A method and device for planning a landing trajectory of a carrier-based aircraft in dynamic docking

Abstract

The application discloses a landing trajectory planning method and device for dynamic docking of a carrier-based aircraft, and aims to solve the technical problem that the landing accuracy of a carrier-based sub-aircraft cannot be ensured due to the existing landing trajectory planning method for dynamic docking of the carrier-based aircraft. The method comprises the following steps: obtaining a scalar distance between a sub-aircraft and a parent aircraft; when the distance between the two aircrafts exceeds a preset allowable error and a preset switching threshold, a docking preparation point is calculated by using a geometric positioning method, the sub-aircraft is quickly guided to return to the airspace near the parent aircraft, the positioning accuracy is further improved by using an iterative positioning method, the sub-aircraft is accurately returned to the upper side of the parent aircraft, and finally, a target landing trajectory corresponding to the carrier-based sub-aircraft is generated by combining a preset constraint condition, a preset trajectory optimization problem and a preset quasi-Newton optimizer according to the flight state data of the carrier-based sub-aircraft at the current time.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned aerial vehicles, and particularly relates to a landing trajectory planning method and device for dynamic docking of a carrier-based aircraft. BACKGROUND

[0002] With the increasing maturity of unmanned aerial vehicle technology, the potential of carrier-based aircraft as an aerial vehicle gradually emerges. These carrier-based aircraft can transport multiple small unmanned aerial vehicles to a designated target area to perform tasks such as monitoring, package delivery, and search and rescue.

[0003] After the completion of the task, the small unmanned aerial vehicle can autonomously return to the mother aircraft, which makes up for the deficiency of its limited operating range. In this process, the trajectory planning technology for the carrier-based aircraft to guide the unmanned aerial vehicle to safely return and land becomes the key. However, most current researches focus on static ground nest docking, which usually relies on a two-dimensional code or other obvious markers for visual positioning and uses visual target recognition to obtain the relative pose during the return process. These nests are generally fixed and therefore less difficult.

[0004] The existing trajectory planning method for dynamic docking and landing of a carrier-based aircraft relies on GPS positioning or visual positioning to achieve docking of the unmanned aerial vehicle. However, in an outdoor environment, changes in light can interfere with the accurate identification of images by the visual positioning system, causing the obtained relative position information of the sub-aircraft to deviate. On the other hand, terrain obstructions not only hinder the reception of GPS signals, affecting the accuracy of positioning, but also further exacerbate the difficulty of visual positioning. Under the influence of these adverse factors, both GPS positioning and visual positioning will produce large errors. Moreover, as the distance between the sub-aircraft and the mother aircraft increases, the visual positioning error will gradually accumulate. This series of problems will gradually reduce the positioning accuracy, making it difficult to ensure the landing accuracy of the carrier-based sub-aircraft. SUMMARY

[0005] The present application provides a landing trajectory planning method and device for dynamic docking of a carrier-based aircraft, which solves the technical problem that the existing trajectory planning method for dynamic docking and landing of a carrier-based aircraft makes it difficult to ensure the landing accuracy of the carrier-based sub-aircraft.

[0006] The present application provides a landing trajectory planning method and device for dynamic docking of a carrier-based aircraft, which solves the technical problem that the existing trajectory planning method for dynamic docking and landing of a carrier-based aircraft makes it difficult to ensure the landing accuracy of the carrier-based sub-aircraft.

[0007] The landing trajectory planning method for dynamic docking of a carrier-based aircraft provided by the present application comprises:

[0008] If the scalar distance is greater than the preset allowable error, the scalar distance and the preset switching distance are compared.

[0009] if the scalar distance is greater than the preset switching distance, based on a geometric positioning method, using the inertial coordinate reference position of the carrier-based sub-machine, a plurality of waypoints corresponding to the carrier-based sub-machine, and the position difference distance between the carrier-based sub-machine at each waypoint and the carrier-based mother machine, the inertial coordinate position of the docking preparation point is calculated;

[0010] The inertial coordinate real-time position of the carrier-based sub-machine is updated according to the inertial coordinate position of the docking preparation point using a proportional differential controller, the intermediate inertial coordinate real-time position of the carrier-based sub-machine is determined, and the difference between the intermediate inertial coordinate real-time position of the carrier-based sub-machine and the inertial coordinate position of the docking preparation point is calculated. The difference between the docking point distance of the sub-machine is judged whether it reaches the preset switching distance;

[0011] If it is reached, based on the iterative positioning method, the flight influence data of the carrier-based sub-machine, the inertial coordinate position of the docking preparation point, and the intermediate inertial coordinate real-time position of the carrier-based sub-machine are used to update the scalar distance, the updated scalar distance is determined, and it is judged whether the updated scalar distance reaches the preset distance threshold;

[0012] If it is reached, based on the preset constraint condition, the preset trajectory optimization problem and the preset quasi-Newton optimizer are used to plan the trajectory according to the current time flight state data of the carrier-based sub-machine, and the target landing trajectory corresponding to the carrier-based sub-machine is generated.

[0013] Optionally, the inertial coordinate position of the docking preparation point is calculated based on the geometric positioning method, using the inertial coordinate reference position of the carrier-based sub-machine, a plurality of waypoints corresponding to the carrier-based sub-machine, and the position difference distance between the carrier-based sub-machine at each waypoint and the carrier-based mother machine, comprising:

[0014] According to the plurality of waypoints corresponding to the carrier-based sub-machine and the position difference distance between the carrier-based sub-machine at each waypoint and the carrier-based mother machine, the local coordinate position of the carrier-based mother machine is calculated;

[0015] The local coordinate position of the carrier-based mother machine and the inertial coordinate reference position of the carrier-based sub-machine are added to determine the inertial coordinate position of the carrier-based mother machine relative to the carrier-based sub-machine;

[0016] The inertial coordinate position of the docking preparation point is calculated using the inertial coordinate position of the carrier-based mother machine relative to the carrier-based sub-machine and the local coordinate position of the carrier-based mother machine.

[0017] Optionally, the inertial coordinate real-time position of the carrier-based sub-machine is updated according to the inertial coordinate position of the docking preparation point using a proportional differential controller, and the intermediate inertial coordinate real-time position of the carrier-based sub-machine is determined, comprising:

[0018] determining an error term by subtracting the inertial coordinate position of the docking preparation point from the inertial coordinate real-time position of the carrier-borne sub-machine;

[0019] outputting a speed control signal according to the error term by using the proportional differential controller;

[0020] updating the inertial coordinate real-time position of the carrier-borne sub-machine based on the speed control signal to determine an intermediate inertial coordinate real-time position of the carrier-borne sub-machine.

[0021] Optionally, the flight influence data comprises a disturbance and a two-dimensional vector; and updating the scalar distance based on the iterative positioning method, using the flight influence data of the carrier-borne sub-machine, the inertial coordinate position of the docking preparation point, and the intermediate inertial coordinate real-time position of the carrier-borne sub-machine, to determine an updated scalar distance, comprises:

[0022] initializing the two-dimensional vector to determine an initial two-dimensional vector;

[0023] updating the two-dimensional vector based on the initial two-dimensional vector to determine an updated two-dimensional vector;

[0024] updating the disturbance according to the updated two-dimensional vector to determine an updated disturbance;

[0025] calculating the position of the docking preparation point relative to the carrier-borne sub-machine using the inertial coordinate position of the docking preparation point and the intermediate inertial coordinate real-time position of the carrier-borne sub-machine;

[0026] calculating a relative position according to the position of the docking preparation point relative to the carrier-borne sub-machine using a preset relative position adaptive estimator;

[0027] calculating a carrier-borne sub-machine control flight speed according to the relative position and the updated disturbance using a preset bounded motion controller;

[0028] updating the intermediate inertial coordinate real-time position of the carrier-borne sub-machine based on the carrier-borne sub-machine control flight speed to determine a target inertial coordinate real-time position of the carrier-borne sub-machine;

[0029] updating the scalar distance according to the target inertial coordinate real-time position of the carrier-borne sub-machine to determine an updated scalar distance.

[0030] Optionally, the method further comprises:

[0031] initializing the inertial coordinate position of the docking preparation point to determine an initial inertial coordinate position of the docking preparation point if the scalar distance is less than or equal to the preset switching distance;

[0032] initializing the two-dimensional vector to determine an initial two-dimensional vector;

[0033] updating the two-dimensional vector based on the initial two-dimensional vector, to determine an updated two-dimensional vector;

[0034] updating the applied disturbance according to the updated two-dimensional vector, to determine an updated applied disturbance;

[0035] calculating a position of the docking preparation point relative to the carrier-borne sub-machine using the initial docking preparation point inertial coordinate position and the real-time inertial coordinate position of the carrier-borne sub-machine;

[0036] calculating a relative position according to the position of the docking preparation point relative to the carrier-borne sub-machine using a preset relative position adaptive estimator;

[0037] calculating a carrier-borne sub-machine control flight speed according to the relative position and the updated applied disturbance using a preset bounded motion controller;

[0038] updating the real-time inertial coordinate position of the carrier-borne sub-machine based on the carrier-borne sub-machine control flight speed, to determine a target real-time inertial coordinate position of the carrier-borne sub-machine;

[0039] updating the scalar distance according to the target real-time inertial coordinate position of the carrier-borne sub-machine, to determine an updated scalar distance.

[0040] Optionally, the trajectory planning for the carrier-borne sub-machine according to the current time flight state data of the carrier-borne sub-machine based on the preset constraint condition, the preset trajectory optimization problem and the preset quasi-Newton optimizer to generate the target landing trajectory corresponding to the carrier-borne sub-machine comprises:

[0041] transforming the constraint penalty function corresponding to the preset constraint condition and the preset trajectory optimization problem through integral method and nonlinear transformation method to determine an unconstrained optimization problem;

[0042] inputting the unconstrained optimization problem and the preset constraint condition into a preset quasi-Newton optimizer for calculation to generate the target landing trajectory corresponding to the carrier-borne sub-machine.

[0043] The second aspect of the present application provides a landing trajectory planning device for carrier-borne dynamic docking, comprising:

[0044] an acquisition module configured to acquire a scalar distance between a carrier-borne sub-machine and a carrier-borne mother machine, and compare the scalar distance with a preset allowable error;

[0045] a comparison module configured to compare the scalar distance with a preset switching distance if the scalar distance is greater than the preset allowable error;

[0046] a geometric positioning module, configured to, if the scalar distance is greater than the preset switching distance, calculate an inertial coordinate position of a docking preparation point based on a geometric positioning method, using a reference position of the carrier-based sub-machine in inertial coordinates, a plurality of waypoints corresponding to the carrier-based sub-machine, and a position difference distance between each of the waypoints and the carrier-based mother machine;

[0047] a position updating module, configured to update a real-time position of the carrier-based sub-machine in inertial coordinates according to the inertial coordinate position of the docking preparation point using a proportional differential controller, determine an intermediate real-time position of the carrier-based sub-machine in inertial coordinates, and calculate a difference between a sub-machine docking point distance between the intermediate real-time position of the carrier-based sub-machine in inertial coordinates and the inertial coordinate position of the docking preparation point, and determine whether the difference between the sub-machine docking point distance reaches the preset switching distance;

[0048] an iterative positioning module, configured to, if the difference between the sub-machine docking point distance reaches the preset switching distance, update the scalar distance based on an iterative positioning method, using flight influence data of the carrier-based sub-machine, the inertial coordinate position of the docking preparation point, and the intermediate real-time position of the carrier-based sub-machine in inertial coordinates, determine an updated scalar distance, and determine whether the updated scalar distance reaches a preset distance threshold;

[0049] a trajectory planning module, configured to, if the difference between the sub-machine docking point distance reaches the preset switching distance, perform trajectory planning based on a preset constraint condition, using a preset trajectory optimization problem and a preset quasi-Newton optimizer according to current flight state data of the carrier-based sub-machine, and generate a target landing trajectory corresponding to the carrier-based sub-machine.

[0050] The third aspect of the present application provides a computer device, including a memory and a processor, the memory stores a computer program, and the computer program is executed by the processor to make the processor execute the steps of the landing trajectory planning method for dynamic docking of a carrier-based aircraft according to any one of the above aspects.

[0051] The fourth aspect of the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed to implement the steps of the landing trajectory planning method for dynamic docking of a carrier-based aircraft according to any one of the above aspects.

[0052] The fifth aspect of the present application provides a computer program product, which includes a computer program stored on a non-transitory computer readable storage medium, and the computer program includes program instructions, wherein when the program instructions are executed by a computer, the computer executes the steps of the landing trajectory planning method for dynamic docking of a carrier-based aircraft according to any one of the above aspects.

[0053] From the above technical solutions, the present application has the following advantages:

[0054] The scheme provides a landing trajectory planning method for dynamic docking of a carrier-based aircraft. BRIEF DESCRIPTION OF DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments or the prior art description will be briefly introduced as follows. Obviously, the drawings in the following description only show some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort on the basis of these drawings.

[0056] Figure 1 A step flow chart of a landing trajectory planning method for dynamic docking of a carrier-based aircraft provided by the embodiment one of the present application;

[0057] Figure 2 A trajectory diagram for guiding back to a docking point under a noisy condition only by using an iterative positioning method provided by the embodiment one of the present application;

[0058] Figure 3 A schematic diagram of relative position deviation of the iterative positioning method provided for the first embodiment of the present application under a noisy condition;

[0059] Figure 4 An enlarged schematic diagram of a graph of relative position deviation of the iterative positioning method provided for the first embodiment of the present application under a noisy condition;

[0060] Figure 5 A trajectory graph of guiding back to a docking point of the guiding positioning algorithm (geometric positioning method and iterative positioning method) provided for the first embodiment of the present application under a noisy condition;

[0061] Figure 6 A schematic diagram of relative position deviation of the guiding positioning algorithm provided for the first embodiment of the present application under a noisy condition;

[0062] Figure 7 An enlarged schematic diagram of a graph of relative position deviation of the guiding positioning algorithm provided for the first embodiment of the present application under a noisy condition;

[0063] Figure 8 A schematic diagram of a smooth approximation function varying with a parameter provided for the first embodiment of the present application;

[0064] Figure 9 A schematic diagram of a docking of a mother machine and a child machine in a simulation environment provided for the first embodiment of the present application;

[0065] Figure 10 A schematic diagram of the mother machine in a camera view of the child machine in a landing process provided for the first embodiment of the present application;

[0066] Figure 11 A schematic diagram of a landing trajectory generated by using the trajectory planning algorithm of the present application in a ROS+gazebo environment provided for the first embodiment of the present application;

[0067] Figure 12 A flowchart of guiding positioning of a ship-borne child machine provided for the first embodiment of the present application;

[0068] Figure 13 A schematic diagram of a running flowchart of a ship-borne machine overall system provided for the second embodiment of the present application;

[0069] Figure 14 A structural block diagram of a landing trajectory planning device for ship-borne machine dynamic docking provided for the third embodiment of the present application. DETAILED DESCRIPTION

[0070] The embodiments of the present application provide a landing trajectory planning method and device for ship-borne machine dynamic docking, and solve the technical problem that landing precision of a ship-borne child machine is difficult to guarantee caused by an existing landing trajectory planning method for ship-borne machine dynamic docking.

[0071] In order to make the application purposes, features and advantages of the present application more obvious and easy to understand, the technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings of the embodiments of the present application. Obviously, the following described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0072] Please refer to Figure 1 , Figure 1 A step flow chart of a landing trajectory planning method for a ship-borne dynamic docking is provided for Embodiment One of the present application.

[0073] The landing trajectory planning method for a ship-borne dynamic docking provided by the present application comprises:

[0074] Step 101: Obtain the scalar distance between the ship-borne child machine and the ship-borne parent machine, and compare the scalar distance with the preset allowable error.

[0075] It should be noted that, due to the influence of sensor noise, the estimated value obtained by using the iterative positioning method when the ship-borne child machine is far away from the docking preparation point may have a large error, therefore, in order to improve the efficiency of the long-range guidance task, the geometric positioning method is used to calculate the rough position of the docking point in the initial stage, and when the distance between the child machine and the parent machine reaches the threshold value, i.e., the distance difference between the child machine and the parent machine reaches the preset switching distance, the iterative method is switched to for more accurate estimation and guidance to the docking point (inertial coordinate position of the docking preparation point).

[0076] Further, the distance between the child machine (ship-borne child machine) and the parent machine (ship-borne parent machine) is obtained by using a UWB sensor (Ultra - Wideband) (the scalar distance), and the displacement of the child machine in the nearest time interval is calculated based on a VIO module (Visual - Inertial Odometry) , the scalar distance and the preset allowable error d a are compared, wherein the preset allowable error, the preset switching distance, etc. can be set as needed, and the present application does not make specific limitation thereon.

[0077] Step 102: If the scalar distance is greater than the preset allowable error, the scalar distance and the preset switching distance are compared.

[0078] It should be noted that if the scalar distance is less than or equal to the preset allowable error, the remote guidance algorithm is exited, and the sub-machine is directly entered into the landing stage. The current flight state data of the carrier sub-machine when the scalar distance is less than or equal to the preset allowable error is obtained. Based on the preset constraint condition, the preset trajectory optimization problem and the preset quasi-Newton optimizer are used to perform trajectory planning according to the current flight state data to generate the target landing trajectory corresponding to the carrier sub-machine.

[0079] In step 103, if the scalar distance is greater than the preset switching distance, the inertial coordinate position of the docking preparation point is calculated based on the geometric positioning method, using the inertial coordinate reference position of the carrier sub-machine, the plurality of waypoints corresponding to the carrier sub-machine, and the position difference distance between the carrier sub-machine at each waypoint and the carrier mother machine.

[0080] It should be noted that if the scalar distance is less than or equal to the preset allowable error, the remote guidance algorithm is exited, and the sub-machine is directly entered into the landing stage. The current flight state data of the carrier sub-machine when the scalar distance is less than or equal to the preset allowable error is obtained. Based on the preset constraint condition, the preset trajectory optimization problem and the preset quasi-Newton optimizer are used to perform trajectory planning according to the current flight state data to generate the target landing trajectory corresponding to the carrier sub-machine. greater than the preset switching distance d c , a local coordinate system is established using the inertial coordinate reference position of the sub-machine, and 4 waypoints are designed with reference to the regular tetrahedron. After reaching each waypoint, hover for 1 second to obtain the position p i of the waypoint in the local coordinate system, and measure the distance to the mother machine to obtain the average distance d i . Then, the geometric positioning method is used to calculate the position p d of the docking preparation point in the inertial coordinate system.

[0081] Specifically, step 103 can include the following sub-steps S31-S33:

[0082] Step S31, calculating the local coordinate position of the carrier mother machine according to the plurality of waypoints corresponding to the carrier sub-machine and the position difference distance between the carrier sub-machine at each waypoint and the carrier mother machine;

[0083] Step S32, adding the local coordinate position of the carrier mother machine and the inertial coordinate reference position of the carrier sub-machine to determine the inertial coordinate position of the carrier mother machine relative to the carrier sub-machine;

[0084] Step S33, calculating the inertial coordinate position of the docking preparation point using the inertial coordinate position of the carrier mother machine relative to the carrier sub-machine and the local coordinate position of the carrier mother machine.

[0085] The waypoint is the target waypoint of the carrier sub-machine in the local coordinate system, and the plurality of waypoints includes a first waypoint, a second waypoint, a third waypoint, and a fourth waypoint.

[0086] It should be noted that according to the spatial four-point positioning method, if the positions of four points in Cartesian space that are not in the same plane are known, as well as their distances to a certain point, the position of the point can be calculated. Let the reference position of the sub-machine in the inertial coordinate system (the inertial coordinate reference position of the carrier sub-machine) be , is the horizontal coordinate of the reference position of the carrier-borne sub-vehicle in the inertial coordinate system, is the longitudinal coordinate of the reference position of the carrier-borne sub-vehicle in the inertial coordinate system, is the vertical coordinate of the reference position of the carrier-borne sub-vehicle in the inertial coordinate system.

[0087] Further, the position of the carrier-borne mother-vehicle relative to the carrier-borne sub-vehicle (the inertial coordinate position of the carrier-borne mother-vehicle relative to the carrier-borne sub-vehicle) is calculated by VIO (Visual Positioning Module) as . Specifically, a local coordinate system with the same attitude as the inertial coordinate system is established with as the origin. Next, the corresponding four non-coplanar waypoints of the sub-vehicle can be referred to as a regular tetrahedron shape. The positions of the waypoints in the local coordinate system are described as p i , specifically:

[0088] , , , , ;

[0089] wherein, is the first waypoint; is the second waypoint; is the third waypoint; is the fourth waypoint; is the distance moved by the carrier-borne sub-vehicle at the second waypoint in the local coordinate system; is the distance moved by the carrier-borne sub-vehicle at the third waypoint in the local coordinate system; is the distance moved by the carrier-borne sub-vehicle at the fourth waypoint in the local coordinate system; is the distance moved by the carrier-borne sub-vehicle at the i-th waypoint in the local coordinate system; is a constant greater than 1.

[0090] It is worth mentioning that the carrier-borne sub-vehicle can select a suitable l i based on the obstacle information of the surrounding environment, so as to obtain a collision-free waypoint, , is the horizontal coordinate of the i-th waypoint, is the longitudinal coordinate of the i-th waypoint, is the vertical coordinate of the i-th waypoint.

[0091] Let the position of the mother-vehicle in the local coordinate system be , is the horizontal coordinate of the local coordinate position of the carrier-borne mother-vehicle, is the longitudinal coordinate of the local coordinate position of the carrier-borne mother-vehicle, The vertical coordinate of the local coordinate position of the carrier-based mother aircraft. The slave aircraft can directly obtain the distance of the mother aircraft using UWB at each waypoint, and the average value of the distance is obtained by sampling for a certain period of time, that is, the position difference distance d between the waypoint and the carrier-based mother aircraft. i , according to the distance formula:

[0092] ;

[0093] in, is the position difference between the i-th waypoint and the carrier-based aircraft; is the i-th waypoint; is the local coordinate position of the carrier-based aircraft; is the horizontal coordinate of the i-th waypoint; is the horizontal coordinate of the local coordinate position of the carrier-based aircraft; is the ordinate of the i-th waypoint; is the ordinate of the local coordinate position of the carrier-based aircraft; is the vertical coordinate of the i-th waypoint; is the vertical coordinate of the local coordinate position of the carrier-based aircraft.

[0094] After simplifying the above formula, we can get:

[0095] ;

[0096] remember , we can get the difference between the formulas:

[0097] ;

[0098] in, for and the difference between the distances to the origin; for and The difference between the distances to the origin.

[0099] Convert the formula into matrix expression:

[0100] ;

[0101] ;

[0102] ;

[0103] ;

[0104] Among them, A is the position constraint matrix, and each row of the matrix represents and the xyz coordinate difference between them; b is an error matrix, each element of b represents the relative deviation in error between the measured distance and the known point, i.e. the difference between the distances to the origin.

[0105] Further, when the four points of the child machine are not on the same plane, the matrix A is invertible. Therefore, the position of the parent machine in the local coordinate system (the local coordinate position of the carrier parent machine) can be obtained by solving the formula , and then converting to the child machine inertial coordinate system, i.e. , to obtain the inertial coordinate position of the carrier parent machine relative to the carrier child machine.

[0106] Further, due to the existence of UWB sensor noise, the estimated value decreases in accuracy when the distance is long, so the estimated value is comprehensively considered together with the initial position of the parent machine calculated by VIO , combined with the reference position of the child machine in the inertial coordinate system , and the position p of the docking preparation point relative to the parent machine set in advance dm , to obtain the position (inertial coordinate position of the docking preparation point) p of the docking preparation point in the inertial coordinate system. d , wherein the weight satisfies Since the initial position of the parent machine obtained by VIO is measured when the child machine takes off, and the parent machine is likely to have a positioning deviation when it hovers at the docking point during the subsequent execution of the task by the child machine, the weight should be adjusted to a larger value; wherein the calculation formula of the inertial coordinate position of the docking preparation point is:

[0107] ;

[0108] wherein is the inertial coordinate position of the docking preparation point; is the weight.

[0109] Step 104, using a proportional-derivative controller to update the inertial coordinate real-time position of the carrier child machine according to the inertial coordinate position of the docking preparation point, determine the intermediate inertial coordinate real-time position of the carrier child machine, and calculate the difference between the intermediate inertial coordinate real-time position of the carrier child machine and the inertial coordinate position of the docking preparation point. The docking point distance difference of the child machine is judged whether the docking point distance difference of the child machine reaches the preset switching distance.

[0110] Specifically, step 104 can include the following sub-steps S41-S43:

[0111] ​​Step S41, subtracting the inertial coordinate position of the docking preparation point from the inertial coordinate real-time position of the ship-borne sub-machine to determine an error term;

[0112] Step S42, outputting a speed control signal according to the error term by using a proportional differential controller;

[0113] Step S43, updating the inertial coordinate real-time position of the ship-borne sub-machine based on the speed control signal to determine an intermediate inertial coordinate real-time position of the ship-borne sub-machine.

[0114] It should be noted that after obtaining the target docking position (inertial coordinate position of the docking preparation point) based on the above steps, the present application uses a PD controller (Proportional-Derivative Controller) to control the sub-machine to move towards the inertial coordinate position of the docking preparation point.

[0115] Specifically, during the movement, the inertial coordinate real-time position of the ship-borne sub-machine is , the target position is p d , the error term is , and the PD controller outputs a target speed (speed control signal) , wherein is the horizontal coordinate of the inertial coordinate real-time position of the ship-borne sub-machine, is the vertical coordinate of the inertial coordinate real-time position of the ship-borne sub-machine, is the vertical coordinate of the inertial coordinate real-time position of the ship-borne sub-machine, is the proportional gain, is the differential gain, is the error term at time k.

[0116] Further, the sub-machine uses the control signal to adjust its speed and move towards the target position, thereby updating the current position to obtain the intermediate inertial coordinate real-time position of the ship-borne sub-machine, and the distance between the sub-machine and the docking point (the difference between the intermediate inertial coordinate real-time position of the ship-borne sub-machine and the inertial coordinate position of the docking preparation point) is When the difference between the sub-machine docking point distance does not reach the switching distance, the intermediate inertial coordinate real-time position of the ship-borne sub-machine is taken as the new inertial coordinate real-time position of the ship-borne sub-machine, and the step S41 is executed, until the difference between the sub-machine docking point distance reaches the preset switching distance, and then switching to the iterative method to more accurately estimate and guide to the docking point (inertial coordinate position of the docking preparation point).

[0117] It is worth mentioning that the PD controller calculates the error term at the current time, and generates a speed control signal according to the proportional and differential gain, and the control signal is used by the sub-machine to adjust the speed thereof, move towards the target position, update the current position, and enter the next iteration, and when the preset switching distance is reached, the next stage is entered, and the iterative method is used for accurate docking.

[0118] Step 105, if reached, based on the iterative positioning method, using the flight influence data of the ship-borne sub-machine, the inertial coordinate position of the docking preparation point, the intermediate inertial coordinate real-time position of the ship-borne sub-machine, updating the scalar distance, determining the updated scalar distance, and judging whether the updated scalar distance reaches the preset distance threshold.

[0119] The flight influence data includes a disturbance and a two-dimensional vector.

[0120] It should be noted that when is greater than the preset allowable error, and the difference between the docking point distances of the sub-machines reaches the preset switching distance, firstly, the and parameters are updated, and secondly, the relative position is calculated according to the relative position adaptive estimator (preset relative position adaptive estimator). Thus, the ship-borne sub-machine is controlled by the preset bounded motion controller, so as to update the inertial coordinate real-time position of the ship-borne sub-machine, and continuously reduce the distance between the ship-borne sub-machine and the ship-borne mother machine, until the relative estimated distance (updated scalar distance) between the sub-machines reaches the target threshold; wherein the target point (inertial coordinate real-time position of the ship-borne sub-machine) after iteration can be expressed as Finally, the target point is navigated through the flight control algorithm, and the scalar distance (updated scalar distance) between the ship-borne sub-machine and the ship-borne mother machine is reacquired through the UWB, until the relative estimated distance (updated scalar distance) between the sub-machines reaches the target threshold.

[0121] It is worth mentioning that after the speed is calculated, the ship-borne sub-machine is controlled based on the preset bounded motion controller, and the ship-borne sub-machine is controlled to fly towards the mother machine, so as to continuously reduce the scalar distance between the ship-borne sub-machine and the ship-borne mother machine, so that the relative estimated distance (updated scalar distance) between the sub-machines reaches the target threshold.

[0122] Specifically, step 105 can include the following sub-steps S51-S58:

[0123] Step S51, initializing the two-dimensional vector, and determining the initial two-dimensional vector;

[0124] Step S52, updating the two-dimensional vector based on the initial two-dimensional vector, and determining the updated two-dimensional vector;

[0125] Step S53, update the applied disturbance according to the updated two-dimensional vector, determine the updated applied disturbance;

[0126] It should be noted that the applied disturbance satisfies , that is The norm of is limited to 1, which ensures that the generated internal signal will not increase indefinitely, so that the control input is always bounded; and for any , there is a constant such that , and , that is The linear span of covers all cases from k=0 to k=K-1, that is, the generated signal is rich enough to cover the entire signal space; where is a three-dimensional Euclidean space, that is, the set of all possible vectors in three-dimensional space, sup is the supremum, that is, the upper bound of the maximum in all possible values, and span is the linear span, that is, the entire linear space generated by a set of vectors.

[0127] Further, the processing process of the updated applied disturbance can be represented as:

[0128] ;

[0129] wherein is the updated applied disturbance at k time; is a transformation function; is a two-dimensional real number space, that is, the set of all two-dimensional real number vectors; N is a natural number set, that is, the set of all positive integers.

[0130] Further, the two-dimensional vector satisfies , and Let the initial two-dimensional vector , then the processing process of the updated two-dimensional vector can be represented as:

[0131] ;

[0132] wherein is the updated two-dimensional vector at k time; is an update matrix; is an initial two-dimensional vector; a is the degree of discretization and periodicity of the signal period; is the initial phase of the signal, and the initial value is generally assigned to 0.

[0133] It is worth mentioning that let , then the mapping is:

[0134] ;

[0135] wherein, is a transformation function; x is a two-dimensional vector in the transformation function; is a first parameter in the transformation function; is a first value of the two-dimensional vector; is a second value of the two-dimensional vector; is a second parameter in the transformation function.

[0136] Step S54, using the docking preparation point inertial coordinate position and the inertial coordinate real-time position of the ship-borne sub-machine, the position of the docking preparation point relative to the ship-borne sub-machine is calculated;

[0137] Step S55, using the pre-set relative position adaptive estimator, the relative position is calculated according to the position of the docking preparation point relative to the ship-borne sub-machine;

[0138] Step S56, using the pre-set bounded motion controller, the ship-borne sub-machine control flight speed is calculated according to the relative position and the updated disturbance;

[0139] Step S57, based on the ship-borne sub-machine control flight speed, the inertial coordinate real-time position of the ship-borne sub-machine is updated, and the updated inertial coordinate real-time position of the ship-borne sub-machine is determined;

[0140] Step S58, according to the updated inertial coordinate real-time position of the ship-borne sub-machine, the scalar distance is updated, and the updated scalar distance is determined.

[0141] It should be noted that the present application introduces a relative position estimator based on UWB and VIO to calculate the position of the mother machine relative to the sub-machine. First, the motion model of the unmanned aerial vehicle is simplified as a first-order integrator model, the position of the center of mass is selected as the state variable x of the system, and the velocity is selected as the input u of the system, that is , , is the horizontal coordinate of the unmanned aerial vehicle, is the longitudinal coordinate of the unmanned aerial vehicle, is the vertical coordinate of the unmanned aerial vehicle, is the flight speed corresponding to the horizontal coordinate of the unmanned aerial vehicle, is the flight speed corresponding to the longitudinal coordinate of the unmanned aerial vehicle, is the flight speed corresponding to the vertical coordinate of the unmanned aerial vehicle; T is a transpose, and the sampling period of the unmanned aerial vehicle state is t s , the motion model can be expressed as: .

[0142] Further, the position of the sub-machine in the inertial coordinate system at time k is obtained by VIO and the velocity , then the displacement of the sub-machine from time k to (k+1) time ; wherein the displacement is obtained by onboard sensors, rather than by differentiating two consecutive absolute positions.

[0143] Further, it is assumed that the mother drone remains hovering while docking, and the position of the mother drone in the inertial frame at this time is denoted as The distance from the mother drone to the child drone at time k is obtained using UWB: .

[0144] Further, in the remote guidance phase, the child drone needs to navigate to a position near the mother drone, and then use a vision-based relative positioning algorithm for accurate positioning. The present application describes this position as the docking preparation point p d , and the positions of the docking preparation point relative to the child drone and the mother drone are denoted as q k , p dm , respectively, i.e. , The navigation goal in the remote guidance phase can be expressed as making the child drone reach the docking preparation point within a certain time, i.e. .

[0145] Based on the above, we have , then:

[0146] ;

[0147] Further, we have: ;

[0148] Further, based on the above definitions, a relative position adaptive estimator is constructed, i.e. the expression corresponding to the pre-set relative position adaptive estimator, which is:

[0149] ;

[0150] wherein is the relative position at time k, representing the estimated position of the docking preparation point relative to the ship-borne child drone at time k; is a mapping or transformation function; is the relative position at time k-1; is the displacement of the ship-borne child drone at time k-1; is a constant, satisfying , is the maximum allowed speed of the unmanned aerial vehicle; is the estimation error at time k.

[0151] It is worth mentioning that the function is used to limit the modulus of the estimated value to be less than the measured distance, and is defined as:

[0152] ;

[0153] where, is a mapping or transformation function; is a mapping or transformation function; is a three-dimensional vector of the incoming function; U is a scalar of the incoming function.

[0154] Further, the estimation error e k is defined as:

[0155]

[0156] After simplifying, we get:

[0157]

[0158] It is worth mentioning that the inertial coordinate position of the docking preparation point and the real-time position of the inertial coordinate of the carrier sub-machine are substituted into the formula , to get the inertial coordinate position of the docking preparation point and the real-time position of the inertial coordinate of the carrier sub-machine , using as the initial , and using and the scalar distance between the carrier sub-machine and the carrier mother machine to calculate the estimation error, and then substituting the estimation error and into the corresponding expression of the pre-set relative position adaptive estimator to calculate the relative position.

[0159] Further, in order to make the adaptive estimator formula converge, a sustained disturbance needs to be added to the flight of the sub-machine until the sub-machine reaches the docking preparation point p d . Therefore, a bounded motion controller is designed for the sub-machine, and its corresponding expression is:

[0160]

[0161] where, is the control flight speed of the carrier sub-machine at time k; is a mapping or transformation function; is a first constant; is the relative position at time k; is the position of the docking preparation point relative to the carrier mother machine; is a second constant; is the updated disturbance applied at time k.

[0162] It is worth mentioning that represents a feedback control term related to the error of the target position, which is used to guide the system to move towards the target direction, represents an internal signal ​​​a related term for increasing flexibility or increasing dynamic characteristics within the system; is the applied disturbance at time k, i.e., the updated applied disturbance at time k; constant , satisfies .

[0163] Further, if the updated scalar distance does not reach the preset distance threshold, the updated scalar distance is taken as a new scalar distance, the target inertial coordinate real-time position of the carrier-borne sub-machine is taken as a new intermediate inertial coordinate real-time position of the carrier-borne sub-machine, the updated two-dimensional vector is taken as a new initial two-dimensional vector, the updated applied disturbance is taken as a new applied disturbance, and the step S52 is executed by jumping until the updated scalar distance reaches the preset distance threshold, and then the current time flight state data of the carrier-borne sub-machine when the updated scalar distance reaches the preset distance threshold is acquired, the trajectory planning is performed in combination with the preset constraint condition, the preset trajectory optimization problem and the preset quasi-Newton optimizer, and the target landing trajectory corresponding to the carrier-borne sub-machine is generated.

[0164] Optionally, the method further comprises:

[0165] If the scalar distance is less than or equal to the preset switching distance, the inertial coordinate position of the docking preparation point is initialized to determine an initial inertial coordinate position of the docking preparation point.

[0166] The two-dimensional vector is initialized to determine an initial two-dimensional vector.

[0167] The two-dimensional vector is updated based on the initial two-dimensional vector to determine an updated two-dimensional vector.

[0168] The applied disturbance is updated according to the updated two-dimensional vector to determine an updated applied disturbance.

[0169] The initial inertial coordinate position of the docking preparation point and the inertial coordinate real-time position of the carrier-borne sub-machine are used to calculate the position of the docking preparation point relative to the carrier-borne sub-machine.

[0170] The preset relative position adaptive estimator is used to calculate the relative position according to the position of the docking preparation point relative to the carrier-borne sub-machine.

[0171] The preset bounded motion controller is used to calculate the control flight speed of the carrier-borne sub-machine according to the relative position and the updated applied disturbance.

[0172] The inertial coordinate real-time position of the carrier-borne sub-machine is updated based on the control flight speed of the carrier-borne sub-machine to determine a target inertial coordinate real-time position of the carrier-borne sub-machine.

[0173] The scalar distance is updated according to the target inertial coordinate real-time position of the carrier-borne sub-machine to determine an updated scalar distance.

[0174] It should be noted that if the scalar distance is less than or equal to the preset switching distance, the geometric positioning method is not used to calculate the inertial coordinate position of the docking preparation point at this time, and the inertial coordinate position of the docking preparation point can be randomly assigned as needed to obtain an initial inertial coordinate position of the docking preparation point, and then the scalar distance between the parent and child machines is gradually reduced. The principle of determining the updated scalar distance here is consistent with the principle of determining the updated scalar distance in step 105, and the present application will not be described in more detail.

[0175] Further, please refer to Figures 2-4 , only the iterative positioning method proposed in the present application is used to guide the docking point back under noisy conditions to obtain the relative position deviation under noisy conditions (the solid line in Figure 3 , Figure 4 The actual deviation is the estimated deviation calculated by the algorithm) After zooming in on part of the graph, it can be observed that the error is within a radius of two meters.

[0176] Further, please refer to Figures 5-7 , the guiding positioning algorithm proposed in the present application is used to guide the docking point back under noisy conditions to obtain the relative position deviation under noisy conditions (the solid line in Figure 5 , Figure 6 The actual deviation is the estimated deviation calculated by the algorithm) After zooming in on part of the graph, it can be observed that the positioning accuracy is not much different from that of the iterative positioning method alone, but the advantage is that the jitter is smaller and the oscillation frequency is lower. As can be seen from the simulation, the guiding positioning algorithm proposed in the present application has faster convergence speed under noisy conditions while maintaining considerable convergence accuracy.

[0177] Step 106, if it is reached, based on the preset constraint condition, a preset trajectory optimization problem and a preset quasi-Newton optimizer are used to perform trajectory planning according to the current flight state data of the ship-borne child machine to generate a target landing trajectory corresponding to the ship-borne child machine.

[0178] Specifically, step 106 can include the following sub-steps S61-S62:

[0179] Step S61, the constraint penalty function corresponding to the preset constraint condition and the preset trajectory optimization problem are transformed by integral method and nonlinear transformation method to determine an unconstrained optimization problem;

[0180] Step S62, the unconstrained optimization problem and the preset constraint condition are input into the preset quasi-Newton optimizer for calculation to generate a target landing trajectory corresponding to the ship-borne child machine.

[0181] The preset constraint condition includes dynamic constraint, angular velocity construction constraint, collision distance constraint, relative velocity constraint, acceleration constraint caused by drag force, and field of view angle constraint.

[0182] The current time flight state data includes the position and attitude of the ship-borne sub-machine in the inertial coordinate system, the flight speed of the ship-borne sub-machine, the acceleration of the ship-borne sub-machine, the lateral and longitudinal axis angular velocity of the ship-borne sub-machine, the vertical axis angular velocity of the ship-borne sub-machine, the vertical axis height distance between the ship-borne sub-machine and the ship-borne mother machine, the relative speed between the ship-borne sub-machine and the ship-borne mother machine, the wind speed and the standard deviation.

[0183] The preset quasi-Newton optimizer is a quasi-Newton L-BFGS (Limited-memory BFGS) optimizer.

[0184] It should be noted that the trajectory optimization problem (preset trajectory optimization problem) in the docking process is specifically:

[0185] ;

[0186] Wherein, is the preset trajectory optimization problem; c is a coefficient matrix in the trajectory; T is a trajectory time; is the s-th order derivative of the system state z(t) with respect to time t; is the weight of the time cost; is the starting point constraint of the trajectory; is the end point constraint of the trajectory; is a dynamic constraint condition, that is, a preset constraint condition; is a constraint set of a dynamic system, indicating a set that needs to be met by the system state and the control input, for limiting the feasible state space of the system; is the state of the system at time t; s.t.T indicates that the total time of the trajectory needs to be greater than zero.

[0187] It is worth mentioning that, The definition of is a vector. It contains the state of the trajectory and its previous (s-1) order derivatives: Wherein, is the original system vector of the system, is the first derivative of the system state, which is usually the speed or change rate, describing the change of the state variable, is the high-order derivative of the system state.

[0188] Further, in the remote guidance phase, the present application selects the smoothness order s = 4, i.e. minimizes the squared Euclidean norm of the fourth derivative of position. According to the differential flatness theory of multi-rotor UAVs, the moment in the system input is a function of the fourth derivative of position. Therefore, MinimumSnap corresponds to minimizing the differential of the thrust, which can effectively reduce the energy consumption of the UAV, thereby improving its endurance. In the precision landing phase, the present application selects the smoothness order s = 3, i.e. minimizes the squared Euclidean norm of the jerk. According to the differential flatness theory of multi-rotor UAVs, the angular velocity is a function of the third derivative of position. Therefore, MinimumJerk corresponds to minimizing the angular velocity, which is conducive to visual tracking during landing and reduces the probability of tag loss.

[0189] Further, in the precision landing phase, although the mother machine remains in a hovering state, it may be affected by external disturbances (such as wind, airflow interference of the child machine, etc.), resulting in unstable state. In order to ensure that the child machine can accurately and stably dock with the mother machine, the planner needs to perform high-frequency trajectory re-planning. The purpose of trajectory re-planning is to continuously adjust the flight trajectory of the child machine according to the real-time acquired state information of the mother machine, so that the final state of the child machine is consistent with that of the mother machine. The end state of the child machine can be represented by the following equation:

[0190]

[0191] wherein, is the inertial coordinate position of the carrier child machine; is the position of the landing target point in the inertial coordinate system; is the inertial coordinate attitude of the carrier child machine; is the attitude of the landing target point in the inertial coordinate system; is the flight speed of the carrier child machine; is the flight speed of the carrier mother machine; is a very small traction speed in the direction; is the direction vector of the z-axis in the landing coordinate system; is the acceleration of the carrier child machine.

[0192] It is worth mentioning that the above-mentioned target attitude of the end point is to align the attitude of the child machine with the two-dimensional code on the mother machine, and the target position of the end point is a small distance in the positive direction of the z-axis at the calculated position of the two-dimensional code, because when approaching the final target landing point, the field of view angle of the camera is limited and the complete two-dimensional code cannot be seen, i.e. the relative position cannot be obtained, at this time a very small traction speed in the z-axis direction is added to the end speed, so that the child machine moves to the target docking position, and then the motor is turned off to make the child machine stop on the mother machine.

[0193] ​Further, in order to ensure the smooth trajectory of the UAV in space, avoid rapid displacement or sudden acceleration changes, set the maximum allowed flight speed as v max , the maximum allowed acceleration as a max , establish the dynamics constraint:

[0194] ;

[0195] Further, construct the punishment function corresponding to the dynamics constraint:

[0196] ;

[0197] wherein, is the flight speed punishment function corresponding to the dynamics constraint; is the acceleration punishment function corresponding to the dynamics constraint; is the defined mapping transformation, used to quantify the punishment size.

[0198] The definition of function is:

[0199] ;

[0200] wherein, x is the first parameter in the input mapping; y is the second parameter in the input mapping.

[0201] Further, since when two UAVs gradually approach docking, the attitude needs to be stable to avoid docking errors caused by rapid changes in attitude, the angular velocity constraint is mainly used to control the attitude change speed of the UAV. Attitude control is particularly important for stable flight and turning of the UAV. If the angular velocity is not limited, it may cause the UAV to rotate violently during attitude adjustment, affecting the overall stability, especially when approaching other aircraft or docking, therefore, the angular velocity of the UAV is divided into two parts, considering the roll angle and pitch angle, and limiting the angular velocity of the yaw angle alone, set w xy,max as the maximum allowed angular velocity of the x-y axis, w z,max as the maximum allowed angular velocity of the z axis, construct the constraint, that is, the angular velocity construct constraint:

[0202] ;

[0203] wherein, is the horizontal and longitudinal axis angular velocity of the shipborne sub-machine at time t; is the vertical axis angular velocity of the shipborne sub-machine at time t.

[0204] Further, construct the punishment function corresponding to the angular velocity constraint:

[0205] ;

[0206] wherein, is a penalty function corresponding to the angular velocity constraint; is a second-order continuously differentiable smooth approximation function of max(x, 0).

[0207] Further, referring to Figure 8 , the shape of the function under different polishing factors can control the precision of the constraint by adjusting the size of the function, the function is defined as:

[0208] ;

[0209] wherein, is a smooth approximation function; x is the input of the smooth approximation function.

[0210] Further, since the docking process is carried out in the air, the unsuccessful landing of the sub-machine may touch the propeller of the mother machine and cause a crash, so the sub-machine needs to be stably maintained above the mother machine during docking, and if vibration or deviation occurs, it will cause loss. The z-axis height constraint of the sub-mother machine is constructed, and the minimum safe height difference is d min , this height difference is related to the effective distance of the sub-machine recognizing the two-dimensional code in the final landing stage, if this value is too small, the two-dimensional code will be lost in the field of view and cannot be positioned, and if it is too large, it will produce collision risk, the z-axis height of the sub-machine and the mother machine in the docking stage is z c and z m , the collision distance constraint is constructed:

[0211] ;

[0212] wherein, is the vertical axis height distance between the ship-borne sub-machine and the ship-borne mother machine at time t.

[0213] Further, the penalty function corresponding to the collision distance constraint is constructed:

[0214] ;

[0215] wherein, is a penalty function corresponding to the collision distance constraint; is a defined mapping transformation, which is used to quantify the penalty size.

[0216] Further, the function is defined as:

[0217] ;

[0218] wherein, x is Input of the function.

[0219] Further, during the docking process, not only the speed of each UAV should be controlled, but also the relative speed between the two UAVs. The relative speed between the two UAVs near the docking point can be limited to ensure that the speed of the two UAVs is slow and stable when approaching the docking point. Reducing impact: when the two UAVs gradually approach each other, keeping a small change in the relative speed can reduce the impact or collision risk caused by the speed difference, and the maximum allowed relative speed v rel,max is set as

[0220] ;

[0221] Further, the relative speed constraint corresponding to the penalty function is constructed:

[0222] ;

[0223] wherein, is the penalty function corresponding to the relative speed constraint.

[0224] Further, considering the interference of the airflow generated by the mother UAV on the child UAV during docking, an integrated wind speed and direction sensor module is equipped on the UAV. The present application assumes that the wind speed v w and the standard deviation can be measured by the sensor, (the speed here is a three-dimensional vector), and the acceleration of the UAV is affected by the drag force f. Generally, the size of the drag force is proportional to the square of the speed. A general expression is:

[0225] ;

[0226] wherein, the total wind speed relative to the speed of the UAV V is defined as:

[0227] ;

[0228] wherein, is the wind speed; is the speed of the child UAV.

[0229] Further, the generated drag force can be expressed as:

[0230] ;

[0231] wherein, c is the drag coefficient related to the characteristics of the UAV; represents the range of turbulent disturbance in the direction of the wind speed

[0232] Further, in order to ensure that the unmanned aerial vehicle can maintain stable acceleration in the wind field, and avoid losing control due to the influence of turbulence, the application can constrain the acceleration caused by the drag force, that is, the acceleration constraint caused by the drag force, which is specifically:

[0233] ;

[0234] Wherein, m is the mass of the sub-machine; is the maximum allowed interference acceleration.

[0235] Based on the above, we can get:

[0236] ;

[0237] Further, the penalty function corresponding to the acceleration constraint caused by the drag force is constructed:

[0238] ;

[0239] Wherein, is the penalty function corresponding to the acceleration constraint caused by the drag force.

[0240] Further, due to the two-dimensional code detection system relying on vision in the docking process, the field of view angle constraint needs to be performed on the sub-machine, so that the mother machine appears in the field of view range in real time, which ensures that the sensor of the unmanned aerial vehicle can capture real-time information of the docking point and the surrounding environment, avoids collision, especially in the face of dynamic environment and targets close to each other. Assuming that r cm is the relative position vector from the sub-machine to the mother machine, u camera is the line of sight direction unit vector of the camera of the sub-machine, is the field of view angle range of the unmanned aerial vehicle, is the included angle between the relative position vector and the direction between the camera.

[0241] Further, the camera of the sub-machine needs to ensure that the mother machine is always located in its field of view, so the field of view angle constraint can be expressed as: That is:

[0242] ;

[0243] ;

[0244] Further, the penalty function corresponding to the field of view angle constraint is constructed:

[0245] ;

[0246] Wherein, is the penalty function corresponding to the field of view angle constraint.

[0247] Further, the integral method is used to calculate the violation of the continuous time constraint condition, so as to reduce the infinite constraint conditions to a limited number, that is:

[0248]

[0249] wherein, is the total penalty function of the constraint condition obtained by integrating the continuous time under the i-th constraint condition; is the penalty function corresponding to the preset constraint condition.

[0250] Suppose the weight of the constraint penalty is Then the constraint penalty received by the sub-machine during the docking process can be summarized as:

[0251]

[0252] wherein, is the total penalty accumulation of all constraints; is the total penalty of the flight speed; is the weight factor of the flight speed constraint.

[0253] Further, the trajectory optimization problem with geometric constraints is converted into an unconstrained optimization problem by integral method and nonlinear transformation, that is, the unconstrained optimization problem J, which is specifically:

[0254]

[0255] Further, referring to Figures 9-11 All constraints are transmitted into the quasi-Newton L-BFGS (Limited-memory BFGS) optimizer for calculation, so as to obtain a final landing trajectory that satisfies the dynamic constraint and considers the air flow interference, that is, the target landing trajectory corresponding to the ship-borne sub-machine.

[0256] As a comparison of technical effects, the existing technology can be referred to. The existing research on vehicle-mounted dynamic docking focuses on guiding positioning and trajectory planning, although it also relies on visual recognition technology, but unlike airborne docking, vehicle-mounted docking does not need to consider air flow interference. The research on airborne docking is still relatively rare, and the existing airborne docking station mostly relies on additional mechanical arms to complete, which not only increases the cost, but also improves the control complexity. Some colleges and universities try to conduct experiments indoors through motion capture systems, but in outdoor environments, especially in environments with high positioning errors and air flow interference, airborne docking is still in the bottleneck stage and urgently needs further research and breakthroughs.

[0257] ​​​Most of the docking algorithms in the past require the landing platform to be visible throughout the entire mission, because the UAV relies on visual information to output control variables. Although the on-board visual system can effectively solve the docking problem of the UAV, its coverage is limited, and it is difficult to meet the long-range guidance needs of the UAV. The application scenario of on-board docking is that the sub-machine returns to the docking after executing the task, and at this time, the visual system cannot play a role, and the GPS positioning has a large error, and in the GNSS denial environment, it cannot be accurately positioned, and the high-precision RTK needs to be arranged corresponding to the base station to improve the precision, which does not meet the application scenario of the on-board docking system, so it is necessary to use a relative positioning algorithm to guide the sub-machine to the vicinity of the mother machine. At the same time, for the nest docking and vehicle docking system, the airflow of the UAV and the nest or the mobile vehicle end influences each other less, but for the on-board docking system, the airflow of the mother machine and the sub-machine interferes with each other more, when the sub-machine is above the mother machine preparing to dock, the airflow generated by the two will interfere with each other and cause the attitude or positioning to drift, thereby reducing the success rate of docking, and the existing docking system control method rarely considers the interference caused by the airflow influence.

[0258] In view of the above problems, the present application provides a landing trajectory planning method for shipborne dynamic docking, please refer to Figure 12 The present application combines the relative positioning algorithm of the geometric positioning method and the iterative method, combines the ultra wide band technology (Ultra WideBand, UWB) and visual positioning, under the influence of sensor noise, the estimated value obtained by the iterative positioning method may have a large error and a slow convergence speed as the distance error increases at a long distance, therefore, in order to improve the efficiency of the long-range guidance task, the geometric positioning method is used to calculate the rough position of the docking preparation point in the initial stage, and when the distance between the sub-machine and the mother machine is reduced to a certain range, the iterative positioning method is switched to obtain more accurate position estimation. Specifically, when the sub-machine enters the guidance stage after completing the task in the non-visible area of the two-dimensional code and preparing to return, the scalar distance between the sub-machine and the mother machine is measured in real time, when the distance is too large, the position of the docking point and the corresponding pose are estimated by the geometric positioning method, when the position deviation between the sub-machine and the estimated docking point reaches a given threshold, that is, the preset switching distance, the iterative positioning method is switched to continuously reduce the scalar distance between the sub-machine and the mother machine, thereby entering the docking landing stage.

[0259] Further, the present application is equipped with an integrated wind speed and direction sensor module on the sub-machine, the acceleration of the sub-machine under the influence of turbulence is constrained by observing the data of the sensor, a penalty containing field of view angle constraint and airflow interference constraint is established, so as to reduce the influence of airflow on the system during docking.

[0260] In the embodiment of the present application, the above scheme of the present application provides a landing trajectory planning method for dynamic docking of a carrier-based aircraft. First, a scalar distance between a carrier-based child aircraft and a carrier-based parent aircraft is obtained, and the scalar distance is compared with a preset allowable error. Then, if the scalar distance is greater than the preset allowable error, the scalar distance is compared with a preset switching distance. If the scalar distance is greater than the preset switching distance, the inertial coordinate position of a docking preparation point is calculated based on a geometric positioning method, using the inertial coordinate reference position of the carrier-based child aircraft, a plurality of waypoints corresponding to the carrier-based child aircraft, and the position difference distance between the carrier-based child aircraft at each waypoint and the carrier-based parent aircraft. The inertial coordinate real-time position of the carrier-based child aircraft is updated in real time according to the inertial coordinate position of the docking preparation point using a proportional-derivative controller, the intermediate inertial coordinate real-time position of the carrier-based child aircraft is determined, and the difference between the intermediate inertial coordinate real-time position of the carrier-based child aircraft and the inertial coordinate position of the docking preparation point is calculated to determine whether the difference between the docking point distance of the carrier-based child aircraft reaches the preset switching distance. If the difference reaches the preset switching distance, the scalar distance is updated based on an iterative positioning method, using the flight influence data of the carrier-based child aircraft, the inertial coordinate position of the docking preparation point, and the intermediate inertial coordinate real-time position of the carrier-based child aircraft, the updated scalar distance is determined, and it is determined whether the updated scalar distance reaches a preset distance threshold. Finally, if the updated scalar distance reaches the preset distance threshold, a target landing trajectory corresponding to the carrier-based child aircraft is generated based on the preset constraint condition, using a preset trajectory optimization problem and a preset quasi-Newton optimizer according to the current flight state data of the carrier-based child aircraft. Based on the above scheme, when the distance between the child aircraft and the parent aircraft is too large, i.e., the scalar distance between the child aircraft and the parent aircraft exceeds the allowable error and the switching threshold, the inertial coordinate position of the docking preparation point is first calculated based on the geometric positioning method, and the distance between the child aircraft and the parent aircraft is reduced to a certain range based on the proportional-derivative controller, and then the iterative positioning method is switched to continue to reduce the scalar distance between the child aircraft and the parent aircraft, which can quickly guide the child aircraft to return to the airspace near the parent aircraft, so that the child aircraft can accurately return to the parent aircraft directly above, thereby improving the positioning accuracy and further improving the landing accuracy of the carrier-based child aircraft.

[0261] For better illustration, refer to Figure 13 Fig. 2 shows a schematic diagram of the operation flow of the carrier-based aircraft system provided in the second embodiment of the present application, and the process is specifically as follows:

[0262] The guidance homing algorithm of the sub-machine, the Apritag relative positioning algorithm based on vision and the trajectory optimization method considering the influence of air flow are proposed based on the application. At the beginning of the task, the carrier-based unmanned aerial vehicle receives the task instruction and carries the sub-machine to the target area. After arriving at the parking point of the mother machine, the sub-machine takes off to perform tasks such as monitoring, package delivery, search and rescue. After the completion of the task, due to the drift of the positioning system of the sub-machine and the mother machine, the direct use of the coordinate difference of the two may lead to inaccurate guidance, and even there is a risk of collision with the mother machine. Through the guidance homing algorithm proposed in the application, combined with the geometric method of space four-point positioning and the iterative method based on vision and distance, the sub-machine can be accurately and efficiently guided to return to the upper side of the mother machine, ready for docking.

[0263] Further, after the homing is successful, the sub-machine is above the mother machine, and a two-dimensional code is placed on the upper side of the mother machine to assist detection. At this time, based on the Apritag (two-dimensional code detection system) relative positioning algorithm, the pose information of the two-dimensional code of the mother machine relative to the sub-machine can be obtained. The pose information will be converted to the inertial coordinate system of the sub-machine, so as to obtain the real-time position of the mother machine, that is, the target point of trajectory generation. Finally, combined with factors such as dynamic constraint, camera field angle constraint and air flow interference, the trajectory is optimized to generate a smooth trajectory, successfully guiding the sub-machine to complete accurate landing and safe docking on the mother machine.

[0264] Compared with the prior art, in an outdoor environment, the relative position of the sub-machine is often affected by factors such as light changes and terrain obstructions, resulting in large deviations of the GPS or vision positioning system, especially after the sub-machine completes the task, the two-dimensional code cannot be recognized at a long distance, and the homing cannot be performed by relying on the relative positioning method based on Apritag, which brings great challenges to the air docking. Once the positioning error exceeds the safety limit, the sub-machine and the mother machine may collide in the air, thereby causing serious consequences.

[0265] Therefore, the application provides a landing trajectory planning method for dynamic docking of a carrier-based aircraft, aiming to improve the robustness and accuracy of guidance. The geometric method positions through four points that are not coplanar in space, has the advantages of fast convergence and easy implementation, but as the distance between the mother aircraft and the child aircraft increases and the visual positioning error accumulates, the positioning accuracy gradually decreases, which cannot meet the high-precision requirement. In contrast, the iterative method can gradually converge to higher accuracy with the increase of the number of iterations, but it needs a long time and is relatively complex to implement. Therefore, when the distance between the two aircrafts exceeds the switching threshold, the geometric method is first used to quickly guide the child aircraft back to the vicinity of the mother aircraft, and then the iterative method is used to further improve the positioning accuracy, so that the child aircraft accurately returns to the top of the mother aircraft, which not only shortens the guidance time, but also improves the positioning accuracy. After the child aircraft is guided to the top of the mother aircraft, the Apritag-based two-dimensional code assisted positioning system can accurately obtain the position of the mother aircraft in the child aircraft coordinate system, providing accurate data for subsequent trajectory optimization, and ensuring the safe landing of the child aircraft.

[0266] In the embodiment of the application, unlike the landing platform on the ground or water surface, the airborne platform will encounter more complex airflow interference during docking. When docking in the air, the downward airflow of the child aircraft will have a significant impact on the mother aircraft. This uneven turbulence will cause irregular changes in the attitude (such as tilt angle, yaw angle, etc.) of the upper unmanned aerial vehicle, resulting in a decrease in the height and left-right shaking of the mother aircraft, which will affect its flight stability. This airflow disturbance not only increases the attitude instability of the mother aircraft, but also directly affects the landing accuracy of the child aircraft. Therefore, the application introduces an Apritag-based two-dimensional code assisted positioning system. Due to high-frequency airflow fluctuations, relying solely on the positioning data of the child aircraft cannot ensure high-precision docking. The airflow above the mother aircraft will be compressed, thereby reducing the maneuverability of the child aircraft when approaching the mother aircraft, causing significant hysteresis, making it difficult for traditional trajectory optimization methods to effectively cope with this complex airflow disturbance. In order to overcome these challenges, the application combines camera field of view angle constraints and airflow interference compensation technology. During the landing process, by adjusting the camera field of view angle constraints in real time, the positioning error caused by the change of the viewing angle can be effectively avoided. At the same time, the airflow interference compensation technology can dynamically adjust the flight trajectory of the child aircraft, reducing the negative impact of airflow on maneuverability. Through the integration of these two technologies, the application can ensure effective correction of the child aircraft landing process in a complex airflow environment, significantly improving the safety and robustness of the air docking. Finally, the child aircraft can more accurately and stably land above the mother aircraft, thereby completing the safe and accurate air docking task.

[0267] Please refer to Figure 14 , Figure 14 A structural block diagram of a landing trajectory planning device for dynamic docking of a carrier-based aircraft is provided in the third embodiment of the application.

[0268] The application provides a landing trajectory planning device for dynamic docking of a carrier-based aircraft, which comprises:

[0269] The acquisition module 1401 is configured to acquire a scalar distance between the carrier-based child aircraft and the carrier-based parent aircraft, and compare the scalar distance with a preset allowable error.

[0270] The comparison module 1402 is configured to compare the scalar distance with a preset switching distance if the scalar distance is greater than the preset allowable error.

[0271] The geometric positioning module 1403 is configured to calculate an inertial coordinate position of a docking preparation point based on a geometric positioning method, by using a reference inertial coordinate position of the carrier-based child aircraft, a plurality of waypoints corresponding to the carrier-based child aircraft, and position difference distances between the carrier-based child aircraft and the carrier-based parent aircraft at the plurality of waypoints, if the scalar distance is greater than the preset switching distance.

[0272] The position updating module 1404 is configured to update a real-time inertial coordinate position of the carrier-based child aircraft according to the inertial coordinate position of the docking preparation point by using a proportional differential controller, determine an intermediate inertial coordinate real-time position of the carrier-based child aircraft, and calculate a distance difference between the intermediate inertial coordinate real-time position of the carrier-based child aircraft and the inertial coordinate position of the docking preparation point, and judge whether the distance difference reaches the preset switching distance.

[0273] The iterative positioning module 1405 is configured to update the scalar distance based on an iterative positioning method by using flight influence data of the carrier-based child aircraft, the inertial coordinate position of the docking preparation point and the intermediate inertial coordinate real-time position of the carrier-based child aircraft, determine an updated scalar distance, and judge whether the updated scalar distance reaches a preset distance threshold, if the distance difference reaches the preset switching distance.

[0274] The trajectory planning module 1406 is configured to perform trajectory planning based on a preset constraint condition by using a preset trajectory optimization problem and a preset quasi-Newton optimizer according to current time flight state data of the carrier-based child aircraft to generate a target landing trajectory corresponding to the carrier-based child aircraft, if the distance difference reaches the preset switching distance.

[0275] Further, the geometric positioning module 1403 is specifically configured to:

[0276] Calculate a local coordinate position of the carrier-based parent aircraft according to the plurality of waypoints corresponding to the carrier-based child aircraft and the position difference distances between the carrier-based child aircraft and the carrier-based parent aircraft at the plurality of waypoints.

[0277] Add the local coordinate position of the carrier-based parent aircraft and the reference inertial coordinate position of the carrier-based child aircraft to determine an inertial coordinate position of the carrier-based parent aircraft relative to the carrier-based child aircraft.

[0278] Calculate the inertial coordinate position of the docking preparation point by using the inertial coordinate position of the carrier-based parent aircraft relative to the carrier-based child aircraft and the local coordinate position of the carrier-based parent aircraft.

[0279] Further, the position updating module 1404 is specifically configured to:

[0280] Subtracting the inertial coordinate position of the docking preparation point from the inertial coordinate real-time position of the ship-borne sub-machine to determine an error term;

[0281] Using a proportional differential controller to output a speed control signal according to the error term;

[0282] Updating the inertial coordinate real-time position of the ship-borne sub-machine based on the speed control signal to determine an intermediate inertial coordinate real-time position of the ship-borne sub-machine.

[0283] Further, the flight influence data includes an applied disturbance and a two-dimensional vector; and the iterative positioning module 1405 is specifically configured to:

[0284] Initializing the two-dimensional vector to determine an initial two-dimensional vector;

[0285] Updating the two-dimensional vector based on the initial two-dimensional vector to determine an updated two-dimensional vector;

[0286] Updating the applied disturbance according to the updated two-dimensional vector to determine an updated applied disturbance;

[0287] Using the inertial coordinate position of the docking preparation point and the intermediate inertial coordinate real-time position of the ship-borne sub-machine to calculate the position of the docking preparation point relative to the ship-borne sub-machine;

[0288] Using a preset relative position adaptive estimator to calculate the relative position according to the position of the docking preparation point relative to the ship-borne sub-machine;

[0289] Using a preset bounded motion controller to calculate a ship-borne sub-machine control flight speed according to the relative position and the updated applied disturbance;

[0290] Updating the intermediate inertial coordinate real-time position of the ship-borne sub-machine based on the ship-borne sub-machine control flight speed to determine a target inertial coordinate real-time position of the ship-borne sub-machine;

[0291] Updating the scalar distance according to the target inertial coordinate real-time position of the ship-borne sub-machine to determine an updated scalar distance.

[0292] Further, the method further includes:

[0293] The first module is configured to initialize the inertial coordinate position of the docking preparation point to determine an initial inertial coordinate position of the docking preparation point if the scalar distance is less than or equal to a preset switching distance;

[0294] The second module is configured to initialize the two-dimensional vector to determine an initial two-dimensional vector;

[0295] The third module is configured to update the two-dimensional vector based on the initial two-dimensional vector, and determine an updated two-dimensional vector;

[0296] The fourth module is configured to update the applied disturbance according to the updated two-dimensional vector, and determine an updated applied disturbance;

[0297] The fifth module is configured to calculate the position of the docking preparation point relative to the carrier-borne sub-machine by using the initial docking preparation point inertial coordinate position and the inertial coordinate real-time position of the carrier-borne sub-machine.

[0298] The sixth module is configured to calculate the relative position by using a preset relative position adaptive estimator according to the position of the docking preparation point relative to the carrier-borne sub-machine.

[0299] The seventh module is configured to calculate the carrier-borne sub-machine control flight speed by using a preset bounded motion controller according to the relative position and the updated applied disturbance.

[0300] The eighth module is configured to update the inertial coordinate real-time position of the carrier-borne sub-machine based on the carrier-borne sub-machine control flight speed, and determine a target inertial coordinate real-time position of the carrier-borne sub-machine.

[0301] The ninth module is configured to update the scalar distance according to the target inertial coordinate real-time position of the carrier-borne sub-machine, and determine an updated scalar distance.

[0302] Further, the trajectory planning module 1406 is specifically configured to:

[0303] The constraint penalty function corresponding to the preset constraint condition and the preset trajectory optimization problem are transformed by using the integral method and the nonlinear transformation method, and an unconstrained optimization problem is determined.

[0304] The unconstrained optimization problem and the preset constraint condition are input into the preset quasi-Newton optimizer for calculation, and a target landing trajectory corresponding to the carrier-borne sub-machine is generated.

[0305] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described device and module can refer to the corresponding process in the foregoing method embodiments, which will not be described herein.

[0306] The embodiment of the present application further provides a computer device, which comprises a memory and a processor, and the memory stores a computer program; the computer program is executed by the processor, so that the processor executes the steps of the landing trajectory planning method for dynamic docking of a carrier-borne machine in the foregoing embodiment one.

[0307] The embodiment of the present application further provides a computer readable storage medium, which stores a computer program / instruction, and the computer program / instruction is executed by a processor to implement the steps of the landing trajectory planning method for dynamic docking of a carrier-borne machine in the foregoing embodiment one.

[0308] The embodiment of the present application further provides a computer program product comprising computer programs / instructions, which, when executed by a processor, implement the steps of the method for planning a landing trajectory of a carrier-based aircraft in dynamic docking according to the above embodiment one.

[0309] In several embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented in other manners. For example, the division of the above-described apparatus embodiments is merely a logical function division, and there can be another division manner for the actual implementation, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between different units, or the among different units, can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0310] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., can be located in one place, or can be distributed on multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0311] The above description and the above embodiments are merely used to illustrate the technical solutions of the present application, but not to limit the present application; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: the technical solutions recorded in the foregoing embodiments can still be modified, or some technical features can be replaced by equivalents; and these modifications or replacements do not make the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for planning a landing trajectory for a carrier-based aircraft dynamic interface, characterized in that, The method comprises the following steps: acquiring a scalar distance between a carrier sub-machine and a carrier mother-machine, and comparing the scalar distance with a preset allowable error; if the scalar distance is greater than the preset allowable error, comparing the scalar distance with a preset switching distance; if the scalar distance is greater than the preset switching distance, calculating an inertial coordinate position of a docking preparation point based on a geometric positioning method, using a reference inertial coordinate position of the carrier sub-machine, a plurality of waypoints corresponding to the carrier sub-machine, and a position difference distance between the carrier sub-machine and the carrier mother-machine at each waypoint; updating a real-time inertial coordinate position of the carrier sub-machine according to the inertial coordinate position of the docking preparation point using a proportional differential controller, determining an intermediate inertial coordinate real-time position of the carrier sub-machine, and calculating a difference in a sub-machine docking point distance between the intermediate inertial coordinate real-time position of the carrier sub-machine and the inertial coordinate position of the docking preparation point, and judging whether the difference in the sub-machine docking point distance reaches the preset switching distance; if so, updating the scalar distance based on an iterative positioning method, using flight influence data of the carrier sub-machine, the inertial coordinate position of the docking preparation point, and the intermediate inertial coordinate real-time position of the carrier sub-machine, determining an updated scalar distance, and judging whether the updated scalar distance reaches a preset distance threshold; if so, generating a target landing trajectory corresponding to the carrier sub-machine based on a preset trajectory optimization problem and a preset quasi-Newton optimizer according to current flight state data of the carrier sub-machine under a preset constraint condition.

2. The method of claim 1, wherein, The method of calculating the inertial coordinate position of the docking preparation point based on the geometric positioning method comprises the following steps: calculating a local coordinate position of the carrier mother-machine according to the plurality of waypoints corresponding to the carrier sub-machine and the position difference distance between the carrier sub-machine and the carrier mother-machine at each waypoint; adding the local coordinate position of the carrier mother-machine and the reference inertial coordinate position of the carrier sub-machine to determine an inertial coordinate position of the carrier mother-machine relative to the carrier sub-machine; and calculating the inertial coordinate position of the docking preparation point using the inertial coordinate position of the carrier mother-machine relative to the carrier sub-machine and the local coordinate position of the carrier mother-machine.

3. The method of claim 1, wherein, The method of updating the real-time inertial coordinate position of the carrier sub-machine according to the inertial coordinate position of the docking preparation point using the proportional differential controller to determine the intermediate inertial coordinate real-time position of the carrier sub-machine comprises the following steps: determining an error term by subtracting the inertial coordinate position of the docking preparation point from the real-time inertial coordinate position of the carrier sub-machine; outputting a speed control signal according to the error term using the proportional differential controller; updating the real-time inertial coordinate position of the carrier sub-machine based on the speed control signal to determine the intermediate inertial coordinate real-time position of the carrier sub-machine.

4. The method of claim 1, wherein, The flight influence data includes a disturbance and a two-dimensional vector; based on the iterative positioning method, the flight influence data of the ship-borne sub-machine, the docking preparation point inertial coordinate position, and the intermediate inertial coordinate real-time position of the ship-borne sub-machine are used to update the scalar distance, and the updated scalar distance is determined, including: The two-dimensional vector is initialized to determine an initial two-dimensional vector; The two-dimensional vector is updated based on the initial two-dimensional vector to determine an updated two-dimensional vector; The disturbance is updated according to the updated two-dimensional vector to determine an updated disturbance; The docking preparation point relative to the ship-borne sub-machine is calculated based on the docking preparation point inertial coordinate position and the intermediate inertial coordinate real-time position of the ship-borne sub-machine; The relative position is calculated based on the docking preparation point relative to the ship-borne sub-machine by using a preset relative position adaptive estimator; The ship-borne sub-machine control flight speed is calculated based on the relative position and the updated disturbance by using a preset bounded motion controller; The intermediate inertial coordinate real-time position of the ship-borne sub-machine is updated based on the ship-borne sub-machine control flight speed to determine the target inertial coordinate real-time position of the ship-borne sub-machine; The scalar distance is updated according to the target inertial coordinate real-time position of the ship-borne sub-machine to determine the updated scalar distance.

5. The method of claim 4, wherein, Further comprising: If the scalar distance is less than or equal to the preset switching distance, the docking preparation point inertial coordinate position is initialized to determine an initial docking preparation point inertial coordinate position; The two-dimensional vector is initialized to determine an initial two-dimensional vector; The two-dimensional vector is updated based on the initial two-dimensional vector to determine an updated two-dimensional vector; The disturbance is updated according to the updated two-dimensional vector to determine an updated disturbance; The docking preparation point relative to the ship-borne sub-machine is calculated based on the initial docking preparation point inertial coordinate position and the inertial coordinate real-time position of the ship-borne sub-machine; The relative position is calculated based on the docking preparation point relative to the ship-borne sub-machine by using a preset relative position adaptive estimator; The ship-borne sub-machine control flight speed is calculated based on the relative position and the updated disturbance by using a preset bounded motion controller; The inertial coordinate real-time position of the ship-borne sub-machine is updated based on the ship-borne sub-machine control flight speed to determine the target inertial coordinate real-time position of the ship-borne sub-machine; The scalar distance is updated according to the target inertial coordinate real-time position of the ship-borne sub-machine to determine the updated scalar distance.

6. The method of claim 1, wherein, Based on the preset constraint condition, the preset trajectory optimization problem and the preset quasi-Newton optimizer are used to perform trajectory planning according to the current time flight state data of the ship-borne sub-machine to generate the target landing trajectory corresponding to the ship-borne sub-machine, including: The constraint penalty function corresponding to the preset constraint condition and the preset trajectory optimization problem are transformed by integral method and nonlinear transformation method to determine an unconstrained optimization problem; The unconstrained optimization problem and the preset constraint condition are input into a preset quasi-Newton optimizer for calculation to generate the target landing trajectory corresponding to the ship-borne sub-machine.

7. A landing trajectory planning device for dynamic docking of carrier-based aircraft, characterized in that: Including: The acquisition module is configured to acquire a scalar distance between the carrier sub-aircraft and the carrier mother-aircraft, and compare the scalar distance with a preset allowable error; The comparison module is configured to compare the scalar distance with a preset switching distance if the scalar distance is greater than the preset allowable error; The geometric positioning module is configured to calculate an inertial coordinate position of a docking preparation point based on a geometric positioning method, using an inertial coordinate reference position of the carrier sub-aircraft, a plurality of waypoints corresponding to the carrier sub-aircraft, and a position difference distance between the carrier sub-aircraft and the carrier mother-aircraft at each of the waypoints, if the scalar distance is greater than the preset switching distance; The position updating module is configured to update a real-time inertial coordinate position of the carrier sub-aircraft based on the inertial coordinate position of the docking preparation point using a proportional-differential controller, determine an intermediate inertial coordinate real-time position of the carrier sub-aircraft, and calculate a difference between a docking point distance of the carrier sub-aircraft and the inertial coordinate position of the docking preparation point, to determine whether the difference reaches the preset switching distance; The iterative positioning module is configured to update the scalar distance based on an iterative positioning method, using flight influence data of the carrier sub-aircraft, the inertial coordinate position of the docking preparation point, and the intermediate inertial coordinate real-time position of the carrier sub-aircraft, to determine an updated scalar distance, and determine whether the updated scalar distance reaches a preset distance threshold, if the difference reaches the preset switching distance; The trajectory planning module is configured to perform trajectory planning based on a preset constraint condition, using a preset trajectory optimization problem and a preset quasi-Newton optimizer according to current flight state data of the carrier sub-aircraft, to generate a target landing trajectory corresponding to the carrier sub-aircraft, if the updated scalar distance reaches the preset distance threshold.

8. A computer device, comprising: The computer program is executed by the processor to cause the processor to perform the steps of the carrier aircraft dynamic docking landing trajectory planning method according to any one of claims 1-6.

9. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed to implement the carrier aircraft dynamic docking landing trajectory planning method according to any one of claims 1-6.

10. A computer program product, characterised in that, The computer program product comprises a computer program stored on a non-transitory computer-readable storage medium, and the computer program comprises program instructions, wherein when the program instructions are executed by a computer, the computer performs the carrier aircraft dynamic docking landing trajectory planning method according to any one of claims 1-6.

Images