technical solution, preferably, the construction of the positioning adjustment model in step S2, according to the surrounding environment information and the positioning information, the optimal positioning source is calculated based on the positioning adjustment model, wherein the following sub-steps are included:

[0017] S21, preset accuracy scoring rules, match the positioning errors corresponding to the visual positioning module, the ALIGN-GPS positioning module and the UWB positioning module, and obtain corresponding positioning accuracy scores respectively;

[0018] S22, presetting a scenario scoring rule, matching the reliability of each positioning source according to the current scenario, and obtaining corresponding reliability scores respectively;

[0019] S23, preset a signal strength scoring rule, match the corresponding signal strengths of the ALIGN-GPS positioning module and the UWB positioning module, and obtain corresponding availability scores respectively; preset a light intensity scoring rule, match the light intensity of the visual positioning module, and obtain corresponding availability scores;

[0020] S24, preset a delay scoring rule, match the corresponding data delays of the ALIGN-GPS positioning module and the UWB positioning module, and obtain corresponding delay scores respectively; preset a frame rate scoring rule, match the video frame rate of the visual positioning module, and obtain corresponding delay scores;

[0021] S25, constructing a positioning adjustment model, and calculating a total score corresponding to each positioning source based on the positioning adjustment model according to the positioning accuracy score, the reliability score, the availability score, and the delay score;

[0022] S26, according to the total scores of the positioning sources, selecting the positioning source with the highest total score as the optimal positioning source for the current scene.

[0023] On the basis of the above technical solution, preferably, the calculation expression of the positioning adjustment model is:

[0024] S=W1*Q i +W2*M i +W3*N i +W4*L i

[0025] In the formula, S is the total score corresponding to the i-th positioning source, Q i is the positioning accuracy score corresponding to the i-th positioning source, M i is the reliability score corresponding to the i-th positioning source, N i is the availability score corresponding to the i-th positioning source, L i is the delay score corresponding to the ith positioning source, W1, W2, W3 and W4 are the weights corresponding to each score, W1+W2+W3+W4=1.

[0026] On the basis of the above technical solution, preferably, the accuracy scoring rule, signal strength scoring rule and delay scoring rule are all piecewise functions;

[0027] The accuracy scoring rule expression is:

[0028]

[0029] In the formula, J i is the positioning error value corresponding to the i-th positioning source;

[0030] If the photosensitive sensor detects that the location is outdoor, the corresponding reliability score of the ALIGN-GPS positioning module is 0.9, the corresponding reliability score of the UWB positioning module is 0.7, and the corresponding reliability score of the visual positioning module is 0.4. If the photosensitive sensor detects that the location is indoor, the corresponding reliability score of the ALIGN-GPS positioning module is 0.1, the corresponding reliability score of the UWB positioning module is 0.3, and the corresponding reliability score of the visual positioning module is 0.6.

[0031] If the positioning source is the ALIGN-GPS positioning module or the UWB positioning module, the signal strength scoring rule expression is:

[0032]

[0033] Where, X i The signal strength corresponding to the ALIGN-GPS positioning module and the UWB positioning module;

[0034] If the positioning source is a visual positioning module, the signal strength scoring rule expression is:

[0035]

[0036] In the formula, G i is the light intensity corresponding to the visual positioning module;

[0037] If the positioning source is the ALIGN-GPS positioning module or the UWB positioning module, the delay scoring rule expression is:

[0038]

[0039] In the formula, C i The data delay corresponding to the ALIGN-GPS positioning module and the UWB positioning module;

[0040] If the positioning source is the visual positioning module, the delayed scoring rule expression is:

[0041]

[0042] Where Y i It is the data delay corresponding to the visual positioning module.

[0043] On the basis of the above technical solution, preferably, the drone in step S3 includes a drone control module, which obtains relative position information of the landing target according to the optimal positioning source, and inputs the relative position information into the drone control module, so that the drone lands according to the relative position information, including the following sub-steps:

[0044] S31, obtaining relative position information of the landing target according to the optimal positioning source;

[0045] S32, using a PID controller to output the target speed of the X, Y, and Z axes of the drone, and sending it to the drone control module, so that the drone lands according to the relative position information;

[0046] The PID controller output formula is:

[0047] e X (t) = X target (t)-X drone (t)

[0048] e Y (t) = Y target (t)-Y drone (t)

[0049] e Z (t) = Z target (t)-Z drone (t)

[0050] v X (t) = KP X *e X (t)+KI X *∫e X (t)dt+KD X *de X (t) / dt

[0051] v Y (t) = KP Y *eY(t)+KI Y *∫e Y (t)dt+KD Y *de Y (t) / dt

[0052] v Z (t) = KP Z *eZ(t)+KI Z *∫e Z (t)dt+KD Z *de Z (t) / dt

[0053] Where, X target (t), Y target (t) and Z target (t) are the X, Y, and Z coordinates of the relative position of the landing target in space, drone (t), Y drone (t), Z drone (t) are the X, Y, and Z coordinates of the current position of the drone in space; e X (t), eY (t), e Z (t) are the distances between the target relative position and the current position of the drone along the X, Y, and Z axes, respectively, and KP X , KP Y , KP Z They are the proportional gains of the X, Y, and Z axes respectively; KI X KI Y KI Z They are the integral gains of the X, Y, and Z axes respectively; KD X , KD Y , KD Z are the differential gains of the X, Y, and Z axes respectively; v X (t), v Y (t), v Z (t) is the final output of the controlled UAV flight speed, and v X (t), v Y (t) The maximum speed does not exceed 10m / s, v Z (t) The maximum speed shall not exceed 2 m / s.

[0054] On the basis of the above technical solution, preferably, the step S32 also includes dynamically adjusting the PID values ​​of the X, Y, and Z axes of the drone so that the drone lands on a moving object, wherein:

[0055] Let the moving speed of the landing target on the X, Y, and Z axes be V respectively. X(target) 、V Y(target) and V Z(target) , the relationship between the moving speed of the landing target and the PID parameters is:

[0056] KP X =0.0072V X(target) 2 +0.0521V X(target) +0.2321

[0057] KP Y =0.0072V Y(target) 2 +0.0521V Y(target) +0.2321

[0058] KP Z =0.0130V Z(target) 3 -0.0901V Z(target) 2 +0.2376V Z(target) +0.1737

[0059] KI X =0.018*e0.2495VX(target) +0.000026

[0060] KI Y =0.018*e0.2495V Y(target) +0.000026

[0061] KI Z =0.036*e0.2883V Z(target) +0.000015

[0062] KD X =0.0000533V X(target) +0.0000167

[0063] KD Y =0.0000533V Y(target) +0.0000167

[0064] KD Z =0.0000533V Z(target) +0.0000167.

[0065] On the basis of the above technical scheme, preferably, step S4 is also included to monitor the flight status of the UAV during the landing process to provide safety protection for the UAV, wherein the UAV also includes a status monitoring module for real-time monitoring of the flight status of the UAV during the landing process, including speed, altitude, attitude and position information. If an abnormal state of the UAV is detected, the UAV is controlled to return or hover, and the abnormal state includes abnormal positioning of the UAV, loss of landing target information and landing timeout.

[0066] In a second aspect, the present invention further provides an unmanned landing system based on scene adaptation, which is implemented by the unmanned landing method based on scene adaptation as described above, and the system comprises:

[0067] The environment perception module is used for the drone to carry a sensor unit and several positioning sources. The sensor unit collects the surrounding environment information, and the several positioning sources respectively locate the relative position to obtain the positioning information of each positioning source;

[0068] The adaptive algorithm module is used to construct a positioning adjustment model, and obtain the optimal positioning source based on the positioning adjustment model calculation according to the surrounding environment information and positioning information;

[0069] The drone control module is used to obtain the relative position information of the landing target according to the optimal positioning source, and input the relative position information into the drone control module so that the drone can land according to the relative position information.

[0070] Compared with the prior art, the unmanned landing method and system based on scene adaptation of the present invention have the following advantages:

[0071] Beneficial effects:

[0072] (1) By comprehensively evaluating the positioning accuracy, scenarios, signal strength, and data delay of different positioning sources, and selecting the positioning source with the highest total score as the current optimal positioning source, the positioning accuracy of the UAV can be improved, which helps the UAV to more accurately determine its own position and reduce errors in key tasks such as landing. The overall reliability, stability, and adaptability of the UAV positioning system are enhanced, and it can achieve accurate and reliable autonomous landing in complex or dynamically changing environments.

[0073] (2) By dynamically adjusting the PID values ​​of the three axes of the drone, the drone can accurately land on a moving object. During the flight of the drone, the control system will obtain the moving speed of the landing target in real time, calculate the current optimal PID parameter value, and then apply the parameter value to the PID controller to output the target speed of the drone's X, Y, and Z axes, thereby controlling the drone to follow and land on the moving target, allowing the drone to respond to the landing task of the moving target more flexibly, improving the system's adaptability and control accuracy;

[0074] (3) By monitoring the flight status of the UAV during the landing process, the UAV can be protected and the reliability and safety of the flight can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0076] Figure 1 It is a flowchart of the unmanned landing method based on scene adaptation of the present invention;

[0077] Figure 2 It is a structural block diagram of the unmanned landing system based on scene adaptation of the present invention;

[0078] Figure 3 The present invention is a flow chart of adjustment rules for the positioning adjustment model of the scene-adaptive unmanned landing method. DETAILED DESCRIPTION

[0079] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0080] like Figure 1-3 As shown, an unmanned landing method based on scene adaptation of the present invention comprises the following steps:

[0081] S1, the UAV is equipped with a sensor unit and several positioning sources. The sensor unit collects surrounding environment information, and several positioning sources respectively locate the relative position to obtain the positioning information of each positioning source.

[0082] Among them, in step S1, the sensor unit includes a photosensitive sensor and a laser radar, and the positioning source includes a visual positioning module, an ALIGN-GPS positioning module and a UWB positioning module.

[0083] The photosensor is used to collect the illumination data of the surrounding environment and determine whether the current scene is indoors or outdoors; the lidar is used to collect the position of obstacles in the surrounding environment; the visual positioning module is used to capture images and videos of the surrounding environment, obtain the light intensity, positioning error and video frame rate of the surrounding environment; the ALIGN-GPS positioning module is used to collect the positioning error, signal strength and data delay of the relative positioning position; the UWB relative positioning module is used to collect the positioning error, signal strength and data delay of the relative positioning position.

[0084] It should be noted that the photosensitive sensor can distinguish whether the current scene is indoors or outdoors by measuring the light intensity of the surrounding environment. If it is an indoor scene, the light intensity is low and stable. If it is an outdoor scene, the light intensity varies greatly, providing a basis for the subsequent selection of a suitable landing positioning algorithm; the lidar uses a laser beam to scan the surrounding environment, construct a three-dimensional point cloud map, and accurately identify the position, shape and size of obstacles, providing an important basis for the drone's obstacle avoidance and path planning.

[0085] The visual positioning module in the positioning source not only captures images and videos of the surrounding environment, but also evaluates the light intensity by analyzing the image and video content, and verifies the positioning error and video frame rate with the photosensitive sensor data; the ALIGN-GPS positioning module provides satellite-based positioning information, including positioning error, signal strength and data delay; the UWB positioning module calculates the distance by measuring the flight time of radio waves between sending and receiving, thereby achieving high-precision relative positioning.

[0086] S2, constructing a positioning adjustment model, and obtaining the optimal positioning source based on the positioning adjustment model calculation according to the surrounding environment information and positioning information.

[0087] Step S2 includes the following sub-steps:

[0088] S21, preset accuracy scoring rules, match the positioning errors corresponding to the visual positioning module, the ALIGN-GPS positioning module and the UWB positioning module, and obtain corresponding positioning accuracy scores respectively;

[0089] S22, presetting a scenario scoring rule, matching the reliability of each positioning source according to the current scenario, and obtaining corresponding reliability scores respectively;

[0090] S23, preset a signal strength scoring rule, match the corresponding signal strengths of the ALIGN-GPS positioning module and the UWB positioning module, and obtain corresponding availability scores respectively; preset a light intensity scoring rule, match the light intensity of the visual positioning module, and obtain corresponding availability scores;

[0091] S24, preset a delay scoring rule, match the corresponding data delays of the ALIGN-GPS positioning module and the UWB positioning module, and obtain corresponding delay scores respectively; preset a frame rate scoring rule, match the video frame rate of the visual positioning module, and obtain corresponding delay scores;

[0092] S25, constructing a positioning adjustment model, and calculating a total score corresponding to each positioning source based on the positioning adjustment model according to the positioning accuracy score, the reliability score, the availability score, and the delay score;

[0093] S26, according to the total scores of the positioning sources, selecting the positioning source with the highest total score as the optimal positioning source for the current scene.

[0094] The calculation expression of the positioning adjustment model is:

[0095] S=W1*Q i +W2*M i +W3*N i +W4*L i

[0096] In the formula, S is the total score corresponding to the i-th positioning source, Q i is the positioning accuracy score corresponding to the i-th positioning source, M i is the reliability score corresponding to the i-th positioning source, N i is the availability score corresponding to the i-th positioning source, L i is the delay score corresponding to the ith positioning source, W1, W2, W3 and W4 are the weights corresponding to each score, W1+W2+W3+W4=1.

[0097] The accuracy scoring rule, signal strength scoring rule and delay scoring rule are all piecewise functions;

[0098] The accuracy scoring rule expression is:

[0099]

[0100] In the formula, J i is the positioning error value corresponding to the i-th positioning source;

[0101] Among them, the score is based on the error range of the positioning source. The smaller the error, the higher the score. The accuracy scoring rules correspond to Table 1:

[0102] Table 1

[0103]

[0104] If the photosensitive sensor detects that the location is outdoor, the corresponding reliability score of the ALIGN-GPS positioning module is 0.9, the corresponding reliability score of the UWB positioning module is 0.7, and the corresponding reliability score of the visual positioning module is 0.4. If the photosensitive sensor detects that the location is indoor, the corresponding reliability score of the ALIGN-GPS positioning module is 0.1, the corresponding reliability score of the UWB positioning module is 0.3, and the corresponding reliability score of the visual positioning module is 0.6.

[0105] Scoring is performed based on the stability of the positioning source in different environments. The scenario scoring rules correspond to Table 2. Table 2

[0106]

[0107] If the positioning source is the ALIGN-GPS positioning module or the UWB positioning module, the signal strength scoring rule expression is:

[0108]

[0109] Where, X i It is the signal strength corresponding to the ALIGN-GPS positioning module and the UWB positioning module.

[0110] If the positioning source is a visual positioning module, the signal strength scoring rule expression is:

[0111]

[0112] In the formula, G i is the light intensity corresponding to the visual positioning module.

[0113] Score based on the signal strength and availability of the positioning source in the current environment. The signal strength scoring rules correspond to Table 3:

[0114] Table 3

[0115]

[0116] If the positioning source is the ALIGN-GPS positioning module or the UWB positioning module, the delay scoring rule expression is:

[0117]

[0118] In the formula, C i The data delay corresponding to the ALIGN-GPS positioning module and the UWB positioning module;

[0119] If the positioning source is the visual positioning module, the delayed scoring rule expression is:

[0120]

[0121] Where Y i It is the data delay corresponding to the visual positioning module.

[0122] Scoring is based on the data transmission and processing delay of the positioning source. The lower the delay, the higher the score. The delay scoring rules correspond to Table 4:

[0123]

[0124] It should be noted that by conducting a comprehensive evaluation based on the positioning accuracy, scenarios, signal strength and data delay of different positioning sources, and selecting the positioning source with the highest total score as the current optimal positioning source, the positioning accuracy of the drone can be improved, which will help the drone to more accurately determine its own position and reduce errors in key tasks such as landing; it will enhance the overall reliability, stability and adaptability of the drone positioning system and enable it to cope with various complex and changing environments and tasks.

[0125] For example, suppose the drone is performing an indoor landing mission. At this time, the ALIGN-GPS positioning module fails due to signal blocking, the visual positioning module degrades due to insufficient light, and the UWB positioning module shows good stability and accuracy. Through the calculation of the positioning adjustment model, the drone will automatically select the UWB positioning module as the current optimal positioning source and perform precise landing operations based on it. The drone will be able to overcome environmental limitations and successfully complete the mission.

[0126] S3, the drone includes a drone control module, which obtains relative position information of the landing target according to the optimal positioning source, and inputs the relative position information into the drone control module, so that the drone lands according to the relative position information.

[0127] Wherein, step S3 includes the following sub-steps:

[0128] S31, obtaining relative position information of the landing target according to the optimal positioning source;

[0129] S32, using a PID controller to output the target speed of the X, Y, and Z axes of the drone, and sending it to the drone control module, so that the drone lands according to the relative position information;

[0130] The PID controller output formula is:

[0131] e X (t) = X target (t)-X drone (t)

[0132] e Y (t) = Y target (t)-Y drone (t)

[0133] e Z (t) = Z target (t)-Z drone (t)

[0134] v X (t) = KP X *e X (t)+KI X *∫e X (t)dt+KD X *de X (t) / dt

[0135] v Y (t) = KP Y *eY(t)+KI Y *∫e Y (t)dt+KD Y *de Y (t) / dt

[0136] v Z (t) = KP Z *eZ(t)+KI Z *∫e Z (t)dt+KD Z *de Z (t) / dt

[0137] Where, X target (t), Y target (t) and Z target (t) are the X, Y, and Z coordinates of the relative position of the landing target in space, drone (t), Ydrone (t), Z drone (t) are the X, Y, and Z coordinates of the current position of the drone in space; e X (t), e Y (t), e Z (t) are the distances between the target relative position and the current position of the drone along the X, Y, and Z axes, respectively, and KP X , KP Y , KP Z They are the proportional gains of the X, Y, and Z axes respectively; KI X KI Y KI Z They are the integral gains of the X, Y, and Z axes respectively; KD X , KD Y , KD Z are the differential gains of the X, Y, and Z axes respectively; v X (t), v Y (t), v Z (t) is the final output of the controlled UAV flight speed, and v X (t), v Y (t) The maximum speed does not exceed 10m / s, v Z (t) The maximum speed shall not exceed 2 m / s.

[0138] The relative position information of the landing target is obtained according to the selected positioning source, and then the target speed of the X, Y, and Z axes of the drone is output through the PID controller and sent to the drone control module, so that the drone continues to approach the landing target. When the distance to the landing target gets closer and closer, the target speed of the three axes of the drone will become smaller and smaller until the final precise landing is completed.

[0139] In addition, step S32 also includes dynamically adjusting the PID values ​​of the X, Y, and Z axes of the drone to make the drone land on a moving object, wherein:

[0140] Let the moving speed of the landing target on the X, Y, and Z axes be V respectively. X(target) 、V Y(target) and V Z(target) , the relationship between the moving speed of the landing target and the PID parameters is:

[0141] KP X =0.0072V X(target) 2 +0.0521V X(target) +0.2321

[0142] KP Y =0.0072V Y(target) 2 +0.0521VY(target) +0.2321

[0143] KP Z =0.0130V Z(target) 3 -0.0901V Z(target) 2 +0.2376V Z(target) +0.1737

[0144] KI X =0.018*e0.2495V X(target) +0.000026

[0145] KI Y =0.018*e0.2495V Y(target) +0.000026

[0146] KI Z =0.036*e0.2883V Z(target) +0.000015

[0147] KD X =0.0000533V X(target) +0.0000167

[0148] KD Y =0.0000533V Y(target) +0.0000167

[0149] KD Z =0.0000533V Z(target) +0.0000167.

[0150] It should be noted that the precise landing algorithm can also be used for precise landing of moving targets. By dynamically adjusting the PID values ​​of the three axes of the drone, the drone can accurately land on a moving object. During the flight of the drone, the control system will obtain the moving speed of the landing target in real time, and calculate the current optimal PID parameter value based on the above relationship expression. Then, the parameter value is applied to the PID controller to output the target speed of the drone's X, Y, and Z axes, thereby controlling the drone to follow and land on the moving target, allowing the drone to respond to the landing task of the moving target more flexibly and improve the system's adaptability and control accuracy.

[0151] The method also includes step S4, monitoring the flight status of the UAV during the landing process to provide safety protection for the UAV, wherein the UAV also includes a status monitoring module for real-time monitoring of the flight status of the UAV during the landing process, including speed, altitude, attitude and position information. If an abnormal state of the UAV is detected, the UAV is controlled to return or hover, and the abnormal state includes abnormal positioning of the UAV, loss of landing target information and landing timeout.

[0152] It should be noted that the status monitoring module monitors the current flight speed of the UAV to ensure that it is within a safe range. Too fast a speed may cause the UAV to lose control or collide. The UAV's altitude information is tracked in real time to prevent it from flying too low and touching the ground or flying too high and losing control. The pitch, yaw and roll angles of the UAV are checked to ensure that its attitude is stable and meets the landing requirements. The precise position of the UAV is obtained based on the optimal positioning source and compared with the landing target position to ensure that the UAV is on the correct route.

[0153] When the drone cannot accurately obtain its own position or the positioning signal is unstable, the system will consider the positioning abnormal. At this time, the drone should be controlled to enter a hovering state and try to reacquire the positioning information. If the correct positioning cannot be restored for a long time, consider returning home; if the drone cannot continuously receive the location information of the landing target, the system will consider that the target information is lost. At this time, the drone should suspend the landing operation and try to re-establish the information connection with the target. If it cannot be restored, it will decide whether to return home based on the current position and remaining power; the landing time threshold is preset. If the drone fails to complete the landing operation within the preset time, it is considered to have timed out. The drone chooses to return home or find a safe hovering position. The drone can achieve self-monitoring and safety protection during the landing process to improve the reliability and safety of the flight.

[0154] In a second aspect, the present invention further provides an unmanned landing system based on scene adaptation, which is implemented by the unmanned landing method based on scene adaptation as described above, and the system comprises:

[0155] The environment perception module is used for the drone to carry a sensor unit and several positioning sources. The sensor unit collects the surrounding environment information, and the several positioning sources respectively locate the relative position to obtain the positioning information of each positioning source;

[0156] The adaptive algorithm module is used to construct a positioning adjustment model, and obtain the optimal positioning source based on the positioning adjustment model calculation according to the surrounding environment information and positioning information;

[0157] The drone control module is used to obtain the relative position information of the landing target according to the optimal positioning source, and input the relative position information into the drone control module so that the drone can land according to the relative position information.

[0158] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An unmanned landing method based on scene adaptation, characterized in that: The method comprises the following steps: S1, the drone is equipped with a sensor unit and several positioning sources, the sensor unit collects surrounding environment information, and the several positioning sources respectively locate the relative positions to obtain the positioning information of each positioning source; the sensor unit includes a photosensitive sensor and a laser radar, and the several positioning sources include a visual positioning module, an ALIGN-GPS positioning module and a UWB positioning module; S2, constructing a positioning adjustment model, and calculating and obtaining the optimal positioning source based on the positioning adjustment model according to the surrounding environment information and positioning information; S21, preset accuracy scoring rules, match the positioning errors corresponding to the visual positioning module, the ALIGN-GPS positioning module and the UWB positioning module, and obtain corresponding positioning accuracy scores respectively; S22, presetting a scenario scoring rule, matching the reliability of each positioning source according to the current scenario, and obtaining corresponding reliability scores respectively; S23, preset a signal strength scoring rule, match the corresponding signal strengths of the ALIGN-GPS positioning module and the UWB positioning module, and obtain corresponding availability scores respectively; preset a light intensity scoring rule, match the light intensity of the visual positioning module, and obtain corresponding availability scores; S24, preset a delay scoring rule, match the corresponding data delays of the ALIGN-GPS positioning module and the UWB positioning module, and obtain corresponding delay scores respectively; preset a frame rate scoring rule, match the video frame rate of the visual positioning module, and obtain corresponding delay scores; S25, constructing a positioning adjustment model, and calculating a total score corresponding to each positioning source based on the positioning adjustment model according to the positioning accuracy score, the reliability score, the availability score, and the delay score; S26, according to the total scores of the positioning sources, selecting the positioning source with the highest total score as the optimal positioning source for the current scene; S3, the drone includes a drone control module, which obtains relative position information of the landing target according to the optimal positioning source, and inputs the relative position information into the drone control module, so that the drone lands according to the relative position information; S31, obtaining relative position information of the landing target according to the optimal positioning source; S32, using the PID controller to output the target speed of the X, Y, and Z axes of the drone, and sending it to the drone control module, so that the drone can land according to the relative position information.

2. The scene-adaptive unmanned landing method according to claim 1, characterized in that: The light sensor is used to collect the illumination data of the surrounding environment and determine whether the current scene is indoors or outdoors; The laser radar is used to collect and obtain the positions of obstacles in the surrounding environment; The visual positioning module is used to capture images and videos of the surrounding environment, and obtain the light intensity, positioning error and video frame rate of the surrounding environment; The ALIGN-GPS positioning module is used to collect positioning error, signal strength and data delay of relative positioning position; The UWB relative positioning module is used to collect positioning error, signal strength and data delay of the relative positioning position.

3. The scene-adaptive unmanned landing method according to claim 1, characterized in that: The calculation expression of the positioning adjustment model is: S=W1*Q i +W2*M i +W3*N i +W4*L i In the formula, S is the total score corresponding to the i-th positioning source, Q i is the positioning accuracy score corresponding to the i-th positioning source, M i is the reliability score corresponding to the i-th positioning source, N i is the availability score corresponding to the i-th positioning source, L i is the delay score corresponding to the ith positioning source, W1, W2, W3 and W4 are the weights corresponding to each score, W1+W2+W3+W4=1.

4. The scene-adaptive unmanned landing method according to claim 3, characterized in that: The accuracy scoring rule, signal strength scoring rule and delay scoring rule are all piecewise functions; The accuracy scoring rule expression is: In the formula, J i is the positioning error value corresponding to the i-th positioning source; If the photosensitive sensor detects that the location is outdoor, the corresponding reliability score of the ALIGN-GPS positioning module is 0.9, the corresponding reliability score of the UWB positioning module is 0.7, and the corresponding reliability score of the visual positioning module is 0.

4. If the photosensitive sensor detects that the location is indoor, the corresponding reliability score of the ALIGN-GPS positioning module is 0.1, the corresponding reliability score of the UWB positioning module is 0.3, and the corresponding reliability score of the visual positioning module is 0.

6. If the positioning source is the ALIGN-GPS positioning module or the UWB positioning module, the signal strength scoring rule expression is: In the formula, X i The signal strength corresponding to the ALIGN-GPS positioning module and the UWB positioning module; If the positioning source is a visual positioning module, the signal strength scoring rule expression is: In the formula, G i is the light intensity corresponding to the visual positioning module; If the positioning source is the ALIGN-GPS positioning module or the UWB positioning module, the delay scoring rule expression is: In the formula, C i The data delay corresponding to the ALIGN-GPS positioning module and the UWB positioning module; If the positioning source is the visual positioning module, the delayed scoring rule expression is: Where Y i It is the data delay corresponding to the visual positioning module.

5. The scene-adaptive unmanned landing method according to claim 1, characterized in that: The output formula of the PID controller in step S3 is: e X (t)=X target (t)-X drone (t) e Y (t)=Y target (t)-Y drone (t) e Z (t)=Z target (t)-Z drone (t) v X (t)=KP X *e X (t)+KI X *∫e X (t)dt+KD X *de X (t) / dt v Y (t)=KP Y *eY(t)+KI Y *∫e Y (t)dt+KD Y *de Y (t) / dt v Z (t)=KP Z *eZ(t)+KI Z *∫e Z (t)dt+KD Z *de Z (t) / dt In the formula, X target (t), Y target (t) and Z target (t) are the X, Y, and Z coordinates of the relative position of the landing target in space, drone (t), Y drone (t), Z drone (t) are the X, Y, and Z coordinates of the current position of the drone in space; e X (t), e Y (t), e Z (t) are the distances between the target relative position and the current position of the drone along the X, Y, and Z axes, respectively, and KP X , KP Y , KP Z They are the proportional gains of the X, Y, and Z axes respectively; KI X KI Y KI Z They are the integral gains of the X, Y, and Z axes respectively; KD X , KD Y , KD Z are the differential gains of the X, Y, and Z axes respectively; v X (t), v Y (t), v Z (t) is the final output of the controlled UAV flight speed, and v X (t), v Y (t) The maximum speed does not exceed 10m / s, v Z (t) The maximum speed shall not exceed 2 m / s.

6. The scene-adaptive unmanned landing method according to claim 5, characterized in that: Step S32 also includes dynamically adjusting the PID values ​​of the X, Y, and Z axes of the drone to land the drone on a moving object, wherein: Let the moving speed of the landing target on the X, Y, and Z axes be V respectively. X(target) 、V Y(target) and V Z(target) , the relationship between the moving speed of the landing target and the PID parameters is: KP X =0.0072V X(target) 2 +0.0521V X(target) +0.2321 KP Y =0.0072V Y(target) 2 +0.0521V Y(target) +0.2321 KP Z =0.0130V Z(target) 3 -0.0901V Z(target) 2 +0.2376V Z(target) +0.1737 KI X =0.018*e0.2495V X(target) +0.000026 KI Y =0.018*e0.2495V Y(target) +0.000026 KI Z =0.036*e0.2883V Z(target) +0.000015 KD X =0.0000533V X(target) +0.0000167 KD Y =0.0000533V Y(target) +0.0000167 KD Z =0.0000533V Z(target) +0.0000167。 7. The scene-adaptive unmanned landing method according to claim 1, characterized in that: The method also includes step S4, monitoring the flight status of the UAV during the landing process to provide safety protection for the UAV, wherein the UAV also includes a status monitoring module for real-time monitoring of the flight status of the UAV during the landing process, including speed, altitude, attitude and position information. If an abnormal state of the UAV is detected, the UAV is controlled to return or hover, and the abnormal state includes abnormal positioning of the UAV, loss of landing target information and landing timeout.

8. An unmanned landing system based on scene adaptation, implemented by the unmanned landing method based on scene adaptation as claimed in any one of claims 1 to 7, characterized in that: The system comprises: The environment perception module is used for the drone to carry a sensor unit and several positioning sources. The sensor unit collects the surrounding environment information, and the several positioning sources respectively locate the relative position to obtain the positioning information of each positioning source; The adaptive algorithm module is used to construct a positioning adjustment model, and obtain the optimal positioning source based on the positioning adjustment model calculation according to the surrounding environment information and positioning information; The drone control module is used to obtain the relative position information of the landing target according to the optimal positioning source, and input the relative position information into the drone control module so that the drone can land according to the relative position information.

Images