Pipeline-based tail-seat vertical take-off and landing UAV control method and device

Abstract

This invention discloses a pipeline-based tail-seat VTOL unmanned aerial vehicle (UAV) planning and control method and apparatus. The method includes: obtaining an initial trajectory based on a spatial model; obtaining a target trajectory and a variable-radius pipeline based on the initial trajectory through cubic spline fitting; setting contour constraints and adaptive parameters based on the variable-radius pipeline; constructing an objective function based on the target trajectory, the contour constraints, and the adaptive parameters; pre-setting sideslip angle constraints for the VTOL UAV; and constructing a pipeline-based model prediction contour control framework based on the objective function, the contour constraints, the adaptive parameters, and the sideslip angle constraints. The pipeline-based model prediction contour control framework is used for planning and controlling the VTOL UAV. This invention can improve the performance and robustness of VTOL UAV planning and control and can be widely applied in the field of UAV control technology.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) control technology, and in particular to a pipeline-based tail-seat vertical take-off and landing (VTOL) UAV control method and apparatus. Background Technology

[0002] Current aircraft control schemes typically involve creating a trajectory with time information by considering a robot model and differential flatness at the front end, followed by tracking using methods such as PID or MPC at the back end. However, for vertical takeoff and landing fixed-wing aircraft, nonlinear aerodynamics makes the model more complex. If this framework is still used, the fixed time information calculated in the early stage based on potentially mismatched models may lead to error accumulation during later tracking, making it difficult to counteract disturbances in real time. Furthermore, the technical solution based on differential flatness has stringent conditions, requiring tedious and precise planning calculations based on differential flatness, resulting in poor real-time performance. Summary of the Invention

[0003] In view of this, the main objective of the embodiments of the present invention is to provide a pipeline-based tail-mounted vertical take-off and landing unmanned aerial vehicle (UAV) control method and device, in order to solve at least one of the problems of the prior art. The present invention can improve the performance and robustness of vertical take-off and landing UAV control.

[0004] To achieve the above objectives, one aspect of the present invention provides a pipeline-based tail-seat vertical takeoff and landing unmanned aerial vehicle (UAV) control method, the method comprising:

[0005] Obtain the initial trajectory based on the spatial model;

[0006] Based on the initial trajectory, the target trajectory and the variable radius pipe are obtained through cubic spline fitting;

[0007] Based on the variable radius pipe, set the contour constraints and adaptive parameters;

[0008] Based on the target trajectory, the contour constraints, and the adaptive parameters, a target function is constructed;

[0009] Pre-set sideslip angle constraints for vertical take-off and landing UAVs;

[0010] Based on the objective function, the contour constraint, the adaptive parameters, and the sideslip angle constraint, a pipeline-based model prediction contour control framework is constructed.

[0011] The pipeline-based model prediction profile control framework is used for planning and controlling vertical take-off and landing (VTOL) drones.

[0012] In some embodiments, obtaining the initial trajectory based on the spatial model includes the following steps:

[0013] Generate the first trajectory based on the convex polyhedron corridor;

[0014] Alternatively, a second trajectory can be generated based on the symbolic distance field.

[0015] In some embodiments, generating the first trajectory based on the convex polyhedron corridor includes the following steps:

[0016] Based on the fuselage dimensions of the vertical takeoff and landing UAV, the obstacle dimensions are expanded to generate environmental information;

[0017] The initial geometric path from the current position to the termination position is obtained through a path search algorithm;

[0018] Based on the environmental information and the initial geometric path, a convex polyhedron corridor is generated.

[0019] Construct a cost function based on the convex polyhedron corridor;

[0020] The first trajectory is generated based on the convex polyhedral corridor and the cost function.

[0021] In some embodiments, constructing the cost function based on the convex polyhedral corridor includes the following steps:

[0022] Obtain the minimum inscribed circle of the convex polyhedron corridor;

[0023] Obtain the first radius of the smallest inscribed circle, and use the length of the first radius as the first distance;

[0024] Obtain the exterior normal vector of each face of the convex polyhedron corridor;

[0025] Obtain the preset plane formed by moving the first preset point along the outer normal vector into the interior of the convex polyhedron corridor by the first distance;

[0026] Obtain the second distance from the first preset point to the preset plane;

[0027] If the second distance is positive, the cost and gradient are calculated using the smoothing loss function to obtain the cost function.

[0028] In some embodiments, the first arc length of the first trajectory is parameterized by a preset resolution to obtain the first discrete data point;

[0029] The target trajectory is obtained by performing cubic spline fitting on the first discrete data points;

[0030] Obtain the first position of each vertical take-off and landing UAV of the second arc length on the target trajectory;

[0031] Obtain the first velocity direction at the first position;

[0032] Obtain a first sectional plane perpendicular to the first velocity direction, and the first sectional plane intersects with the convex polyhedral corridor to form a convex polygon;

[0033] The shortest distance from the first position to each side of the convex polygon is taken as the second radius;

[0034] The variable radius pipe is obtained by performing a cubic spline fit on the second radius with respect to the second arc length.

[0035] In some embodiments, generating the second trajectory based on the symbolic distance field includes the following steps:

[0036] Based on the fuselage dimensions of the vertical takeoff and landing UAV, the obstacle dimensions are expanded to generate environmental information;

[0037] The initial geometric path from the current position to the termination position is obtained through a path search algorithm;

[0038] Based on the environmental information and the initial geometric path, a symbolic distance field is constructed;

[0039] The second trajectory is generated by performing trajectory planning in the symbolic distance field using the MINCO trajectory class.

[0040] In some embodiments, obtaining the target trajectory and the variable radius pipe based on the initial trajectory through cubic spline fitting includes the following steps:

[0041] By using a preset resolution, the third arc length of the second trajectory is parameterized to obtain the second discrete data point;

[0042] The target trajectory is obtained by performing cubic spline fitting on the second discrete data points;

[0043] The third distance between the discrete second preset point on the target trajectory and the nearest obstacle is taken as the third radius;

[0044] The variable radius pipe is obtained by performing cubic spline fitting on the third radius with respect to the third arc length of the target trajectory.

[0045] In some embodiments, setting the contour constraints and adaptive parameters according to the variable radius pipe includes the following steps:

[0046] Set the fourth arc length of the target trajectory to a preset trajectory;

[0047] The derivative of the variable radius pipe is obtained by taking the derivative of the variable radius pipe.

[0048] The contour constraint is obtained based on the preset trajectory, the variable radius pipe, and the derivative of the variable radius pipe;

[0049] Obtain the maximum and minimum radius of the target trajectory;

[0050] Obtain the maximum and minimum weights of the flight speed of the vertical take-off and landing unmanned aerial vehicle;

[0051] The adaptive parameters are obtained based on the maximum radius, minimum radius, maximum weight, and minimum weight.

[0052] In some embodiments, constructing the objective function based on the target trajectory, the contour constraint, and the adaptive parameters includes the following steps:

[0053] Set the fourth arc length of the target trajectory to a preset trajectory;

[0054] The current position of the vertical take-off and landing UAV is constrained by the contour constraint.

[0055] The tracking error at the current position of the vertical take-off and landing UAV from the preset trajectory is obtained, and the longitudinal error and the lateral contour error are obtained.

[0056] The flight speed of the vertical take-off and landing UAV is adjusted using the adaptive parameters.

[0057] The objective function is obtained based on the longitudinal error, the lateral profile error, the flight speed of the vertical take-off and landing UAV, and the rate of change of the state input.

[0058] To achieve the above objectives, another aspect of the present invention proposes a pipeline-based tail-seat vertical take-off and landing unmanned aerial vehicle (UAV) control device, the device comprising:

[0059] The first module is used to obtain the initial trajectory based on the spatial model;

[0060] The second module is used to obtain the target trajectory and the variable radius pipe by cubic spline fitting based on the initial trajectory.

[0061] The third module is used to set contour constraints and adaptive parameters according to the variable radius pipe;

[0062] The fourth module is used to construct an objective function based on the target trajectory, the contour constraints, and the adaptive parameters;

[0063] The fifth module is used to preset the sideslip angle constraints of vertical take-off and landing UAVs;

[0064] The sixth module is used to construct a pipeline-based model prediction contour control framework based on the objective function, the contour constraints, the adaptive parameters, and the sideslip angle constraints.

[0065] The pipeline-based model prediction profile control framework is used for planning and controlling vertical take-off and landing (VTOL) drones.

[0066] To achieve the above objectives, another aspect of the present invention provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned pipeline-based tail-seat vertical take-off and landing unmanned aerial vehicle (UAV) control method.

[0067] To achieve the above objectives, another aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned pipeline-based tail-seat vertical take-off and landing unmanned aerial vehicle (UAV) control method.

[0068] To achieve the above objectives, another aspect of the present invention provides a computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device can read the computer instructions from the computer-readable storage medium and execute the computer instructions, causing the computer device to perform the aforementioned pipeline-based tail-seat vertical takeoff and landing unmanned aerial vehicle (UAV) control method.

[0069] The embodiments of the present invention include at least the following beneficial effects: The present invention provides a pipeline-based tail-seat VTOL UAV control method and apparatus. This scheme obtains the initial trajectory through a spatial model; based on the initial trajectory, a target trajectory and a variable radius pipeline are obtained through cubic spline fitting, which reasonably describes the environmental information and can complete more complex obstacle avoidance tasks; based on the variable radius pipeline, contour constraints are set to make obstacle avoidance more accurate, and an adaptive parameter mechanism is introduced to better balance flight speed and accuracy, improving anti-interference performance; based on the target trajectory, the contour constraints, and the adaptive parameters, an objective function is constructed; the sideslip angle constraint of the VTOL UAV is pre-set to simplify the model and improve the real-time performance of the solution; based on the objective function, the contour constraints, the adaptive parameters, and the sideslip angle constraint, a pipeline-based model prediction contour control framework is constructed, which can improve the performance and robustness of VTOL UAV control. Attached Figure Description

[0070] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0071] Figure 1 This is a flowchart of the pipeline-based tail-seat vertical take-off and landing unmanned aerial vehicle (UAV) control method provided in an embodiment of the present invention;

[0072] Figure 2 This is a structural schematic diagram of a vertical take-off and landing unmanned aerial vehicle provided in an embodiment of the present invention;

[0073] Figure 3 This is a schematic diagram of the vertical take-off and landing unmanned aerial vehicle (UAV) provided in an embodiment of the present invention;

[0074] Figure 4 This is a schematic diagram of the front-end path planning trajectory and environmental information representation based on a convex polyhedral corridor provided in an embodiment of the present invention;

[0075] Figure 5 This is a schematic diagram of the front-end path planning trajectory and environmental information representation based on the symbolic distance field provided in an embodiment of the present invention;

[0076] Figure 6 This is a schematic diagram of the generation trajectory and pipeline-type environment characterization provided in an embodiment of the present invention;

[0077] Figure 7 This is a schematic diagram of trajectory error and contour constraint provided in an embodiment of the present invention;

[0078] Figure 8 This is a schematic diagram of contour error and progress error provided in an embodiment of the present invention;

[0079] Figure 9 This is a schematic diagram of the process framework of the pipeline-based VTOL overall planning and control scheme provided in the embodiment of the present invention;

[0080] Figure 10 This is a schematic diagram of the hardware structure of the electronic device provided in an embodiment of the present invention. Detailed Implementation

[0081] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this invention; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this invention as detailed in the appended claims.

[0082] It should be noted that although functional modules are divided in the system diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the system or the order in the flowchart. The terms "first / S100" and "second / S200" in the specification, claims, and the foregoing drawings may be used herein to describe various concepts, but unless specifically stated otherwise, these concepts are not limited by these terms. These terms are used only to distinguish one concept from another. For example, first information may also be referred to as second information without departing from the scope of the embodiments of the invention, and similarly, second information may also be referred to as first information. Depending on the context, the words "if" or "when" as used herein may be interpreted as "when," "in response to a determination," or "in the event of a determination."

[0083] The terms “at least one,” “multiple,” “each,” “any,” etc., used in this invention, “at least one” includes one, two, or more than two; “multiple” includes two or more than two; “each” refers to each of the corresponding multiple; and “any” refers to any one of the multiple.

[0084] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein is for the purpose of describing embodiments of the invention only and is not intended to limit the invention.

[0085] With the continuous development of unmanned technology, rotary-wing drones are being used more and more widely. However, their endurance problem has not been properly solved. Fixed-wing drones, utilizing their aerodynamic advantages, have better endurance and are more advantageous in carrying people and goods, but they have high requirements for take-off and landing sites and cannot hover at any time. Tail-seat vertical take-off and landing drones combine the advantages of high flight efficiency of fixed-wing drones and low dependence on take-off and landing sites of rotary-wing drones, and have broad application prospects. However, completing their navigation tasks requires appropriate planning and control schemes.

[0086] For quadrotor tracking, a layered approach is typically used. Trajectory planning is completed based on the quadrotor's differential flatness, followed by tracking using proportional-integral-derivative (PID) controllers or model predictive control (MPC). For quadrotors, there is also a corridor-based model predictive profile controller (CMPCC), which uses the flight corridor as a hard safety constraint, considers environmental obstacle avoidance, and achieves quadrotor trajectory tracking. For VTOL fixed-wing aircraft, relatively advanced research also involves calculating the differential flatness of the VTOL fixed-wing aircraft, designing trajectory planning based on this, and then using MPC for tracking. For canard-tailed VTOL UAVs, a model predictive profile controller (MPCC) is used to implement trajectory tracking.

[0087] Current aircraft control schemes typically involve creating a trajectory with time information by considering a robot model and differential flatness at the front end, followed by tracking using methods such as PID or MPC at the back end. However, for vertical takeoff and landing (VTOL) fixed-wing aircraft, nonlinear aerodynamics makes the model more complex. If this framework is still used, the fixed time information calculated in the early stage based on potentially mismatched models may lead to error accumulation during later tracking, making it difficult to counteract disturbances in real time. Furthermore, the technical solution based on differential flatness has stringent conditions, requiring tedious and precise planning calculations based on differential flatness, resulting in poor real-time performance, and differential flatness cannot be derived for all aircraft types. While existing tracking schemes for canard-tailed VTOL UAVs only require a geometric path, they track the UAV trajectory by manually providing a fixed reference geometric path, without calculating an excellent geometric path using a high-level planning framework. Moreover, this algorithm does not incorporate environmental considerations in real time for obstacle avoidance, making it difficult to overcome complex environments. This algorithm is limited to only one tracking scheme. Compared to canard MPCC, CMPCC considers environmental information, but its trajectory and corridor are pre-defined using a teach-repeat-replan approach. This means the trajectory doesn't closely follow the safe zone, leaving too little safety margin for tracking. In narrow areas, it's prone to grazing the edge of the corridor, posing a certain risk. It doesn't fully utilize space, resulting in lower safety for UAVs. Furthermore, the contour constraint representation is imprecise and cannot fully leverage prior environmental information to provide guidance on trajectory tracking.

[0088] In view of this, such as Figure 1 As shown, this embodiment of the invention provides a pipeline-based tail-seat vertical takeoff and landing unmanned aerial vehicle (UAV) control method, which may include, but is not limited to, steps S100 to S600:

[0089] Step S100: Obtain the initial trajectory based on the spatial model;

[0090] Step S200: Based on the initial trajectory, obtain the target trajectory and the variable radius pipe through cubic spline fitting;

[0091] Step S300: Set contour constraints and adaptive parameters according to the variable radius pipe;

[0092] Step S400: Construct an objective function based on the target trajectory, the contour constraint, and the adaptive parameters;

[0093] Step S500: Preset the sideslip angle constraint of the vertical take-off and landing UAV;

[0094] Step S600: Construct a pipeline-based model prediction contour control framework based on the objective function, the contour constraint, the adaptive parameters, and the sideslip angle constraint.

[0095] The pipeline-based model prediction profile control framework is used for planning and controlling vertical take-off and landing (VTOL) drones.

[0096] In steps S100 to S600 of some embodiments, a pipeline environmental representation scheme is designed within the framework of planning and controlling the VTOL UAV to describe the safe zone. A spatial model is used to obtain a trajectory closer to the safe space in the environment and a larger pipeline generated around this trajectory. This not only reasonably describes the environmental information, enabling more complex obstacle avoidance tasks, but also provides more positional margin for later tracking. Furthermore, the generated path representation is concise, contains no time information, and does not require differential flatness, thus avoiding mismatch errors in the front-end planning stage, and is usable by all UAV models. Since the pipeline is differentiable, more precise contour constraints are designed for it, providing some predictability for the future environment. An adaptive parameter mechanism is introduced into the pipeline-based Model Predict Contouring Control (PMPCC) scheme, which better balances flight speed and accuracy, resulting in better anti-disturbance performance. Additionally, sideslip angle constraints are introduced using the physical characteristics of the VTOL UAV to simplify the model and ensure real-time solution.

[0097] In some embodiments, for example Figure 2 By studying the vertical takeoff and landing UAVs shown, we can establish, for example... Figure 3 The mechanical system of the vertical takeoff and landing (VTOL) UAV shown is used to construct a mathematical model of the VTOL UAV. For example, the mathematical model may include a motion mathematical model in the geodetic coordinate system (world frame) and aerodynamic forces in the body coordinate system (body frame).

[0098] Motion Mathematical Model (World System):

[0099]

[0100] In the formula, The derivative of the UAV's position in the inertial coordinate system; v represents the UAV's flight velocity in the inertial coordinate system; R represents the UAV's attitude from the world frame; a T x represents the acceleration produced by the thrust; b ,y b ,z b The coordinate system axes represent the machine body; The derivative of the UAV's flight velocity in the inertial coordinate system; g represents the acceleration due to gravity; f represents the reciprocal of the mass of the drone; a This represents the aerodynamic forces in the carrier coordinate system.

[0101] Aerodynamic forces in the carrier / body coordinate system:

[0102]

[0103] In the formula, f a f represents the aerodynamic force in the carrier coordinate system; ax f ay f az ρ represents the components of aerodynamic force in the airflow coordinate system; α represents the angle of attack of the UAV; β represents the sideslip angle; L represents the lift force on the UAV; D represents the drag force on the UAV; Y represents the sideslip force on the UAV; ρ 空 V represents air density; V represents the modulus of airspeed in an inertial frame of reference. a ||;S represents wing area;C L Indicates the lift coefficient; C D Indicates the drag coefficient; C Y Indicates the sideslip force coefficient; (·) T Indicates the transpose operation; w represents the wind speed in the inertial frame; v represents the flight speed of the UAV in the inertial coordinate system; v a This represents the airspeed in an inertial frame of reference; (·) B Indicates body series; Indicates the airspeed of the engine system; These represent the airspeed v in the engine system. a Components on the x, y, and z axes.

[0104] Among them, the lift coefficient, drag coefficient, and sideslip coefficient are usually nonlinear functions obtained through wind tunnel experiments.

[0105] For example, assuming the aircraft model is symmetrical in the XZ plane of the fuselage, then the following expression holds:

[0106]

[0107] Assuming coordinated flight, the sideslip angle β = 0, that is... In the absence of wind, the y-direction velocity of the machine system is 0, and the sideslip force coefficient C Y (α,β)=0, the model can be simplified by using this sideslip angle constraint.

[0108] In step S100 of some embodiments, based on the pipeline description of the safe area, a trajectory closer to the safe space in the environment is generated through two different spatial model schemes. Optionally, the spatial model can be a convex polyhedral corridor or a symbolic distance field. Based on the convex polyhedral corridor, a first trajectory can be generated, and based on the symbolic distance field, a second trajectory can be generated. The first trajectory and the second trajectory are the initial trajectory.

[0109] In some embodiments, the step of generating the first trajectory based on the convex polyhedral corridor may include, but is not limited to, steps S111 to S115:

[0110] Step S111: Based on the fuselage size of the vertical take-off and landing UAV, expand the size of the obstacle to generate environmental information;

[0111] Step S112: Obtain the initial geometric path from the current position to the termination position using a path search algorithm;

[0112] Step S113: Generate a convex polyhedron corridor based on the environmental information and the initial geometric path;

[0113] Step S114: Construct a cost function based on the convex polyhedron corridor;

[0114] Step S115: Generate the first trajectory based on the convex polyhedron corridor and the cost function.

[0115] In steps S111 to S113 of some embodiments, the obstacles at the front end are expanded according to the drone's fuselage size. Then, a fast randomized tree search (RRT) or Astar algorithm is used to search for an initial geometric path from the current position to the endpoint. Based on the obstacle environment, a convex polyhedral corridor is generated by combining the initial geometric path and the gcopter method from the open-source project. These convex polyhedral corridors define the area where the drone can move in space, ensuring that the drone's trajectory meets mission requirements without violating spatial constraints.

[0116] In steps S114 to S115 of some embodiments, by constructing a partially optimized cost function during the process of generating the convex polyhedron corridor, a first trajectory as close as possible to the center of the convex polyhedron flight corridor is obtained based on the convex polyhedron corridor and the constructed cost function.

[0117] In some embodiments, step S114 may include, but is not limited to, steps S1141 to S1146:

[0118] Step S1141: Obtain the minimum inscribed circle of the convex polyhedron corridor;

[0119] Step S1142: Obtain the first radius of the smallest inscribed circle, and use the length of the first radius as the first distance;

[0120] Step S1143: Obtain the exterior normal vector of each face of the convex polyhedron corridor;

[0121] Step S1144: Obtain the preset plane formed by moving the first preset point along the outer normal vector into the interior of the convex polyhedron corridor by the first distance;

[0122] Step S1145: Obtain the second distance from the first preset point to the preset plane;

[0123] Step S1146: If the second distance is positive, then use the smooth loss function to calculate the cost and gradient to obtain the cost function.

[0124] In some embodiments, steps S1141 to S1146, such as Figure 4 As shown, the gcopter method, an open-source project for efficient and versatile multi-rotor aircraft trajectory optimizers, can obtain the interior point of each convex polyhedron, i.e., the center O of the smallest inscribed circle, and also the first radius mindist of this smallest inscribed circle. The length of this first radius mindist is used as the first distance. For example, when the first preset point on the preset flight trajectory is outside the faces of the convex polyhedron, its cost can be calculated as follows:

[0125] 1. For each face of a convex polyhedron, calculate the outward normal vector (normal) of that face;

[0126] 2. Calculate the distance from the first preset point to the surface that moves mindisy into the convex polyhedron along the outer normal vector of the surface. That is, first obtain the preset plane formed by moving the first preset point a first distance into the interior of the convex polyhedron along the outer normal vector, and then obtain the second distance from the first preset point to the preset plane.

[0127] 3. Determine if the second distance is positive. If the second distance is positive, use the smooth loss function to calculate the cost and gradient.

[0128] Optionally, a second distance from the first preset point to the preset plane can be calculated using a dot product and an offset. The sign of the second distance indicates whether the first preset point is on the opposite side of the preset plane. If the second distance is negative, it means that the first preset point is on the side of the preset plane facing the convex polyhedron; if the second distance is positive, it means that the first preset point is outside the convex polyhedron. The expression for the smoothing loss function is as follows:

[0129]

[0130] In the formula, smoothedL1(μ,s) represents the smoothing loss function; s represents the second distance from the first preset point to the preset plane; μ represents the preset segmentation interval, which is generally small.

[0131] Compared to conventional methods where the environmental obstacle cost is consistently zero at any location within the corridor, with only the area outside the corridor having a cost, causing the trajectory to easily approach the corridor edge and hindering tracking, this invention, through the cost and gradient calculation method of the aforementioned cost function, allows the trajectory to no longer be constrained within the convex polyhedron, but rather to be as close as possible to points within the convex polyhedron. Therefore, a trajectory as close as possible to the center of the convex polyhedron flight corridor can be obtained. Since this invention only uses the geometric path information of the trajectory rather than temporal information, the resulting changes in time optimization have no impact.

[0132] In some embodiments, the step of generating the second trajectory based on the symbolic distance field may include, but is not limited to, steps S121 to S124:

[0133] Step S121: Based on the fuselage size of the vertical take-off and landing UAV, expand the size of the obstacle to generate environmental information;

[0134] Step S122: Obtain the initial geometric path from the current position to the termination position using a path search algorithm;

[0135] Step S123: Construct a symbolic distance field based on the environmental information and the initial geometric path;

[0136] Step S124: Perform trajectory planning in the symbol distance field using the MINCO trajectory class to generate the second trajectory.

[0137] In steps S121 to S124 of the game implementation, obstacles at the front end are expanded according to the drone's fuselage size. Then, a Fast Random Tree Search (RRT) or Astar algorithm is used to search for an initial geometric path from the current position to the endpoint. Next, based on the obstacle environment and the initial geometric path, a symbolic distance field (ESDF) is constructed. Using the MINCO trajectory class from the open-source project gcopter, a smooth second trajectory that is far from the obstacles is planned in the symbolic distance field. In the MINCO (Minimum Control) trajectory class, intermediate points and time vectors on the trajectory are used to parameterize the trajectory, meaning that the trajectory can be generated and optimized by adjusting the positions of these intermediate points and the time allocated to each segment. For any given intermediate point and time vector, a trajectory that satisfies the optimality condition can be uniquely determined. These intermediate points, together with the time vector, define the global shape of the trajectory and the specific duration of each segment.

[0138] In step S200 of some embodiments, the target trajectory and the variable radius pipe can be obtained by cubic spline fitting based on the initial trajectory. Optionally, the target trajectory and the variable radius pipe can be obtained by cubic spline fitting based on the first trajectory generated based on the convex polyhedral corridor; or the target trajectory and the variable radius pipe can be obtained by cubic spline fitting based on the second trajectory generated based on the signed distance field.

[0139] In some embodiments, the step of obtaining the target trajectory and the variable radius pipe based on the first trajectory generated by the convex polyhedral corridor through cubic spline fitting may include, but is not limited to, steps S211 to S217:

[0140] Step S211: Parameterize the first arc length of the first trajectory using a preset resolution to obtain the first discrete data point;

[0141] Step S212: Perform cubic spline fitting on the first discrete data points to obtain the target trajectory;

[0142] Step S213: Obtain the first position of each second arc length of the vertical take-off and landing UAV on the target trajectory;

[0143] Step S214: Obtain the first velocity direction at the first position;

[0144] Step S215: Obtain a first sectional surface perpendicular to the first velocity direction, and the first sectional surface intersects with the convex polyhedral corridor to form a convex polygon;

[0145] Step S216: Take the shortest distance from the first position to each side of the convex polygon as the second radius;

[0146] Step S217: Perform cubic spline fitting on the second radius with respect to the second arc length to obtain the variable radius pipe.

[0147] In steps S211 to S217 of some embodiments, after generating the first trajectory based on the convex polyhedral corridor, the arc length is parameterized according to a preset resolution, that is, the first arc length of the first trajectory is parameterized to obtain the first discrete data points. The target trajectory P(θ) can be obtained by fitting the first discrete data points using cubic splines. For example, the following expression can be obtained using cubic spline fitting:

[0148] P(θ)=[X(θ) Y(θ) Z(θ)] T (6)

[0149] In the formula, P(θ) represents the target trajectory, which is the reference trajectory in three-dimensional space used for backend tracking. It consists of reference trajectories in three dimensions: X(θ), Y(θ), and Z(θ), with the independent variable being the arc length θ. At each second arc length θ on the target trajectory, the first position P of the VTOL UAV and the first velocity direction v at that first position are obtained. Then, a tangent Φ perpendicular to the first velocity direction is made at that first position. The tangent intersects with the corridor of the convex polyhedron to form a convex polygon. The shortest distance from the first position P to each side of the convex polygon is taken as the second radius r, as shown below. Figure 4 As shown, the second radius r represents the environmental constraints imposed on the first position P. By performing a cubic spline fit on the second radius r with respect to the second arc length θ, the variable radius pipe R(θ) can be obtained, which leads to the following... Figure 6 Since the cubic spline is differentiable, the derivative of the target trajectory P′(θ) and the derivative of the variable radius pipe R′(θ) can also be obtained by taking the derivative.

[0150] In some embodiments, the step of obtaining the target trajectory and the variable radius pipe by cubic spline fitting based on the second trajectory generated by the symbolic distance field may include, but is not limited to, steps S221 to S224:

[0151] Step S221: Parameterize the third arc length of the second trajectory using a preset resolution to obtain the second discrete data points;

[0152] Step S222: Perform cubic spline fitting on the second discrete data points to obtain the target trajectory;

[0153] Step S223: The third distance between the discrete second preset point on the target trajectory and the nearest obstacle is taken as the third radius;

[0154] Step S224: Perform cubic spline fitting on the third radius with respect to the third arc length of the target trajectory to obtain the variable radius pipe.

[0155] In steps S221 to S224 of some embodiments, after generating the second trajectory based on the symbolic distance field, the arc length is parameterized according to a preset resolution, that is, the third arc length θ of the second trajectory is parameterized to obtain the second discrete data points. The second discrete data points are fitted using cubic splines, and the target trajectory P(θ) can be obtained by the same formula (6). At the second preset point discrete on the target trajectory (i.e., the second position P of the vertical take-off and landing UAV), its ESDF value is obtained. The ESDF value is the third distance from the second position to the nearest obstacle. The third distance is denoted as the third radius r. A sphere is drawn at the second position with the third radius as the radius, and the following can be obtained: Figure 5 The unobstructed safety area shown includes, Figure 5 The x and y coordinates are coordinates in a two-dimensional plane, and the unit can be expressed in meters (m). Therefore, using the third radius as a representation of the environmental constraints at the second position, and performing a cubic spline fit on the third radius with respect to the third arc length, we can obtain the variable radius pipe R(θ), which then yields the following... Figure 6 Since the cubic spline is differentiable, the derivative of the target trajectory P′(θ) and the derivative of the variable radius pipe R′(θ) can also be obtained by taking the derivative.

[0156] In some optional embodiments, a three-dimensional target trajectory P(θ) is optimized using two optional schemes (based on a convex polyhedral corridor and based on a symbolic distance field) to obtain a target trajectory closer to the safe space in the environment. A larger pipe is then generated around this trajectory, and the environment is re-represented as a variable-radius pipe R(θ), such as... Figure 6 All figures use arc length θ as the variable.

[0157] In some embodiments, a pipe-based model prediction contour control (PMPCC) controller framework is constructed, including acquiring the target trajectory to be tracked and the variable radius pipe, as well as the state variables selected by PMPCC, the contour constraints based on the pipe, the adaptive parameters based on the pipe, the objective function of the PMPCC controller, and the VTOL sideslip angle constraints, which will be described in detail later.

[0158] In some embodiments, the state variables selected by PMPCC may include, but are not limited to, the following state variables:

[0159]

[0160] Where x is the state variable of the vertical takeoff and landing UAV; p, v, and q are the position, velocity, and quaternion in the world system, respectively; f tω is the thrust; ω is the angular velocity; θ is the arc length; u is the control input, derived from the rate of change of arc length V. θ Thrust change rate df t It consists of angular acceleration dω. Combining aerodynamic data from wind tunnel tests, substituting the kinematic and dynamic formulas of the above-mentioned mathematical model of motion into the equation of state, we obtain the equation of state. The third position of the vertical takeoff and landing UAV at time step k is represented as p. k =[x k ,y k ,z k ] T .

[0161] In some embodiments, step S300 may include, but is not limited to, steps S310 to S360:

[0162] Step S310: Set the fourth arc length of the target trajectory to a preset trajectory;

[0163] Step S320: Differentiate the variable radius pipe to obtain the derivative of the variable radius pipe;

[0164] Step S330: Obtain the contour constraint based on the preset trajectory, the variable radius pipe, and the derivative of the variable radius pipe;

[0165] Step S340: Obtain the maximum radius and minimum radius of the target trajectory;

[0166] Step S350: Obtain the maximum and minimum weights of the flight speed of the vertical take-off and landing unmanned aerial vehicle;

[0167] Step S360: Obtain the adaptive parameters based on the maximum radius, minimum radius, maximum weight, and minimum weight.

[0168] In some embodiments, steps S310 to S360, such as Figure 7 As shown, in order to make the third position p k As far as possible, within the cone that characterizes environmental constraints, that is, to minimize the lateral profile error. Since the environmental radius R is smaller than that here, the following contour constraints were designed:

[0169]

[0170] In the contour constraint, this embodiment of the invention uses the software package acados, which can provide a fast nonlinear optimal control solver suitable for embedded systems, to solve the optimal control of formula (8). This process is repeated in real time.

[0171] In the formula, This represents the projection profile error obtained by projecting the third position onto the tangent plane; Indicates the preset trajectory The radius of the non-uniform pipe at that location; θ represents the preset trajectory, referring to the preset trajectory in the current optimal control solution; linearized The preset trajectory in the previous optimal control solution; R(θ) linearized ) represents θ linearized The radius of the non-uniform pipe at the location; R′(θ) linearized ) represents θ linearized The derivative of the radius of the non-uniform pipe at that location.

[0172] In this embodiment of the invention, the non-uniform pipe is a complete space allowing aircraft to fly. At each arc length θ, the pipe radius R(θ) and the rate of change of radius, i.e., the derivative of the pipe radius R′(θ), can be obtained, and the non-uniform pipe can be approximated at that location. If the rate of change of radius at that location is ignored, and only the pipe radius is considered, then the pipe at that location is approximately barrel-shaped. If the rate of change of radius at that location is also considered, then the pipe at that location is approximately cone-shaped.

[0173] Since the pipe is differentiable, a continuous R′(θ) can be obtained. This takes into account the tendency of the cone to tighten or expand as θ changes, and simulates the shape changes of the environment in front to a certain extent. This environmental constraint is more accurate than that of a pure bucket and has a certain predictive ability.

[0174] In this embodiment of the invention, the arc length change rate V θ The weighting coefficient is denoted as ρ. This weighting coefficient ρ has preset maximum and minimum values, and it will adaptively adjust based on the maximum and minimum values ​​of the radius R of the entire variable-diameter uniform pipe. Where R is larger, meaning a wider range of aircraft movement is allowed, the arc length variation rate V... θ The weighting coefficient ρ will be larger, which will result in a higher flight speed for the aircraft.

[0175] For example, the rate of change of arc length V in the constructed objective function θ The weighting coefficient ρ can be dynamically adjusted based on the variable radius pipe R(θ), taking the maximum radius R of the entire target trajectory. max and minimum radius R min Take the rate of change of arc length V θ Maximum weight ρ max and minimum weight ρ min The weight coefficients ρ of each time step within the prediction range are set according to the preset trajectory θ from the previous optimal control solution. linearizedThe radius R corresponding to the point is re-linearly mapped so that a larger R results in a larger weight coefficient ρ, allowing the VTOL UAV to travel faster; conversely, a smaller R results in a smaller weight coefficient ρ, causing the VTOL UAV to travel slower. The adaptive parameter weight ρ(θ) then has the following expression:

[0176]

[0177] By introducing an adaptive parameter mechanism, a better balance can be struck between flight speed and accuracy, resulting in improved anti-interference performance.

[0178] In some embodiments, step S400 may include, but is not limited to, steps S410 to S450:

[0179] Step S410: Set the fourth arc length of the target trajectory to a preset trajectory;

[0180] Step S420: Constrain the current position of the vertical take-off and landing UAV using the contour constraint;

[0181] Step S430: Obtain the tracking error of the current position of the vertical take-off and landing UAV at the preset trajectory, and obtain the longitudinal error and the lateral profile error;

[0182] Step S440: Adjust the flight speed of the vertical take-off and landing UAV using the adaptive parameters;

[0183] Step S450: Obtain the objective function based on the longitudinal error, the lateral profile error, the flight speed of the vertical take-off and landing UAV, and the rate of change of the state input.

[0184] In standard MPC, the objective function is the position p of the UAV at prediction time step k. k Its reference position p refer The distance difference. Reference position p refer Typically, tracking data comes from a front-end trajectory that contains time information, usually starting from the start of tracking and recording the position point on the trajectory at the corresponding moment. MPCC, however, does not require time information from the front-end trajectory; instead, it utilizes arc length information θ, and its tracking position is determined by the arc length θ. Theoretically, MPCC can track positions at position p... k Attempting to track the trajectory at a distance p k nearest point Since finding the nearest point is an optimization process, to avoid nested optimizations, in this embodiment of the invention, θ is designed as a preset trajectory in the state space. Current third position p k distance The tracking error at that point is as follows Figure 7 As shown, it is divided into longitudinal errors and lateral profile error For example, a preset trajectory is taken. Draw a tangent plane perpendicular to the tangent at the point. p k To the tangent plane Projection can obtain and This can approximate the longitudinal error. and lateral profile error These two errors will be included as part of the objective function. The other part of the objective function is the rate of change of arc length, V. θ Multiplying by the negative value after weighting allows the aircraft to maintain a certain speed while keeping tracking errors small. Additionally, the rate of change of the state input, dU... k This will also be part of the objective function to ensure the smoothness of the control input. Therefore, the expression for the objective function can be obtained as follows:

[0185]

[0186] Where J represents the objective function; N is the prediction step size; k is the current time step; e l (θ k θ is the arc length at time step k. k Tracking progress error at the location; e c (θ k θ is the arc length at time step k. k Contour error at location; dU k V is the rate of change of the input at time step k. θ It is the rate of change of arc length; Q l Q c Q u ρ are the weighting coefficients of the four costs.

[0187] In some embodiments, such as Figure 8 As shown, let the current state point be P, P(θ) k The state above is the desired tracking point at step size k. The total error between the current state point and the desired tracking point is e = PP(θ). k Let P(θ) be the expected tracking point. k The unit vector of the tangent line at point () is t. k ,but P′(θ k ) is P(θ k ) at θ k The derivative with respect to θ is then:

[0188] e l (θ k )=t k ·(PP(θ k))

[0189] |e c (θ k )|=|t k ×(PP(θ k (11)

[0190] By setting the upper bound of the decision variable θ to the total arc length of the entire target trajectory, stopping at the endpoint can be achieved naturally. For example, by applying an upper bound to the decision variables in formula (7), namely the VTOL UAV state variable x and the control input variable u, the upper bound of x is x max The upper bound of u is u max , where θ max Belongs to x max In Acads, the upper bound θ of the arc length θ in the decision variables is... max Setting it to the total arc length of the entire target trajectory means that the arc length θ traveled must not exceed the total arc length of the entire target trajectory, which allows the aircraft to stop tracking the trajectory once it reaches the endpoint.

[0191] In step S500 of some embodiments, in the analysis of the VTOL UAV, the physical characteristics of the VTOL UAV are utilized, assuming a sideslip angle of 0. This allows lateral forces to be ignored, simplifying the model, reducing the difficulty of the solution, and making the solution easier to complete in real time. For example, the constraint of a sideslip angle of 0 is applied in the following manner:

[0192]

[0193] in, It is the airspeed v in an inertial frame of reference. a The y-axis component in the body frame. Inequality constraints can be implemented using the external cost of acados, and nonlinear optimization is solved using acados.

[0194] In step S600 of some embodiments, the pipeline-based model predictive profile control framework (PMPCC) is constructed using an objective function, profile constraints, adaptive parameters, and sideslip angle constraints. The pipeline-based model predictive profile control framework is used for planning and controlling a vertical takeoff and landing unmanned aerial vehicle (UAV).

[0195] like Figure 9As shown in the figure, an embodiment of the present invention provides a pipeline-based tail-seat VTOL unmanned aerial vehicle (UAV) control method. This method designs a pipeline as an environmental representation to describe the safe zone. Two schemes (based on a convex polyhedral corridor and based on a symbolic distance field) are used to generate a target trajectory P(θ) closer to the safe space in the environment. The environmental representation is reconstructed to obtain a larger pipeline R(θ) surrounding this trajectory. Based on R(θ), more precise contour constraints are designed for tail-seat tracking. An adaptive parameter mechanism is introduced, and tracking is performed using the PMPCC framework. The sideslip angle constraint is introduced using the physical characteristics of VTOL to simplify the model. The required thrust α is calculated using acados. T The underlying Px4 firmware calculates throttle and torque τ, and tracks thrust a in real time, along with angular velocity ω. T In the Px4 firmware, thrust and angular velocity ω are tracked through proportional gain K and PID control.

[0196] This invention also provides a pipeline-based tail-seat VTOL UAV control device, which can implement the above-described pipeline-based tail-seat VTOL UAV control method. The device includes:

[0197] The first module is used to obtain the initial trajectory based on the spatial model;

[0198] The second module is used to obtain the target trajectory and the variable radius pipe by cubic spline fitting based on the initial trajectory.

[0199] The third module is used to set contour constraints and adaptive parameters according to the variable radius pipe;

[0200] The fourth module is used to construct an objective function based on the target trajectory, the contour constraints, and the adaptive parameters;

[0201] The fifth module is used to preset the sideslip angle constraints of vertical take-off and landing UAVs;

[0202] The sixth module is used to construct a pipeline-based model prediction contour control framework based on the objective function, the contour constraints, the adaptive parameters, and the sideslip angle constraints.

[0203] The pipeline-based model prediction profile control framework is used for planning and controlling vertical take-off and landing (VTOL) drones.

[0204] It is understood that the content of the above method embodiments is applicable to the present device embodiments. The specific functions implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0205] This invention also provides an electronic device, which includes a processor and a memory. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned pipeline-based tail-seat vertical takeoff and landing (VTOL) unmanned aerial vehicle (UAV) control method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0206] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0207] refer to Figure 10 , Figure 10 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes:

[0208] The processor 701 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present invention.

[0209] The memory 702 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 702 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 702 and is called and executed by the processor 701 to implement a pipeline-based tail-seat vertical take-off and landing unmanned aerial vehicle (UAV) control method according to an embodiment of the present invention.

[0210] The input / output interface 703 is used to implement information input and output;

[0211] The communication interface 704 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0212] Bus 705 transmits information between various components of the device (e.g., processor 701, memory 702, input / output interface 703, and communication interface 704);

[0213] The processor 701, memory 702, input / output interface 703, and communication interface 704 are connected to each other within the device via bus 705.

[0214] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described pipeline-based tail-seat vertical take-off and landing unmanned aerial vehicle (UAV) control method.

[0215] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0216] This invention also provides a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device can read the computer instructions from the computer-readable storage medium and execute the computer instructions, causing the computer device to perform the aforementioned pipeline-based tail-seat vertical takeoff and landing unmanned aerial vehicle (UAV) control method.

[0217] In summary, the tail-mounted vertical take-off and landing unmanned aerial vehicle (UAV) control method and device based on a pipeline according to embodiments of the present invention has the following advantages:

[0218] 1. This invention provides a complete planning and control scheme, offering two methods to obtain a trajectory closer to a safe space in the environment and to generate a larger pipe around this trajectory. This not only reasonably describes environmental information and enables more complex obstacle avoidance tasks, but also provides more positional margin for subsequent tracking.

[0219] 2. Compared to schemes that use differential flatness for planning and then MPC tracking, the path representation generated in this embodiment is simpler and does not pre-determine time information, thus avoiding mismatch errors in the front-end planning stage and providing better real-time anti-interference capability. It eliminates the need for differential flatness, avoiding tedious and demanding calculations, and is available for all models, making it more widely applicable.

[0220] 3. Compared with the quadcopter CMPCC scheme, the trajectory of the embodiment of the present invention is closer to the safe area, making full use of space, leaving more safety margin for tracking, avoiding rubbing against the edge of the corridor in narrow places, and thus providing better safety.

[0221] 4. Compared with the quadcopter CMPCC scheme, the environmental characterization of the pipeline in this embodiment of the invention is differentiable. Utilizing this characteristic, a more accurate conical profile constraint is designed, which can use the prior information of the area in front to predict the future environment to a certain extent, and the obstacle avoidance is more accurate.

[0222] 5. The embodiments of the present invention introduce an adaptive parameter mechanism based on the pipeline, which can better balance flight speed and accuracy, and has better anti-interference performance.

[0223] 6. The embodiments of the present invention utilize the physical characteristics of VTOL to introduce sideslip angle constraints to simplify the model and improve the real-time performance of the solution.

[0224] In some alternative embodiments, the functions / operations mentioned in the block diagrams may not occur in the order shown in the operation diagrams. For example, depending on the functions / operations involved, two consecutively shown blocks may actually be executed substantially simultaneously, or the blocks may sometimes be executed in reverse order. Furthermore, the embodiments presented and described in the flowcharts of this invention are provided by way of example to provide a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is altered and sub-operations described as part of a larger operation are executed independently.

[0225] Furthermore, although the invention has been described in the context of functional modules, it should be understood that, unless otherwise stated, one or more of the described functions and / or features may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in a separate physical device or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the invention. Rather, given the properties, functions, and internal relationships of the various functional modules in the apparatus disclosed herein, the actual implementation of the module will be understood within the scope of conventional skill of an engineer. Therefore, those skilled in the art can implement the invention as set forth in the claims using ordinary techniques without excessive experimentation. It is also understood that the specific concepts disclosed are merely illustrative and not intended to limit the scope of the invention, which is determined by the full scope of the appended claims and their equivalents.

[0226] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0227] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0228] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0229] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0230] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0231] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

[0232] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the embodiments described. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of the present invention.

Claims

1. A pipeline-based tail-seat vertical takeoff and landing unmanned aerial vehicle (UAV) control method, characterized in that, Includes the following steps: Obtaining the initial trajectory based on the spatial model includes the following steps: Generate the first trajectory based on the convex polyhedron corridor; Alternatively, a second trajectory can be generated based on the symbolic distance field; Based on the initial trajectory, the target trajectory and the variable radius pipe are obtained through cubic spline fitting, including the following steps: The first arc length of the first trajectory is parameterized by a preset resolution to obtain a first discrete data point; cubic spline fitting is performed on the first discrete data point to obtain the target trajectory; the first position of the vertical take-off and landing UAV with a second arc length on the target trajectory is obtained; the first velocity direction of the first position is obtained; a first sectional surface perpendicular to the first velocity direction is obtained, and the first sectional surface intersects with the convex polyhedral corridor to form a convex polygon; the shortest distance from the first position to each side of the convex polygon is taken as the second radius; cubic spline fitting is performed on the second radius with respect to the second arc length to obtain the variable radius pipe; or, The third arc length of the second trajectory is parameterized by a preset resolution to obtain a second discrete data point; cubic spline fitting is performed on the second discrete data point to obtain the target trajectory; the third distance between the discrete second preset point on the target trajectory and the nearest obstacle is taken as the third radius; cubic spline fitting is performed on the third radius with respect to the third arc length of the target trajectory to obtain the variable radius pipe; Based on the variable radius pipe, the contour constraints and adaptive parameters are set, including the following steps: Set the fourth arc length of the target trajectory to a preset trajectory; The derivative of the variable radius pipe is obtained by taking the derivative of the variable radius pipe. The contour constraint is obtained based on the preset trajectory, the variable radius pipe, and the derivative of the variable radius pipe; Obtain the maximum and minimum radius of the target trajectory; Obtain the maximum and minimum weights of the flight speed of the vertical take-off and landing UAV; The adaptive parameters are obtained based on the maximum radius, minimum radius, maximum weight, and minimum weight. Based on the target trajectory, the contour constraints, and the adaptive parameters, a target function is constructed; Pre-set sideslip angle constraints for vertical take-off and landing UAVs; Based on the objective function, the contour constraint, the adaptive parameters, and the sideslip angle constraint, a pipeline-based model prediction contour control framework is constructed. The pipeline-based model prediction profile control framework is used for planning and controlling vertical take-off and landing (VTOL) drones.

2. The pipeline-based tail-seat vertical takeoff and landing UAV control method according to claim 1, characterized in that, The generation of the first trajectory based on the convex polyhedron corridor includes the following steps: Based on the fuselage dimensions of the vertical takeoff and landing UAV, the obstacle dimensions are expanded to generate environmental information; The initial geometric path from the current position to the termination position is obtained through a path search algorithm; Based on the environmental information and the initial geometric path, a convex polyhedron corridor is generated. Construct a cost function based on the convex polyhedron corridor; The first trajectory is generated based on the convex polyhedral corridor and the cost function.

3. The pipeline-based tail-seat vertical takeoff and landing UAV control method according to claim 2, characterized in that, The construction of the cost function based on the convex polyhedron corridor includes the following steps: Obtain the minimum inscribed circle of the convex polyhedron corridor; Obtain the first radius of the smallest inscribed circle, and use the length of the first radius as the first distance; Obtain the exterior normal vector of each face of the convex polyhedron corridor; Obtain the preset plane formed by moving the first preset point along the outer normal vector into the interior of the convex polyhedron corridor by the first distance; Obtain the second distance from the first preset point to the preset plane; If the second distance is positive, the cost and gradient are calculated using the smoothing loss function to obtain the cost function.

4. The pipeline-based tail-seat vertical takeoff and landing UAV control method according to claim 1, characterized in that, The generation of the second trajectory based on the symbolic distance field includes the following steps: Based on the fuselage dimensions of the vertical takeoff and landing UAV, the obstacle dimensions are expanded to generate environmental information; The initial geometric path from the current position to the termination position is obtained through a path search algorithm; Based on the environmental information and the initial geometric path, a symbolic distance field is constructed; The second trajectory is generated by performing trajectory planning in the symbolic distance field using the MINCO trajectory class.

5. The pipeline-based tail-seat vertical takeoff and landing UAV control method according to claim 1, characterized in that, The step of constructing the objective function based on the target trajectory, the contour constraint, and the adaptive parameters includes the following steps: Set the fourth arc length of the target trajectory to a preset trajectory; The current position of the vertical take-off and landing UAV is constrained by the contour constraint. The tracking error at the current position of the vertical take-off and landing UAV from the preset trajectory is obtained, and the longitudinal error and the lateral contour error are obtained. The flight speed of the vertical take-off and landing UAV is adjusted using the adaptive parameters. The objective function is obtained based on the longitudinal error, the lateral profile error, the flight speed of the vertical take-off and landing UAV, and the rate of change of the state input.

6. A pipeline-based tail-seat vertical take-off and landing unmanned aerial vehicle (UAV) control device, characterized in that, include: The first module is used to obtain the initial trajectory based on the spatial model, specifically for: Generate the first trajectory based on the convex polyhedron corridor; Alternatively, a second trajectory can be generated based on the symbolic distance field; The second module is used to obtain the target trajectory and the variable radius pipe based on the initial trajectory through cubic spline fitting, specifically for: The first arc length of the first trajectory is parameterized by a preset resolution to obtain a first discrete data point; cubic spline fitting is performed on the first discrete data point to obtain the target trajectory; the first position of the vertical take-off and landing UAV with a second arc length on the target trajectory is obtained; the first velocity direction of the first position is obtained; a first sectional surface perpendicular to the first velocity direction is obtained, and the first sectional surface intersects with the convex polyhedral corridor to form a convex polygon; the shortest distance from the first position to each side of the convex polygon is taken as the second radius; cubic spline fitting is performed on the second radius with respect to the second arc length to obtain the variable radius pipe; or, The third arc length of the second trajectory is parameterized by a preset resolution to obtain a second discrete data point; cubic spline fitting is performed on the second discrete data point to obtain the target trajectory; the third distance between the discrete second preset point on the target trajectory and the nearest obstacle is taken as the third radius; cubic spline fitting is performed on the third radius with respect to the third arc length of the target trajectory to obtain the variable radius pipe; The third module is used to set contour constraints and adaptive parameters based on the variable radius pipe, specifically for: Set the fourth arc length of the target trajectory to a preset trajectory; The derivative of the variable radius pipe is obtained by taking the derivative of the variable radius pipe. The contour constraint is obtained based on the preset trajectory, the variable radius pipe, and the derivative of the variable radius pipe; Obtain the maximum and minimum radius of the target trajectory; Obtain the maximum and minimum weights of the flight speed of the vertical take-off and landing UAV; The adaptive parameters are obtained based on the maximum radius, minimum radius, maximum weight, and minimum weight. The fourth module is used to construct an objective function based on the target trajectory, the contour constraints, and the adaptive parameters; The fifth module is used to preset the sideslip angle constraints of vertical take-off and landing UAVs; The sixth module is used to construct a pipeline-based model prediction contour control framework based on the objective function, the contour constraints, the adaptive parameters, and the sideslip angle constraints. The pipeline-based model prediction profile control framework is used for planning and controlling vertical take-off and landing (VTOL) drones.

Images