Dynamic motion planning method and system of land-air bimodal unmanned aerial vehicle, land-air bimodal unmanned aerial vehicle and storage medium

Abstract

The invention discloses a dynamic motion planning method and system for an air-ground bimodal unmanned aerial vehicle, the air-ground bimodal unmanned aerial vehicle and a storage medium, and the method comprises the steps: designing a disturbance observer through employing an uncertainty and disturbance estimation method based on a dynamic model of the air-ground bimodal unmanned aerial vehicle, a disturbance observer is used for estimating disturbance factors in the environment, and a disturbance estimation value is obtained; calculating and updating the dynamic boundary of the air-ground bimodal unmanned aerial vehicle in real time by using a motion planner according to the disturbance estimation value; and performing path search and trajectory optimization based on the updated dynamic boundary of the air-ground bimodal unmanned aerial vehicle to obtain a trajectory planning result. According to the invention, the disturbance observer is combined with the land-air dual-mode unmanned aerial vehicle motion planning technology, disturbance information in the environment is perceived at the planning level, and a disturbance adaptive safety boundary adjustment mechanism is provided, so that the planned trajectory meets the real-time dynamic constraint, and the stability of the unmanned aerial vehicle is improved. And the coping capacity of the land-air dual-mode unmanned aerial vehicle to environmental disturbance factors is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle motion planning, and in particular to a dynamic motion planning method and system for a land-air dual-mode unmanned aerial vehicle, a land-air dual-mode unmanned aerial vehicle, and a computer-readable storage medium. Background Art

[0002] UAV motion planning technology aims to enable UAVs to autonomously plan executable flight paths in complex environments to complete specific tasks through algorithms and strategies. Existing motion planning methods are mainly aimed at single-mode UAVs, and do not consider real-time disturbance factors in the environment. There is a lack of effective response measures for invisible disturbance factors in the environment (such as wind and friction), resulting in a high probability of failure of UAV motion planning in highly disturbed environments.

[0003] Although disturbance observers (DOBs) have appeared in existing technologies for estimating disturbance information and improving the tracking performance of controllers, disturbance observers are currently mostly only used in the control field, and their combination with the planning field has not been further explored. That is, disturbance observers have not yet been combined with UAV motion planning technology to consider disturbance information at the planning level, thereby enhancing the system's anti-interference ability.

[0004] Therefore, the prior art still needs to be improved and developed. Summary of the invention

[0005] The main purpose of the present invention is to provide a dynamic motion planning method and system for a land-air dual-mode UAV, a land-air dual-mode UAV and a computer-readable storage medium, aiming to solve the problem that the existing UAV motion planning methods are mainly aimed at single-mode UAVs and do not consider environmental disturbance factors, resulting in a high possibility of failure of UAV motion planning in a large disturbance environment.

[0006] To achieve the above-mentioned object of the invention, the present invention provides a dynamic motion planning method for a land-air dual-mode UAV, and the dynamic motion planning method for the land-air dual-mode UAV comprises: Based on the dynamic model of the dual-mode UAV in flight mode and land mode, a disturbance observer is designed by using uncertainty and disturbance estimation method, and the disturbance observer is used to estimate the disturbance factors in the environment to obtain a disturbance estimation value; Calculate and update the dynamic boundary of the land-air dual-mode UAV in real time using a motion planner according to the disturbance estimation value and the dynamic equations in the dynamic model; Based on the updated dynamic boundary of the land-air dual-mode UAV, the improved motion dynamics A* algorithm is used for path search, and the improved B-spline algorithm is used for trajectory optimization to obtain the trajectory planning result of the land-air dual-mode UAV.

[0007] Preferably, the dynamic model of the land-air dual-mode UAV in flight mode includes: ; ; ; ; ; ; ; ; ; ; ; ; ; in, Indicates the quality of the drone; Represents the position vector of the drone in the three-dimensional space in the world coordinate system, Respectively represent the UAV in the three-dimensional space under the world coordinate system The position of the axis; represents transpose; Represents three-dimensional space; Represents the acceleration vector of the drone in the three-dimensional space in the world coordinate system, Respectively represent the UAV in the three-dimensional space under the world coordinate system The acceleration of the axis; represents the total thrust produced by the four propellers; represents the gravitational acceleration vector, represents the acceleration due to gravity; represents the disturbance vector under the flight mode, Respectively represent the flight mode in the body coordinate system Axis disturbances; represents the attitude angle vector of the drone, They represent the roll angle, pitch angle and yaw angle of the drone respectively; represents the Euler angle vector, Represents the roll angle The cosine and sine of Respectively represent the pitch angle The cosine and sine of Represents the yaw angle The cosine and sine of ; represents the moment of inertia matrix, Indicates is a diagonal matrix with diagonal elements; Respectively represent the UAV in the body coordinate system Moment of inertia about the shaft; represents the rotational torque vector, Respectively represent the UAV in the body coordinate system Rotational torque on the shaft; Represents the attitude angular acceleration vector; represents the coupling vector, represents the attitude angular velocity vector, They represent the roll angular velocity, pitch angular velocity and yaw angular velocity of the UAV in the body coordinate system respectively; represents the unknown disturbance vector, Respectively expressed in the body coordinate system Unknown disturbances on the axis; The dynamic model of the land-air dual-mode UAV in the land mode includes: ; ; ; ; ; ; ; in, represents the driving force vector, Respectively represent the first wheel speed and the second wheel speed; represents the disturbance vector in the land mode, They represent the land mode in the body coordinate system The disturbance of the axis, They represent the longitudinal friction and lateral friction of the UAV in the body coordinate system respectively; represents the yaw angular acceleration; Represent the mass matrix, force coordinate transformation matrix and disturbance coordinate transformation matrix respectively: Indicates the width of the drone: represents the reaction torque, represents the longitudinal friction force on the first wheel, Indicates the lateral friction force on the first wheel: represents the longitudinal friction force on the second wheel, Indicates the lateral friction force on the second wheel: represents the longitudinal friction force on the third wheel, Indicates the lateral friction force on the third wheel: represents the longitudinal friction force on the fourth wheel, Indicates the lateral friction force on the fourth wheel: Respectively represent the distances from the rear wheel and front wheel to the center of mass of the drone.

[0008] Preferably, the dynamic model of the dual-mode UAV in flight mode and land mode is used to design a disturbance observer by using uncertainty and disturbance estimation method, and the disturbance observer is used to estimate the disturbance factors in the environment to obtain the disturbance estimation value, which specifically includes: Define a second-order system: ; in, denote the control output vector and disturbance vector respectively; According to the dynamic equations in the dynamic model of the land-air dual-mode UAV in the flight mode, the control output vector and the disturbance vector in the flight mode are expressed as: ; in, Respectively represent the control output vector and disturbance vector under the flight mode; According to the dynamic equations in the dynamic model of the land-air dual-mode UAV in the land mode, the control output vector and the disturbance vector in the land mode are expressed as: ; in, denote the control output vector and disturbance vector in land mode respectively; Represents the mass matrix The inverse matrix of The difference between the second-order system state vector and the control output vector is used as the disturbance estimate output by the disturbance observer. , where the second-order system state vector corresponds to the acceleration vector in the dynamics equation : ; According to the control output vector , reference control output vector and the disturbance estimate Relationship , we can get: ; Using a low-pass filter to filter the high-frequency noise in the second-order system, we can get: ; in, represents the inverse Laplace transform, is a low-pass filter, Represents the convolution operation; Combination and , we can get: ; make , the disturbance estimate The final expression is: ; in, represents the time constant, represents the Laplace variable, represents the integration time threshold of the disturbance observer, represents the time variable, Represents the first-order system state vector, corresponding to the velocity vector in the dynamic equation; express The corresponding first-order system state vector is, represents the initial first-order system state vector; express The corresponding reference control output vector.

[0009] Preferably, the method of calculating and updating the dynamic boundary of the land-air dual-mode UAV in real time using a motion planner according to the disturbance estimation value and the dynamic equation in the dynamic model specifically includes: Let the disturbance estimate in flight mode be Based on the dynamic model of the dual-mode UAV in flight mode, the dynamic boundary of the dual-mode UAV in flight mode is derived. It is expressed as: ; in, Respectively represent the flight mode in the world coordinate system Perturbation estimates on the axis; They represent the UAV’s flight mode in The minimum and maximum accelerations that the axis can achieve, Indicates the drone is in flight mode The maximum pulling force that the shaft can generate; They represent the UAV’s flight mode in The minimum and maximum accelerations that the axis can achieve, Indicates the drone is in flight mode The maximum pulling force that the shaft can generate; They represent the UAV’s flight mode in The minimum and maximum accelerations that the axis can achieve, Indicates the drone is in flight mode The maximum pulling force that the shaft can generate; Let the perturbation estimate in the land mode be Based on the dynamic model of the dual-mode UAV in the land mode, the dynamic boundary of the dual-mode UAV in the land mode is derived. It is expressed as: ; in, Respectively represent the land mode in the world coordinate system The perturbation estimate on the axis, They represent the UAV in land mode. The minimum and maximum accelerations that the axis can achieve, Indicates that the UAV is in land mode The maximum pulling force that the shaft can generate; They represent the UAV in land mode. The minimum and maximum accelerations that the axis can achieve, Indicates that the UAV is in land mode The maximum pulling force that the shaft can generate; according to and The relationship between the dynamic boundary of the dual-mode UAV under the flight mode is It is expressed as: ; according to and The dynamic boundary of the dual-mode UAV in the land mode is It is expressed as: ; According to the expression formula of the dynamic boundary of the land-air dual-mode UAV, the dynamic boundary of the land-air dual-mode UAV is calculated and updated in real time.

[0010] Preferably, based on the updated dynamic boundary of the land-air dual-mode UAV, the improved motion dynamics A* algorithm is used to perform path search, and the improved B-spline algorithm is used to perform trajectory optimization to obtain the trajectory planning result of the land-air dual-mode UAV, specifically including: Based on the updated dynamic boundary of the dual-mode land-air UAV, a trajectory cost function of the path search problem is designed to obtain an improved motion dynamics A* algorithm; Based on the navigation map, the improved motion dynamics A* algorithm is used to perform path search to obtain an initial path; Based on the updated dynamic boundary of the dual-mode land-air UAV, an optimization cost function of the trajectory optimization problem is designed to obtain an improved B-spline algorithm; Based on the initial path, the improved B-spline algorithm is used to perform trajectory optimization to obtain the trajectory planning result of the land-air dual-mode UAV.

[0011] Preferably, the trajectory cost function of the path search problem is designed based on the updated dynamic boundary of the land-air dual-mode UAV, specifically including: Based on the high energy consumption disadvantage of the dual-mode UAV in the flight mode, an additional energy cost is added to the flight trajectory, and the trajectory cost function is designed. It is expressed as: ; in, represents the acceleration input, represents the norm, represents the height of the motion primitive, represents the height threshold of motion primitives in flight mode, They represent the scaling factors used to adjust the height cost term and the time cost term, respectively. Indicates taking the maximum value; Based on the updated dynamic boundary of the dual-mode land-air UAV, the discrete acceleration input is obtained by discretizing the acceleration boundary at equal intervals. ; Based on the trajectory cost function, the discrete acceleration input All generated The total cost of motion primitives It is expressed as: ; in, represents the total number of motion primitives, Indicates the serial number, Respectively represent The acceleration and terminal height of each motion primitive, is the time over which the motion primitive is integrated forward.

[0012] Preferably, the optimization cost function of the trajectory optimization problem is designed based on the updated dynamic boundary of the land-air dual-mode UAV to obtain an improved B-spline algorithm, which specifically includes: Parameterize the initial trajectory as Control Points of Order B-spline curve, where Represents the first control points, Respectively represent Control points in world coordinate system The position of the axis, Represent the 0th control point, the 1st control point, and the control points, the optimized cost function of the design It is expressed as: ; in, represents the smoothness cost, represents the collision cost, represents the dynamic feasibility cost, Respectively The corresponding weights; Among them, the smoothness cost of the design It is expressed as: ; in, represents the curvature penalty term, represents the elastic uniform term, For the control points, For the control points, is the normalization parameter, represents the regularization weight; The collision cost of design It is expressed as: ; ; in, Indicates Control Points The distance to the nearest obstacle, represents the collision cost mapping function, Indicates the safety distance threshold; Design of dynamic feasibility cost based on the updated dynamic boundary of the dual-mode UAV , the design's dynamic feasibility cost It is expressed as: ; ; in, Indicates The acceleration control points along The acceleration value of the axis, represents the dynamic feasibility cost mapping function, They respectively represent the flight mode edges The lower and upper limits of the axis' acceleration, They represent the land mode along The lower and upper acceleration limits of the axis.

[0013] In order to achieve the above-mentioned invention object, the present invention also provides a dynamic motion planning system for a land-air dual-mode UAV, and the dynamic motion planning system for the land-air dual-mode UAV comprises: A disturbance estimation module is used to design a disturbance observer based on the dynamic model of the dual-mode UAV in the flight mode and the land mode by using uncertainty and disturbance estimation method, and to estimate the disturbance factors in the environment by using the disturbance observer to obtain a disturbance estimation value; A dynamic boundary adjustment module is used to calculate and update the dynamic boundary of the land-air dual-mode UAV in real time using a motion planner according to the disturbance estimation value and the dynamic equation in the dynamic model; Path search and trajectory optimization module: used to perform path search based on the updated dynamic boundary of the land-air dual-mode UAV using the improved motion dynamics A* algorithm, and to perform trajectory optimization using the improved B-spline algorithm, so as to obtain the trajectory planning result of the land-air dual-mode UAV.

[0014] To achieve the above-mentioned purpose of the invention, the present invention also provides a land-air dual-mode UAV, which includes: a disturbance observer and a motion planner, and the disturbance observer and the motion planner are used to jointly implement the steps of the dynamic motion planning method of the land-air dual-mode UAV as described above.

[0015] In order to achieve the above-mentioned purpose of the invention, the present invention also provides a computer-readable storage medium, which stores a dynamic motion planning program for a land-to-air dual-mode UAV. When the dynamic motion planning program for the land-to-air dual-mode UAV is executed by a disturbance observer and a motion planner, the steps of the dynamic motion planning method for the land-to-air dual-mode UAV as described above are implemented.

[0016] In the present invention, based on the dynamic model of the dual-mode land-air UAV in flight mode and land mode, an uncertainty and disturbance estimation method is adopted to design a disturbance observer, and the disturbance observer is used to estimate the disturbance factors in the environment to obtain a disturbance estimation value; according to the disturbance estimation value and the dynamic equations in the dynamic model, a motion planner is used to calculate and update the dynamic boundary of the dual-mode land-air UAV in real time; based on the updated dynamic boundary of the dual-mode land-air UAV, an improved motion dynamics A* algorithm is used to perform path search, and an improved B-spline algorithm is used to perform trajectory optimization, so as to obtain a trajectory planning result of the dual-mode land-air UAV. The present invention designs a disturbance observer in combination with the dynamic model of the land-air dual-mode UAV at the control level, and uploads the estimated disturbance information to the motion planner; at the planning level, a disturbance adaptive safety boundary adjustment mechanism is proposed, which uses the disturbance information to adjust the dynamic boundary of the land-air dual-mode UAV in real time, and then performs path search and trajectory optimization based on the adjusted dynamic boundary of the land-air dual-mode UAV, so as to perceive the disturbance information in the environment at the planning level, plan a trajectory that meets the real-time dynamic constraints of the UAV, improve the coping ability of the land-air dual-mode UAV to environmental disturbance factors, and significantly enhance the adaptability of the land-air dual-mode UAV in a large disturbance environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flow chart of a preferred embodiment of the method for dynamic motion planning of a land-air dual-mode UAV of the present invention; Figure 2 It is another flow chart of a preferred embodiment of the method for dynamic motion planning of a land-air dual-mode UAV of the present invention; Figure 3 It is a structural diagram of a preferred embodiment of the dynamic motion planning system of the land-air dual-mode UAV of the present invention; Figure 4 It is a structural diagram of a preferred embodiment of the land-air dual-mode UAV of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention clearer and more specific, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0019] UAV motion planning technology aims to enable UAVs to autonomously plan executable flight paths in complex environments to complete specific tasks through algorithms and strategies. Existing motion planning methods are mainly aimed at single-mode UAVs and do not consider the real-time disturbance factors in the environment, which may lead to two problems: first, the planner may generate conservative trajectories. For example, in the case of tailwind, the UAV can achieve greater acceleration, but the planner with fixed dynamic boundaries cannot fully utilize the wind; second, the planner may generate unexecutable trajectories. For example, in the case of headwind, the actual maximum acceleration of the UAV will be weakened, and the planner fails to adjust this, resulting in the generated trajectory exceeding the dynamic boundary of the UAV, which in turn causes the controller to produce a large tracking error. In general, the existing UAV motion planning technology only uses visual cameras or lidars to perceive environmental obstacle information at the planning level, without considering the real-time disturbance factors in the environment. There is a lack of effective countermeasures for invisible disturbance factors in the environment (such as wind and friction), resulting in a high probability of failure of UAV motion planning in a large disturbance environment.

[0020] Although disturbance observers (DOBs) have appeared in existing technologies for estimating disturbance information and improving the tracking performance of controllers, disturbance observers are currently mostly only used in the control field, and their combination with the planning field has not been further explored. That is, disturbance observers have not yet been combined with UAV motion planning technology to consider disturbance information at the planning level, thereby enhancing the system's anti-interference ability.

[0021] In order to solve the above technical problems, the present invention provides a dynamic motion planning method for a land-air dual-mode UAV. Based on the dynamic model of the land-air dual-mode UAV in flight mode and land mode, an uncertainty and disturbance estimation method is used to design a disturbance observer, and the disturbance observer is used to estimate the disturbance factors in the environment to obtain a disturbance estimation value; according to the disturbance estimation value and the dynamic equations in the dynamic model, a motion planner is used to calculate and update the dynamic boundary of the land-air dual-mode UAV in real time; based on the updated dynamic boundary of the land-air dual-mode UAV, an improved motion dynamics A* algorithm is used to perform path search, and an improved B-spline algorithm is used to perform trajectory optimization to obtain a trajectory planning result of the land-air dual-mode UAV. The present invention designs a disturbance observer in combination with the dynamic model of the land-air dual-mode UAV at the control level, and uploads the estimated disturbance information to the motion planner; at the planning level, a disturbance adaptive safety boundary adjustment mechanism is proposed, which uses the disturbance information to adjust the dynamic boundary of the land-air dual-mode UAV in real time, and then performs path search and trajectory optimization based on the adjusted dynamic boundary of the land-air dual-mode UAV, so as to perceive the disturbance information in the environment at the planning level, plan a trajectory that meets the real-time dynamic constraints of the UAV, improve the coping ability of the land-air dual-mode UAV to environmental disturbance factors, and significantly enhance the adaptability of the land-air dual-mode UAV in a large disturbance environment.

[0022] The application content is further explained below through the description of embodiments in conjunction with the accompanying drawings.

[0023] A preferred embodiment of the dynamic motion planning method of the land-air dual-mode UAV of the present invention is as follows: Figure 1 and Figure 2 As shown, specifically including: S1. Based on the dynamic model of the dual-mode UAV in flight mode and land mode, a disturbance observer is designed by using uncertainty and disturbance estimation method, and the disturbance observer is used to estimate the disturbance factors in the environment to obtain a disturbance estimation value.

[0024] In one implementation of this embodiment, the dynamic model of the land-air dual-mode UAV in the flight mode includes: ; ; ; ; ; ; ; ; ; ; ; ; ; in, Indicates the quality of the drone; Represents the position vector of the drone in the three-dimensional space in the world coordinate system, Respectively represent the UAV in the three-dimensional space under the world coordinate system The position of the axis; represents transpose; Represents three-dimensional space; Represents the acceleration vector of the drone in the three-dimensional space in the world coordinate system, Respectively represent the UAV in the three-dimensional space under the world coordinate system The acceleration of the axis; represents the total thrust produced by the four propellers; represents the gravitational acceleration vector, represents the acceleration due to gravity; represents the disturbance vector under the flight mode, Respectively represent the flight mode in the body coordinate system Axis disturbances; represents the attitude angle vector of the drone, They represent the roll angle, pitch angle and yaw angle of the drone respectively; represents the Euler angle vector, Represents the roll angle The cosine and sine of Respectively represent the pitch angle The cosine and sine of Represents the yaw angle The cosine and sine of ; represents the moment of inertia matrix, Indicates is a diagonal matrix with diagonal elements and the rest of the elements are zero; Respectively represent the UAV in the body coordinate system Moment of inertia about the shaft; represents the rotational torque vector, Respectively represent the UAV in the body coordinate system Rotational torque on the shaft; Represents the attitude angular acceleration vector; represents the coupling vector, represents the attitude angular velocity vector, They represent the roll angular velocity, pitch angular velocity and yaw angular velocity of the UAV in the body coordinate system respectively; represents the unknown disturbance vector, Respectively expressed in the body coordinate system Unknown disturbances on the axis; The dynamic model of the land-air dual-mode UAV in the land mode includes: ; ; ; ; ; ; ; in, represents the position vector in the world coordinate system in the land mode (the land mode height is 0), where Respectively represent the UAV in the world coordinate system The position of the axis; represents the driving force vector, Represents the first wheel speed and the second wheel speed respectively. The wheels on the same side have the same speed. ; represents the disturbance vector in the land mode, They represent the land mode in the body coordinate system The disturbance of the axis, They represent the longitudinal friction and lateral friction of the drone in the body coordinate system. It can be understood that Respectively represent the UAV in the body coordinate system The disturbance force on the axis can also be expressed as That is, the drone is in the body coordinate system It is expressed by the longitudinal friction and lateral friction on the shaft; represents the yaw angular acceleration; Represent the mass matrix, force coordinate transformation matrix and disturbance coordinate transformation matrix respectively: Indicates the width of the drone: represents the reaction torque, represents the longitudinal friction force on the first wheel, Indicates the lateral friction force on the first wheel: represents the longitudinal friction force on the second wheel, Indicates the lateral friction force on the second wheel: represents the longitudinal friction force on the third wheel, Indicates the lateral friction force on the third wheel: represents the longitudinal friction force on the fourth wheel, Indicates the lateral friction force on the fourth wheel: Respectively represent the distances from the rear wheel and front wheel to the center of mass of the drone.

[0025] In an implementation of this embodiment, the dynamic model of the dual-mode UAV in flight mode and land mode is used to design a disturbance observer using uncertainty and disturbance estimation method, and the disturbance observer is used to estimate the disturbance factors in the environment to obtain the disturbance estimation value, which specifically includes: The difference between the actual output and the expected output of the system is mainly caused by disturbances. The core idea of ​​the disturbance observer is to reversely calculate the disturbance estimate by comparing the actual output with the expected output and combining the system dynamics model: Define a second-order system: ; in, represents the state vector of the second-order system, which corresponds to the acceleration vector of the dynamic equation of the land-air dual-mode UAV in the land-air dual-mode UAV dynamic model, and therefore can be represented by the same symbol; denote the control output vector and disturbance vector respectively; According to the dynamic equations in the dynamic model of the land-air dual-mode UAV in the flight mode , the control output vector and disturbance vector in the flight mode are expressed as: ; in, Respectively represent the control output vector and disturbance vector under the flight mode; According to the dynamic equations in the dynamic model of the land-air dual-mode UAV in the land mode , the control output vector and disturbance vector in land mode are expressed as: ; in, denote the control output vector and disturbance vector in land mode respectively; Represents the mass matrix The inverse matrix of formula describes a control system in reality. , that is, giving the system a control quantity , which is ultimately reflected in the system status However, in the real environment, there are disturbance factors such as wind and friction. , the final control quantity and disturbance factor are reflected on the system state together. The basic idea of ​​the disturbance estimator is to calculate the difference between the actual system state and the control quantity given to the system. The difference is the estimated disturbance value, that is, ; Therefore, the difference between the second-order system state vector (i.e., the actual system state) and the control output vector (i.e., the control quantity given to the system) is used as the disturbance estimate output by the disturbance observer. , where the second-order system state vector corresponds to the acceleration vector in the dynamics equation : ; According to the control output vector , reference control output vector and the disturbance estimate Relationship (Let the reference control the output Subtract the estimated disturbance , to realize the system compensation for disturbance), we can get: ; Using a low-pass filter to filter the high-frequency noise in the second-order system, we can get: ; in, represents the inverse Laplace transform, is a low-pass filter, Represents the convolution operation; because the disturbance information is usually a low-frequency signal, the above formula adds a low-pass filter to the disturbance of the system to filter out useless high-frequency noise in the system; Combination and , we can get: ; make , the disturbance estimate The final expression is: ; in, represents the time constant, represents the Laplace variable, represents the integration time threshold of the disturbance observer, represents the time variable, Represents the first-order system state vector, corresponding to the velocity vector in the dynamic equation; express The corresponding first-order system state vector is, represents the initial first-order system state vector; express The corresponding reference control output vector.

[0026] Based on the dynamic model of the dual-mode UAV of land and air, the present invention adopts the uncertainty and disturbance estimation method (UDE, Uncertainty and Disturbance Estimator) to design disturbance observers for the flight mode and land mode of the dual-mode UAV of land and air respectively, and the disturbance observer estimates the disturbance factors in the environment, wherein the uncertainty and disturbance estimation method first calculates the difference between the actual system state and the control quantity given to the system, then adds a low-pass filter to the value, and finally obtains the disturbance estimation value. In summary, based on the dynamic model of the dual-mode UAV of land and air, the present invention designs a disturbance observer for the flight and land dual modes, and the disturbance observer calculates the deviation signal between the actual system state and the applied control quantity in real time, performs low-pass filtering on it, thereby obtaining a dynamic estimation of the external disturbance. In fact, this invention designs a disturbance observer for a land-air dual-modal UAV based on the uncertainty and disturbance estimation theory for the first time, and combines the disturbance information output by the disturbance observer with the motion planner for the first time to improve the planning effect of the motion planner, thereby realizing the perception of disturbance information in the environment at the planning level (previously, most of the disturbance information was only used in the control field, and the combination with the planning field was not further explored).

[0027] S2. According to the disturbance estimation value and the dynamic equations in the dynamic model, a motion planner is used to calculate and update the dynamic boundary of the land-air dual-mode UAV in real time.

[0028] In an implementation of this embodiment, the method of calculating and updating the dynamic boundary of the land-air dual-mode UAV in real time using a motion planner according to the disturbance estimation value and the dynamic equation in the dynamic model specifically includes: Let the disturbance estimate in flight mode be Based on the dynamic model of the dual-mode UAV in flight mode, the dynamic boundary of the dual-mode UAV in flight mode is derived. It is expressed as: ; in, Respectively represent the flight mode in the world coordinate system Perturbation estimates on the axis; They represent the UAV’s flight mode in The minimum and maximum accelerations that the axis can achieve, Indicates the drone is in flight mode The maximum pulling force that the shaft can generate; They represent the UAV’s flight mode in The minimum and maximum accelerations that the axis can achieve, Indicates the drone is in flight mode The maximum pulling force that the shaft can generate; They represent the UAV’s flight mode in The minimum and maximum accelerations that the axis can achieve, Indicates the drone is in flight mode The maximum pulling force that the shaft can generate; Let the perturbation estimate in the land mode be Based on the dynamic model of the dual-mode UAV in the land mode, the dynamic boundary of the dual-mode UAV in the land mode is derived. It is expressed as: ; in, Respectively represent the land mode in the world coordinate system The perturbation estimate on the axis, They represent the UAV in land mode. The minimum and maximum accelerations that the axis can achieve, Indicates that the UAV is in land mode The maximum pulling force that the shaft can generate; They represent the UAV in land mode. The minimum and maximum accelerations that the axis can achieve, Indicates that the UAV is in land mode The maximum pulling force that the shaft can generate; according to and The relationship between ), the dynamic boundary of the dual-mode UAV under the flight mode It is expressed as: ; according to and The relationship between ), the dynamic boundary of the dual-mode UAV under the land mode is It is expressed as: ; According to the expression formula of the dynamic boundary of the land-air dual-mode UAV, the dynamic boundary of the land-air dual-mode UAV is calculated and updated in real time.

[0029] The present invention uploads the disturbance information estimated by the disturbance observer to the motion planner, and proposes a disturbance adaptive safety boundary adjustment mechanism, which can dynamically adjust the dynamic boundary of the land-air dual-mode UAV in real time and provide accurate dynamic boundary information for path search and trajectory optimization.

[0030] S3. Based on the updated dynamic boundary of the land-air dual-mode UAV, the improved motion dynamics A* algorithm is first used to perform path search, and then the improved B-spline algorithm is used to perform trajectory optimization to obtain the trajectory planning result of the land-air dual-mode UAV.

[0031] In an implementation of this embodiment, based on the updated dynamic boundary of the land-air dual-mode UAV, the improved motion dynamics A* algorithm is used to perform path search, and the improved B-spline algorithm is used to perform trajectory optimization to obtain the trajectory planning result of the land-air dual-mode UAV, specifically including: Based on the updated dynamic boundary of the dual-mode land-air UAV, a trajectory cost function of the path search problem is designed to obtain an improved motion dynamics A* algorithm; Based on the navigation map, the improved motion dynamics A* algorithm is used to perform path search to obtain an initial path; Based on the updated dynamic boundary of the dual-mode land-air UAV, an optimization cost function of the trajectory optimization problem is designed to obtain an improved B-spline algorithm; Based on the initial path, the improved B-spline algorithm is used to perform trajectory optimization to obtain the trajectory planning result of the land-air dual-mode UAV.

[0032] Specifically, the present invention adopts a hierarchical planning method, which uses the Kinodynamic A* algorithm (motion dynamics A* algorithm) at the front end to perform path search and the B-spline algorithm at the back end to perform trajectory optimization. The specific process of using the Kinodynamic A* algorithm to perform path search includes: Step 1: State space modeling: define high-dimensional state nodes including position, velocity, and acceleration, and constrain their dynamic parameter ranges (such as velocity / acceleration upper and lower limits); Step 2: State expansion and pruning: Generate motion primitives through forward integration of discrete acceleration (the acceleration refers to the real-time acceleration boundary calculated above, i.e., the dynamic boundary, and the discretization refers to discretizing the acceleration into n acceleration values ​​of different sizes at equal intervals), and eliminate invalid motion primitives through collision detection and environmental constraints (such as obstacles); Step 3: Design the trajectory cost function, calculate and select the lowest cost motion primitive as the new extended trajectory: After removing the motion primitives that collide with obstacles, it is necessary to calculate the cost of each remaining motion primitive and select the motion primitive with the lowest cost as the new extended trajectory; Step 4: Check whether a trajectory to the end point is found: Check whether the new extended trajectory reaches the end point. If so, use the new extended trajectory as the initial trajectory; if not, use the end point of the new extended trajectory as the new extended node and jump to step 2 to continue the search process.

[0033] The Kinodynamic A* algorithm does not take into account the distance information of obstacles during the search process, which causes the searched trajectory to often be close to obstacles. In addition, the smoothness and dynamic feasibility of each segment of the trajectory need to be improved. Therefore, the B-spline algorithm is also needed to optimize the initial trajectory. The specific process of using the B-spline algorithm for trajectory optimization is as follows: Step 1: First parameterize the initial trajectory with control points Order B-spline curve; Step 2: Design the optimization cost function of the trajectory optimization problem; Step 3: Use numerical methods (such as gradient descent) to iteratively solve the optimization problem, adjust the trajectory position, and finally solve a collision-free, smooth, and dynamically feasible trajectory.

[0034] The present invention designs a Kinodynamic A* path search algorithm, in which the aerial trajectory introduces additional energy cost and tends to plan the ground trajectory; at the same time, a B-spline trajectory optimization algorithm is designed, which optimizes the planned trajectory through the dynamically changing dynamic boundary of the land-air dual-mode UAV to ensure that the trajectory is still feasible when facing disturbances in the real environment.

[0035] In an implementation of this embodiment, the design of a trajectory cost function for the path search problem based on the updated dynamic boundary of the land-air dual-mode UAV specifically includes: Considering the high energy consumption disadvantage of the dual-mode UAV flight mode, by adding additional energy cost to the flight trajectory, the dual-mode UAV tends to plan the ground trajectory and only switches to the flight mode when encountering insurmountable obstacles. The designed trajectory cost function It is expressed as: ; in, represents the acceleration input, represents the norm, represents the height of the motion primitive, represents the height threshold of motion primitives in flight mode, They represent the scaling factors used to adjust the height cost term and the time cost term, respectively. Indicates taking the maximum value; the motion primitive is the basic trajectory fragment generated by discrete control quantities (acceleration), which can be combined for UAV motion planning.

[0036] Based on the updated dynamic boundary of the dual-mode land-air UAV, the discrete acceleration input is obtained by discretizing the acceleration boundary at equal intervals. ; Based on the trajectory cost function, the discrete acceleration input All generated The total cost of motion primitives It is expressed as: ; in, represents the total number of motion primitives, Indicates the serial number, Respectively represent The acceleration and terminal height of each motion primitive, is the time over which the motion primitive is integrated forward.

[0037] The trajectory cost function designed by the present invention is shown in the above formula, which can calculate the cost of all motion primitives from the initial position to the current position. The formula weighs the control amount of the trajectory (acceleration, tending to reduce acceleration to reduce energy consumption), height (tending to search for ground trajectories to reduce system energy consumption) and total trajectory time (tending to shorten system operating time and improve system efficiency), so that the planner searches for a collision-free, low-energy, and high-efficiency trajectory.

[0038] In an implementation of this embodiment, the optimization cost function of the trajectory optimization problem is designed based on the updated dynamic boundary of the land-air dual-mode UAV to obtain an improved B-spline algorithm, which specifically includes: Parameterize the initial trajectory as Control Points of Order B-spline curve, where Represents the first Control points (position control points), Respectively represent Control points in world coordinate system The position of the axis, Represent the 0th control point, the 1st control point, and the control points, the optimized cost function of the design It is expressed as: ; in, represents the smoothness cost, which aims to distribute the control points on a straight line as evenly as possible; represents the collision cost, which pushes the control point away from the obstacle based on the distance information; represents the dynamic feasibility cost, constraining the acceleration of the trajectory to ensure that the dual-mode UAV can execute the trajectory; Respectively The corresponding weights; Among them, the smoothness cost of the design It is expressed as: ; in, represents the curvature penalty term, Represents the elastic uniform term. The smoothness cost is divided into two parts: the curvature penalty term and the elastic uniform term. The curvature penalty term approximates the local curvature of the trajectory through the second-order difference and penalizes the curvature of the trajectory. Its purpose is to suppress sharp turns of the trajectory and improve kinematic feasibility (such as avoiding excessive acceleration or centripetal force). The elastic uniform term encourages uniform distribution of control point spacing to avoid local over-crowding or over-sparseness. For the control points, For the control points; It is a normalization parameter used to eliminate the influence of non-uniform distribution of control points, and its value is the average distance between adjacent control points; represents the regularization weight, which balances the curvature penalty and distribution uniformity; The collision cost of design It is expressed as: ; ; in, Indicates Control Points The distance to the nearest obstacle, represents the collision cost mapping function, Indicates the safety distance threshold, which must be greater than the radius or physical size of the drone. Through the function Mapped to collision cost, the collision costs of all control points are accumulated to form the total collision cost, which is used to quantify the obstacle avoidance constraints in trajectory optimization; when the distance between the control point and the obstacle is When , an exponential penalty is imposed, and the closer the distance, the stronger the penalty; when When , there is no risk of collision and the cost is zero.

[0039] Design of dynamic feasibility cost based on the updated dynamic boundary of the dual-mode UAV , the design's dynamic feasibility cost It is expressed as: ; ; in, Indicates Acceleration control points along The acceleration value of the axis, represents the dynamic feasibility cost mapping function, They respectively represent the flight mode edges The lower and upper limits of the axis' acceleration, They represent the land mode along The design of the dynamic feasibility cost is based on the disturbance adaptive safety boundary adjustment mechanism. The dynamic boundary of the land-air dual-mode UAV system, that is, the acceleration boundary, that is, the lower and upper limits of acceleration, can be calculated in real time by the above formula; Among them, the acceleration control point Calculation method: ; in, is the time between two position control points, For the Speed ​​control points, For the speed control point.

[0040] In another implementation of this embodiment, Figure 2 As shown, a dynamic motion planning method for a land-air dual-mode UAV includes: first determining whether a new instruction is issued by a user, that is, whether a new target position is provided; if so, performing disturbance estimation; then dynamically adjusting the dynamic boundary based on the disturbance estimation value; then performing path search and trajectory optimization based on the dynamically adjusted dynamic boundary to obtain a trajectory result; a controller controls the UAV to fly according to the trajectory result; and then determining whether the target position is reached; if not, returning to continue disturbance estimation; if so, determining whether a new target position is provided; and directly terminating the process if no new target position is provided.

[0041] In summary, the controller (i.e., disturbance observer) of the present invention estimates the disturbance information of the dual-mode UAV in real time and uploads it to the motion planner. The motion planner dynamically adjusts the dynamic boundary of the UAV based on the disturbance information, and finally performs path search and trajectory optimization through the dynamically changing dynamic boundary to ensure that the final trajectory meets the real-time dynamic feasibility requirements.

[0042] In addition, based on the above-mentioned dynamic motion planning method of the land-air dual-mode UAV, the present invention also provides a dynamic motion planning system of the land-air dual-mode UAV, wherein a preferred embodiment of the dynamic motion planning system of the land-air dual-mode UAV is as follows: Figure 3 As shown, specifically including: Disturbance estimation module 01: used to design a disturbance observer based on the dynamic model of the dual-mode UAV in flight mode and land mode by using uncertainty and disturbance estimation method, and use the disturbance observer to estimate the disturbance factors in the environment to obtain a disturbance estimation value; Dynamic boundary adjustment module 02: used for calculating and updating the dynamic boundary of the land-air dual-mode UAV in real time using a motion planner according to the disturbance estimation value and the dynamic equation in the dynamic model; Path search and trajectory optimization module 03: used to perform path search based on the updated dynamic boundary of the land-air dual-mode UAV using the improved motion dynamics A* algorithm, and to perform trajectory optimization using the improved B-spline algorithm, so as to obtain the trajectory planning result of the land-air dual-mode UAV.

[0043] In addition, based on the above-mentioned land-air dual-mode UAV dynamic motion planning method and system, the present invention also provides a land-air dual-mode UAV, wherein a preferred embodiment of the land-air dual-mode UAV is as follows: Figure 4 As shown, it specifically includes a disturbance observer 10 and a motion planner 20. Figure 4 Only some components of the land-air dual-mode UAV are shown, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. In one embodiment, the disturbance observer 10 and the motion planner 20 jointly implement the steps of the dynamic motion planning method of the land-air dual-mode UAV as described above. Specifically, the disturbance observer 10 is used to estimate the disturbance factors in the environment, obtain the disturbance estimation value and upload the disturbance estimation value to the motion planner 20. The motion planner 20 is used to calculate and update the dynamic boundary of the land-air dual-mode UAV in real time based on the disturbance estimation value, perform path search and trajectory optimization based on the updated dynamic boundary of the land-air dual-mode UAV, and obtain the trajectory planning result of the land-air dual-mode UAV.

[0044] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a dynamic motion planning program for a land-to-air dual-mode UAV, and when the dynamic motion planning program for the land-to-air dual-mode UAV is executed by a disturbance observer and a motion planner, the steps of the dynamic motion planning method for the land-to-air dual-mode UAV as described above are implemented.

[0045] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the existence of other identical elements in the process, method, article or terminal including the element.

[0046] Of course, those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing related hardware (such as a processor, a controller, etc.) through a computer program, and the program can be stored in a computer-readable storage medium that can be read by a computer, and the program can include the processes of the above-mentioned method embodiments when executed. The computer-readable storage medium can be a memory, a disk, an optical disk, etc.

[0047] It should be understood that the application of the present invention is not limited to the above examples. For ordinary technicians in this field, improvements or changes can be made based on the above description. All these improvements and changes should fall within the scope of protection of the claims attached to the present invention.

Claims

1. A dynamic motion planning method for a land-air dual-mode UAV, characterized in that: The dynamic motion planning method of the land-air dual-mode UAV includes: Based on the dynamic model of the dual-mode UAV in flight mode and land mode, a disturbance observer is designed by using uncertainty and disturbance estimation method, and the disturbance observer is used to estimate the disturbance factors in the environment to obtain a disturbance estimation value; Calculate and update the dynamic boundary of the land-air dual-mode UAV in real time using a motion planner according to the disturbance estimation value and the dynamic equation in the dynamic model; Based on the updated dynamic boundary of the land-air dual-mode UAV, the improved motion dynamics A* algorithm is used for path search, and the improved B-spline algorithm is used for trajectory optimization to obtain the trajectory planning result of the land-air dual-mode UAV.

2. The method for dynamic motion planning of a land-air dual-mode UAV according to claim 1, characterized in that: The dynamic model of the land-air dual-mode UAV in the flight mode includes: ； ； ； ； ； ； ； ； ； ； ； ； ； in, Indicates the quality of the drone; Represents the position vector of the drone in the three-dimensional space in the world coordinate system, Respectively represent the UAV in the three-dimensional space under the world coordinate system The position of the axis; represents transpose; Represents three-dimensional space; Represents the acceleration vector of the drone in the three-dimensional space in the world coordinate system, Respectively represent the UAV in the three-dimensional space under the world coordinate system The acceleration of the axis; represents the total thrust produced by the four propellers; represents the gravitational acceleration vector, represents the acceleration due to gravity; represents the disturbance vector under the flight mode, Respectively represent the flight mode in the body coordinate system Axis disturbances; represents the attitude angle vector of the drone, They represent the roll angle, pitch angle and yaw angle of the drone respectively; represents the Euler angle vector, Represents the roll angle The cosine and sine of Respectively represent the pitch angle The cosine and sine of Represents the yaw angle The cosine and sine of ; represents the moment of inertia matrix, Indicates is a diagonal matrix with diagonal elements; Respectively represent the UAV in the body coordinate system Moment of inertia about the shaft; represents the rotational torque vector, Respectively represent the UAV in the body coordinate system Rotational torque on the shaft; Represents the attitude angular acceleration vector; represents the coupling vector, represents the attitude angular velocity vector, They represent the roll angular velocity, pitch angular velocity and yaw angular velocity of the UAV in the body coordinate system respectively; represents the unknown disturbance vector, Respectively expressed in the body coordinate system Unknown disturbances on the axis; The dynamic model of the land-air dual-mode UAV in the land mode includes: ； ； ； ； ； ； ； in, represents the driving force vector, Respectively represent the first wheel speed and the second wheel speed; represents the disturbance vector in the land mode, They represent the land mode in the body coordinate system The disturbance of the axis, They represent the longitudinal friction and lateral friction of the UAV in the body coordinate system respectively; represents the yaw angular acceleration; Represent the mass matrix, force coordinate transformation matrix and disturbance coordinate transformation matrix respectively: Indicates the width of the drone: represents the reaction torque, represents the longitudinal friction force on the first wheel, Indicates the lateral friction force on the first wheel: represents the longitudinal friction force on the second wheel, Indicates the lateral friction force on the second wheel: represents the longitudinal friction force on the third wheel, Indicates the lateral friction force on the third wheel: represents the longitudinal friction force on the fourth wheel, Indicates the lateral friction force on the fourth wheel: Respectively represent the distances from the rear wheel and front wheel to the center of mass of the drone.

3. The method for dynamic motion planning of a land-air dual-mode UAV according to claim 2, characterized in that: The method of designing a disturbance observer based on the dynamic model of the dual-mode UAV in flight mode and land mode by using uncertainty and disturbance estimation method, and using the disturbance observer to estimate the disturbance factors in the environment to obtain the disturbance estimation value specifically includes: Define a second-order system: ； in, denote the control output vector and disturbance vector respectively; According to the dynamic equations in the dynamic model of the land-air dual-mode UAV in the flight mode, the control output vector and the disturbance vector in the flight mode are expressed as: ； in, Respectively represent the control output vector and disturbance vector under the flight mode; According to the dynamic equations in the dynamic model of the land-air dual-mode UAV in the land mode, the control output vector and the disturbance vector in the land mode are expressed as: ； in, denote the control output vector and disturbance vector in land mode respectively; Represents the mass matrix The inverse matrix of The difference between the second-order system state vector and the control output vector is used as the disturbance estimate output by the disturbance observer. , where the second-order system state vector corresponds to the acceleration vector in the dynamics equation : ； According to the control output vector , reference control output vector and the disturbance estimate Relationship , we can get: ； Using a low-pass filter to filter the high-frequency noise in the second-order system, we can get: ； in, represents the inverse Laplace transform, is a low-pass filter, Represents the convolution operation; Combination and , we can get: ； make , the disturbance estimate The final expression is: ； in, represents the time constant, represents the Laplace variable, represents the integration time threshold of the disturbance observer, represents the time variable, Represents the first-order system state vector, corresponding to the velocity vector in the dynamic equation; express The corresponding first-order system state vector is, represents the initial first-order system state vector; express The corresponding reference control output vector.

4. The method for dynamic motion planning of a land-air dual-mode UAV according to claim 3 is characterized in that: The method of calculating and updating the dynamic boundary of the land-air dual-mode UAV in real time using a motion planner according to the disturbance estimation value and the dynamic equation in the dynamic model specifically includes: Let the disturbance estimate in flight mode be Based on the dynamic model of the dual-mode UAV in flight mode, the dynamic boundary of the dual-mode UAV in flight mode is derived. It is expressed as: ； in, Respectively represent the flight mode in the world coordinate system Perturbation estimates on the axis; They represent the UAV’s flight mode in The minimum and maximum accelerations that the axis can achieve, Indicates the drone is in flight mode The maximum pulling force that the shaft can generate; They represent the UAV’s flight mode in The minimum and maximum accelerations that the axis can achieve, Indicates the drone is in flight mode The maximum pulling force that the shaft can generate; They represent the UAV’s flight mode in The minimum and maximum accelerations that the axis can achieve, Indicates the drone is in flight mode The maximum pulling force that the shaft can generate; Let the perturbation estimate in the land mode be Based on the dynamic model of the dual-mode UAV in the land mode, the dynamic boundary of the dual-mode UAV in the land mode is derived. It is expressed as: ； in, Respectively represent the land mode in the world coordinate system The perturbation estimate on the axis, They represent the UAV in land mode. The minimum and maximum accelerations that the axis can achieve, Indicates that the UAV is in land mode The maximum pulling force that the shaft can generate; They represent the UAV in land mode. The minimum and maximum accelerations that the axis can achieve, Indicates that the UAV is in land mode The maximum pulling force that the shaft can generate; according to and The relationship between the dynamic boundary of the dual-mode UAV under the flight mode is It is expressed as: ； according to and The dynamic boundary of the dual-mode UAV in the land mode is It is expressed as: ； According to the expression formula of the dynamic boundary of the land-air dual-mode UAV, the dynamic boundary of the land-air dual-mode UAV is calculated and updated in real time.

5. The method for dynamic motion planning of a land-air dual-mode UAV according to claim 4, characterized in that: Based on the updated dynamic boundary of the land-air dual-mode UAV, the improved motion dynamics A* algorithm is used to search for paths, and the improved B-spline algorithm is used to optimize trajectories to obtain the trajectory planning result of the land-air dual-mode UAV, which specifically includes: Based on the updated dynamic boundary of the dual-mode land-air UAV, a trajectory cost function of the path search problem is designed to obtain an improved motion dynamics A* algorithm; Based on the navigation map, the improved motion dynamics A* algorithm is used to perform path search to obtain an initial path; Based on the updated dynamic boundary of the dual-mode land-air UAV, an optimization cost function of the trajectory optimization problem is designed to obtain an improved B-spline algorithm; Based on the initial path, the improved B-spline algorithm is used to perform trajectory optimization to obtain the trajectory planning result of the land-air dual-mode UAV.

6. The method for dynamic motion planning of a land-air dual-mode UAV according to claim 5, characterized in that: The design of the trajectory cost function of the path search problem based on the updated dynamic boundary of the land-air dual-mode UAV specifically includes: Based on the high energy consumption disadvantage of the dual-mode UAV in the flight mode, an additional energy cost is added to the flight trajectory, and the trajectory cost function is designed. It is expressed as: ； in, represents the acceleration input, represents the norm, represents the height of the motion primitive, represents the height threshold of motion primitives in flight mode, They represent the scaling factors used to adjust the height cost term and the time cost term, respectively. Indicates taking the maximum value; Based on the updated dynamic boundary of the dual-mode land-air UAV, the discrete acceleration input is obtained by discretizing the acceleration boundary at equal intervals. ; Based on the trajectory cost function, the discrete acceleration input All generated The total cost of motion primitives It is expressed as: ； in, represents the total number of motion primitives, Indicates the serial number, Respectively represent The acceleration and terminal height of each motion primitive, is the time over which the motion primitive is integrated forward.

7. The method for dynamic motion planning of a land-air dual-mode UAV according to claim 6, characterized in that: The optimization cost function of the trajectory optimization problem is designed based on the updated dynamic boundary of the land-air dual-mode UAV to obtain an improved B-spline algorithm, which specifically includes: Parameterize the initial trajectory as Control Points of Order B-spline curve, where Represents the first control points, Respectively represent Control points in world coordinate system The position of the axis, Represent the 0th control point, the 1st control point, and the control points, the optimized cost function of the design It is expressed as: ； in, represents the smoothness cost, represents the collision cost, represents the dynamic feasibility cost, Respectively The corresponding weights; Among them, the smoothness cost of the design It is expressed as: ； in, represents the curvature penalty term, represents the elastic uniform term, For the control points, For the control points, is the normalization parameter, represents the regularization weight; The cost of design collision It is expressed as: ； ； in, Indicates Control Points The distance to the nearest obstacle, represents the collision cost mapping function, Indicates the safety distance threshold; Design of dynamic feasibility cost based on the updated dynamic boundary of the dual-mode UAV , the design's dynamic feasibility cost It is expressed as: ； ； in, Indicates The acceleration control points along The acceleration value of the axis, represents the dynamic feasibility cost mapping function, They respectively represent the flight mode along The lower and upper limits of the axis' acceleration, They represent the land mode along The lower and upper acceleration limits of the axis.

8. A dynamic motion planning system for a dual-mode land-air UAV, characterized in that: The dynamic motion planning system of the land-air dual-mode UAV includes: A disturbance estimation module is used to design a disturbance observer based on the dynamic model of the dual-mode UAV in the flight mode and the land mode by using uncertainty and disturbance estimation method, and to estimate the disturbance factors in the environment by using the disturbance observer to obtain a disturbance estimation value; A dynamic boundary adjustment module is used to calculate and update the dynamic boundary of the land-air dual-mode UAV in real time using a motion planner according to the disturbance estimation value and the dynamic equation in the dynamic model; Path search and trajectory optimization module: used to perform path search based on the updated dynamic boundary of the land-air dual-mode UAV using the improved motion dynamics A* algorithm, and to perform trajectory optimization using the improved B-spline algorithm, so as to obtain the trajectory planning result of the land-air dual-mode UAV.

9. A dual-mode land and air drone, characterized in that: The land-air dual-mode UAV includes: a disturbance observer and a motion planner, and the disturbance observer and the motion planner are used to jointly implement the steps of the dynamic motion planning method of the land-air dual-mode UAV as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a dynamic motion planning program for a land-to-air dual-mode UAV. When the dynamic motion planning program for the land-to-air dual-mode UAV is executed by a disturbance observer and a motion planner, the steps of the dynamic motion planning method for the land-to-air dual-mode UAV as described in any one of claims 1 to 7 are implemented.

Images