Land-air amphibious unmanned platform cooperative carrying method and system for non-specific target

Abstract

The invention discloses a land-air amphibious unmanned platform cooperative carrying method and system for an unspecific target, and relates to the field of unmanned platform cooperative carrying, and the method comprises the steps: constructing a land-air amphibious unmanned platform model, and determining a carrying performance boundary; adopting a pre-trained convolutional neural network to preliminarily estimate the mass and the centroid position of a carrying target, and determining the initial number and the initial position distribution of the air-ground amphibious unmanned platforms; calculating physical characteristic parameters of the carrying target based on the state information of each air-ground amphibious unmanned platform; determining the final number of the land-air amphibious unmanned platforms based on the quality of the carrying target and the carrying performance boundary, and determining the final position distribution of the land-air amphibious unmanned platforms based on a multi-target optimization model; and determining a control strategy of the air-ground amphibious unmanned platform of the carrying target based on the physical characteristic parameters of the carrying target, and realizing cooperative carrying of the air-ground amphibious unmanned platform. The air-ground cooperative carrying task under the complex environment can be achieved.

Description

Technical Field

[0001] This application relates to the field of collaborative handling of multiple unmanned platforms, and in particular to a method and system for collaborative handling of amphibious unmanned platforms for non-target purposes. Background Technology

[0002] In complex emergency scenarios such as disaster relief and field supply delivery, while unmanned aerial vehicle (UAV) platforms possess three-dimensional maneuverability, they suffer from short endurance, high energy consumption, and limited payload, making them unsuitable for long-duration, heavy-weight transport tasks. Ground-based mobile platforms, while having lower energy consumption and relatively higher payload capacity, are limited by terrain mobility, hindering their ability to effectively overcome obstacles or adapt to complex unstructured environments. In contrast, amphibious UAV platforms combine the advantages of both, enabling efficient and energy-saving transport in ground mode and switching to flight mode when encountering obstacles, demonstrating strong environmental adaptability and mission flexibility. However, a single amphibious UAV platform still has a payload limit, insufficient for transporting large, high-quality targets. Therefore, multi-platform collaborative operations have become an essential way to improve overall transport capabilities, which is of great significance for material transportation in complex emergency scenarios.

[0003] Existing collaborative handling methods for multiple unmanned platforms largely rely on the known physical parameters of the target object to determine the required number of platforms and collaborative control strategies. They also depend on pre-set gripping points or corresponding connection mechanisms, requiring manual marking or dedicated mechanical structures to achieve reliable connections between the platform and the target. These methods are unsuitable for unstructured targets with random patterns and unknown physical properties commonly encountered in disaster relief. Furthermore, existing collaborative control strategies often fail to adequately consider the dynamic coupling and force constraint changes during mode switching between amphibious platforms, lacking online estimation and robust control architectures for unknown targets, thus limiting their practicality in real-world, complex environments. Summary of the Invention

[0004] The purpose of this application is to provide a collaborative transport method and system for amphibious unmanned platforms targeting non-specific targets. This system can achieve collaborative transport tasks in complex environments through multi-platform collaborative perception, online parameter estimation, and distributed optimization control, even without knowing the physical properties of the target or having any pre-connection by human intervention.

[0005] To achieve the above objectives, this application provides the following solution: Firstly, this application provides a collaborative transport method for amphibious unmanned platforms targeting non-specific objectives, including: A land-air amphibious unmanned platform model was constructed, and the transport performance boundary was determined; the transport performance boundary includes the maximum thrust in ground driving mode and the maximum transport force in flight mode. A pre-trained convolutional neural network is used to initially estimate the mass and centroid position of the transported target, thus obtaining the initial mass and initial centroid position of the target. The initial number of amphibious unmanned platforms is determined based on the initial target mass, and the initial position distribution of the amphibious unmanned platforms is determined based on the platform initial distribution optimization model. The initial number and initial location distribution of amphibious unmanned platforms are used to perform the transport task, and the physical characteristic parameters of the transport target are calculated based on the state information of each amphibious unmanned platform. The physical characteristic parameters include the center of mass angular velocity, moment of inertia, relative position of the center of mass, center of mass velocity, and mass. The final number of amphibious unmanned platforms is determined based on the mass of the transport target and the transport performance boundary, and the final location distribution of the amphibious unmanned platforms is determined based on a multi-objective optimization model. The control strategy for the amphibious unmanned platform carrying the transport target is determined based on the physical characteristic parameters of the transport target; the control strategy includes a hybrid control strategy in ground driving mode and a hybrid control strategy in flight mode; Based on the final quantity, the final location distribution, and the control strategy, collaborative transport of amphibious unmanned platforms is achieved.

[0006] Secondly, this application provides a collaborative transport system for amphibious unmanned platforms targeting non-specific objectives, comprising: The model building and transport performance boundary determination module is used to build a land-air amphibious unmanned platform model and determine the transport performance boundary; the transport performance boundary includes the maximum thrust in ground driving mode and the maximum transport force in flight mode; The estimation module is used to initially estimate the mass and centroid position of the transported target using a pre-trained convolutional neural network, thereby obtaining the initial mass and initial centroid position of the target. The initial quantity and initial position distribution determination module is used to determine the initial quantity of the land and air amphibious unmanned platforms based on the target initial mass, and to determine the initial position distribution of the land and air amphibious unmanned platforms based on the platform initial distribution optimization model. The physical characteristic parameter calculation module is used to perform transport tasks on the land and air amphibious unmanned platforms with the initial quantity and initial position distribution, and to calculate the physical characteristic parameters of the transport target based on the state information of each land and air amphibious unmanned platform; the physical characteristic parameters include the center of mass angular velocity, moment of inertia, relative position of the center of mass, center of mass velocity, and mass; The final quantity and final location distribution determination module is used to determine the final quantity of the amphibious unmanned platforms based on the mass of the transport target and the transport performance boundary, and to determine the final location distribution of the amphibious unmanned platforms based on a multi-objective optimization model. The control strategy determination module is used to determine the control strategy of the amphibious unmanned platform for transporting the target based on the physical characteristic parameters of the target; the control strategy includes a hybrid control strategy in ground driving mode and a hybrid control strategy in flight mode; The collaborative transport module is used to achieve collaborative transport of amphibious unmanned platforms based on the final quantity, the final location distribution, and the control strategy.

[0007] According to the specific embodiments provided in this application, this application has the following technical effects: (1) This application calculates the physical characteristic parameters of the transport target based on the status information of each land and air amphibious unmanned platform, thereby determining the final number and final location distribution of the land and air amphibious unmanned platforms, and then uses the number of land and air amphibious unmanned platforms to carry out the transport, so as to achieve applicability to unknown transport targets under the premise of ensuring safety. (2) The control strategy in this application includes a hybrid control strategy in ground driving mode and a hybrid control strategy in flight mode. That is, when there are no obstacles during the transport process, the hybrid control strategy in ground driving mode is used to save energy, and when obstacles are encountered, the hybrid control strategy in flight mode is used to improve maneuverability. Attached Figure Description

[0008] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 A flowchart illustrating a collaborative transport method for amphibious unmanned platforms targeting non-specific objectives, provided as an embodiment of this application; Figure 2 This is a detailed flowchart illustrating a collaborative transport method for amphibious unmanned platforms targeting non-specific objectives, provided as an embodiment of this application. Detailed Implementation

[0010] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0011] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0012] In one exemplary embodiment, such as Figure 1 and Figure 2 As shown, a collaborative transport method for amphibious unmanned platforms targeting non-specific targets is provided. This method is executed by computer equipment, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method is described using a server as an example, and includes the following steps S1 to S7.

[0013] S1: Construct a land-air amphibious unmanned platform model and determine the transport performance boundary; the transport performance boundary includes the maximum thrust in ground driving mode and the maximum transport force in flight mode.

[0014] S2: A pre-trained convolutional neural network is used to initially estimate the mass and centroid position of the transported target, thus obtaining the initial mass and initial centroid position of the target.

[0015] S3: Determine the initial number of amphibious unmanned platforms based on the initial mass of the target, and determine the initial position distribution of the amphibious unmanned platforms based on the platform initial distribution optimization model.

[0016] S4: Perform the transport task on the amphibious unmanned platforms of the initial quantity and initial position distribution, and calculate the physical characteristic parameters of the transport target based on the state information of each amphibious unmanned platform; the physical characteristic parameters include the center of mass angular velocity, moment of inertia, relative position of the center of mass, center of mass velocity and mass.

[0017] S5: Determine the final number of amphibious unmanned platforms based on the mass of the transport target and the transport performance boundary, and determine the final location distribution of the amphibious unmanned platforms based on a multi-objective optimization model.

[0018] S6: Determine the control strategy of the amphibious unmanned platform for transporting the target based on the physical characteristic parameters of the target; the control strategy includes a hybrid control strategy in ground driving mode and a hybrid control strategy in flight mode.

[0019] S7: Based on the final quantity, the final location distribution, and the control strategy, realize the collaborative transport of amphibious unmanned platforms.

[0020] In a specific embodiment, step S1 specifically includes: (a) Constructing a land-air amphibious unmanned platform model.

[0021] For any amphibious unmanned platform (hereinafter referred to as the platform), the position and posture of the platform when it is traveling on the ground are defined as follows: The speed is The platform's position during flight Platform roll, pitch, and yaw attitude angles Platform speed Platform angular velocity Based on Newton-Euler equations, kinematic and dynamic models of amphibious unmanned platforms are established.

[0022] (1) Kinematic model: 1) Ground driving mode: The platform speed satisfies the following kinematic relationship, and the kinematic model is as follows: in, This refers to the Jacobian matrix in ground driving mode.

[0023] 2) Flight Mode: The kinematic model is represented by the transformation relationship of the platform's velocity vector in the world coordinate system and the body coordinate system: in, This is the rotation matrix between the world coordinate system and the body coordinate system. This is the transformation matrix between the Euler angle differential and angular velocity of the machine.

[0024] (2) Dynamic model: 1) Ground driving mode: Considering factors such as ground friction and hub motor thrust, a Newton-Euler equation, i.e., a dynamic model, is established: in, The mass-inertia matrix of the platform in ground driving mode. The matrix represents the Coriolis force and the centripetal force. The ground damping matrix is... The thrust and torque provided by the hub motors of amphibious unmanned platforms. This is an external disturbance.

[0025] 2) Flight Mode: Establish the Newton-Euler equations for the flight platform in the body coordinate system, i.e., the dynamic model: In the formula, For platform quality, For the platform inertia matrix, and Let $\mathbf$ be the net external force and net external torque acting on the platform, respectively, and given by the following formula: In the formula, and These are the aerodynamic forces and aerodynamic torques exerted on the fuselage by the ducted propeller, respectively. This refers to the air resistance experienced by the fuselage during flight. This is the total reaction force experienced by the platform when transporting the target object. This refers to the torque of the propeller gyroscope.

[0026] (ii) Analyze the transport performance boundaries of the amphibious unmanned platform (maximum thrust in ground driving mode and maximum transport force in flight mode).

[0027] (1) Ground driving mode: Maximum thrust is: in, The static friction coefficient of the ground is . This represents the positive pressure exerted by the platform on the ground.

[0028] (2) Flight mode: The maximum attitude angle can be obtained based on the platform hovering conditions. and maximum positive pressure for: Non-hovering state, meaning the platform has a certain acceleration. Maximum attitude angle and maximum positive pressure for: In the formula, For the platform's maximum total thrust, This refers to the frictional force experienced by the platform, and also the transport force applied to the target being transported.

[0029] Static friction must be maintained during flight, which means the conditions for a friction cone must be met: In the formula, The static friction coefficient of the target object can be used to obtain the maximum transport force. for: .

[0030] In a specific embodiment, step S2 specifically includes: preliminarily estimating the mass of the transport target, the number of platforms required, and the positional distribution of the platforms during transport based on visual recognition information.

[0031] Transport target quality and the position of the center of mass Preliminary estimation formula: in, The mapping function is based on a convolutional neural network and is obtained in advance through supervised learning. Image data collected by an airborne camera. This represents the quality estimation error term.

[0032] In a specific embodiment, step S3 specifically includes: Number of platforms in ground driving mode Preliminary determination: in, It is the acceleration due to gravity. For safety reasons, This indicates rounding up to the nearest integer.

[0033] Platform initial distribution optimization model: St in, Let the thrust vectors of each platform be the direction vectors. A random fixed-direction vector, These are the weighting coefficients. , They represent the first The and the first The location of the contact point between the platform and the transport target. This is the minimum anti-collision distance between platforms.

[0034] Each platform arrives at the designated location to perform the transport task. If it fails to move the object, the number of platforms is increased, and the platform positions are reallocated using the initial distribution optimization model until the transport target is moved.

[0035] In one specific embodiment, step S4 specifically includes: Based on the amphibious unmanned platform performing the transport task determined in step S3, define... For the center of mass of the transport target, and These represent the position and velocity of the target's center of mass, respectively. This represents the angular velocity of the center of mass of the target being transported. and They represent the first The position and speed of the contact point between the platform and the transported target and These represent the geometric centers of the contact points, respectively. Position and velocity, and They represent the first The position and speed of the platform's center of gravity Indicates the first The relative positions of the contact points of each platform to the geometric center of the contact points This indicates the relative position of the centroid of the transported object and the geometric center of the contact point. Indicates the first The and the first The relative positions of the contact points between the platform and the transport target.

[0036] (1) angular velocity of the center of mass The estimate.

[0037] The relative positions of the contact points between each platform and the transport target Estimate: In the formula, express The unit vector of the axis. express exist coordinates on, superscript It represents the orthogonal fill space.

[0038] Calculating the angular velocity of the center of mass from rigid body dynamics : (2) Moment of inertia of the transported target Relative position of the center of mass The estimate.

[0039] The rotational dynamics equation of the transported target is: In the formula, and Indicates the first The force and torque applied by the platform to the transported target, Apply the total external force to the platform to move the target. .

[0040] For a vector with constant norm, the following relationship holds: The two equations can be combined to establish a nonlinear observer: in, , , For the observation part, , This is the input known to the system.

[0041] The position of the centroid of the transported target can be obtained from this observer. and moment of inertia .

[0042] (3) Velocity of the center of mass The estimate.

[0043] The rigid body constraints for the amphibious unmanned platform and the transported target are as follows: By performing local estimations and rewriting the formulas for each amphibious unmanned platform, we can obtain: (4) Quality of transported target The estimate.

[0044] The rigid body dynamics of the transported target are as follows: The target quality for handling can be obtained by sorting. The estimated value is: .

[0045] In one specific embodiment, step S5 specifically includes: Considering that the transportation process involves both ground driving mode and flight mode, and the transportation conditions and forms are different in the two modes, the number of platforms and the allocation of positions for the two modes are calculated separately.

[0046] Ground driving mode: Re-determine the final number of platforms required. : The platform's location distribution needs to ensure that the total thrust direction is consistent with the motion direction and to avoid collisions between platforms. A multi-objective optimization model is used to determine this. in, For the first The contact point between the amphibious unmanned platform and the transported target. For the first The direction vector of the thrust of an amphibious unmanned platform. The total number of amphibious unmanned platforms, To determine the direction of the target's movement, These are the weighting coefficients.

[0047] Flight mode: Determines the final number of platforms required when encountering obstacles. : The platform's location distribution needs to satisfy the following conditions: the total frictional force can counteract the weight of the transported target; the direction of the total pressure must be consistent with the direction of motion; and collisions between platforms must be avoided. A multi-objective optimization model is used to determine this. in For the first The contact point between the amphibious unmanned platform and the target being transported in flight mode. For the first The direction vector of thrust applied in the flight mode of an amphibious unmanned platform. Indicates the first The carrying force exerted on the target by the flight platform The unit vector that applies the total force to the flight platform. These are the weighting coefficients.

[0048] In one specific embodiment, step S6 specifically includes: S61: Determine the desired trajectory based on the transportation task.

[0049] Land-air integrated planning strategy: Based on information obtained from cameras on various platforms, determine in real time whether there are obstacles. If there are obstacles, plan the air trajectory and switch the flight mode. After passing the obstacles, switch to the ground driving mode.

[0050] Given the target location of the transport target and target posture Considering the compliance and stability of the trajectory during the transportation process, the desired position of the transported target during the transportation process is designed and generated. Expected speed Expected posture Desired angular velocity for: in, To achieve the desired dynamic trajectory damping, To determine the desired dynamic trajectory bandwidth, and This represents the desired attitude trajectory gain. The compliance and stability of the trajectory during the transport process are improved by adjusting parameters. , , , accomplish.

[0051] Trajectory generation uses a time-optimal method: Each platform uses onboard cameras, LiDAR, and other sensors to build real-time maps of the surrounding environment and detect obstacles. If an obstacle is detected, a set of obstacles is defined. Circumsphere radius of the transport target Determine if the obstacle satisfies: in, Let be any point on the obstacle. This is a safe distance (generally 0.3m).

[0052] If the obstacle is within a safe distance, the platform triggers the obstacle avoidance strategy and switches to flight mode, planning an obstacle avoidance trajectory: in, For obstacle avoidance magnitude vector, The time for obstacle avoidance to begin. To control the duration of obstacle avoidance, This is the unit vector for obstacle avoidance direction.

[0053] S62: Determine the expected contact force applied by the amphibious unmanned platform to the transport target based on the expected trajectory and the physical characteristic parameters of the transport target.

[0054] (1) Force distribution strategy in flight mode.

[0055] The expected net external force on the transported target can be obtained from the expected trajectory. and expected resultant external torque : in, and These are the tracking error damping terms for velocity and angular velocity, respectively.

[0056] The following optimization problem will be addressed. and Distributed to various platforms: in, For the desired contact force of the flight platform, This refers to the relative position between the center of mass of the flight platform and the target being transported.

[0057] Constraints: Each platform must meet the maximum attitude angle constraint: ; Positive pressure is expected on various flight platforms scope: ; in, For the friction of each flight platform, The maximum pressure that each parallel platform can provide.

[0058] (2) Force distribution strategy in ground driving mode.

[0059] In ground driving mode, the desired torque is generally 0, and the desired net external force is mainly used to overcome ground friction and drive the target to move along the desired trajectory. Its expression is: in, For ground friction, This is the speed tracking error damping term.

[0060] Thrust distribution across platforms can be achieved by optimizing the following objectives: Constraints: Maximum thrust limits for each platform: .

[0061] S63: Determine the control strategy for the amphibious unmanned platform under different motion modes based on the expected contact force.

[0062] For platforms in ground driving or flight mode, the system receives two types of instructions from the upper-level planner: one is the platform's desired trajectory, including its position. ,speed and acceleration Secondly, the desired contact force that the platform needs to apply to the target, determined by the force distribution strategy. .

[0063] (1) Hybrid control strategy in ground driving mode.

[0064] The process for determining the hybrid control strategy in the ground driving mode is as follows: determine the virtual equilibrium position based on the desired contact force; generate the desired driving trajectory based on the virtual equilibrium position; and calculate the total control force of the motor using a combination of feedforward control and feedback control based on the desired driving trajectory.

[0065] In ground driving mode, the platform is driven by hub motors. To achieve precise tracking of the macroscopic path and actively control the interaction force between the platform and the transport target, a cascaded control architecture based on dynamic equilibrium point admittance control is designed. This architecture consists of three layers from the outside in: a power controller, an admittance controller, and a trajectory tracking controller.

[0066] Active force controller: to enable the system to actively track non-zero desired contact force. A dynamically adjustable virtual equilibrium position can be designed. : in, The proportional gain of the main power controller The integral gain of the main power controller.

[0067] Admittance controller: based on the current interaction force And the virtual equilibrium position given by the outer ring Generate a smoothed desired driving trajectory: in, , and As the adjustable virtual inertia, damping, and stiffness matrices are defined, the desired driving trajectory can be obtained by solving the differential equation in real time. .

[0068] Track tracking controller: Utilizes a control method combining feedforward and feedback control to calculate the total control force of the motor. : in, The inertia matrix in the dynamic model, For the Coriolis force and centripetal force terms, For gravity, For the proportional gain of the trajectory tracking controller, This is the differential gain of the trajectory tracking controller.

[0069] (2) Hybrid control strategy in flight mode.

[0070] Specifically, the process of determining the hybrid control strategy in the flight mode is as follows: determining the position adjustment amount based on the desired contact force; generating the desired adjustment position based on the position adjustment amount; calculating the total thrust in the world coordinate system using a position control loop based on the desired adjustment position; calculating the desired attitude and the total thrust in the body coordinate system based on the total thrust in the world coordinate system; calculating the three-axis control torque using an attitude control loop based on the desired attitude; and calculating the control signals of each ducted propeller of the amphibious unmanned platform based on the total thrust in the body coordinate system and the three-axis control torque.

[0071] To achieve precise tracking of the desired trajectory and actively control the interaction force between the target and the target being transported in flight mode, a hybrid power / motion control strategy based on dynamic position correction is designed. To resolve the conflict between force control and position control, a force control loop is designed to mitigate force errors. Mapped to a position adjustment amount This adjustment amount is located in the direction of the normal vector of the contact point and is used to fine-tune the platform position to maintain or adjust the positive pressure.

[0072] in, For the proportional gain of the force control loop in flight mode, For the integral gain of the force control loop in flight mode, This is the unit vector normal to the contact point.

[0073] The expected position given by trajectory planning The position adjustment amount calculated by the force controller Combined, to generate the desired adjustment position for position control. : The total thrust in the world coordinate system required to track this trajectory is calculated by the position control loop. : in, For the proportional gain of the flight mode position control loop, The differential gain of the flight mode position control loop, from which the total thrust can be used to calculate the desired attitude required for the corrected trajectory. Total thrust in the body coordinate system .

[0074] The three-axis control torques acting on the platform body are calculated by the attitude control loop. : in, This is the proportional gain of the attitude control loop. This is the differential gain of the attitude control loop.

[0075] Ultimately, the total thrust in the body coordinate system With three-axis control torque The control and distribution module calculates the input signals for the actuators of each ducted propeller, driving the platform to complete the predetermined motion and force control tasks.

[0076] The method provided in this application has the following advantages: (1) A phased collaborative control strategy of “ground parameter estimation-land-air collaborative transport” is proposed. In the ground driving mode, the target mass, center of mass and moment of inertia are estimated online by contact force and motion information to determine the number of platforms required for transport. Then, the transport is carried out by the number of platforms, so as to achieve applicability to unknown targets under the premise of ensuring safety. (2) When there are no obstacles during the transportation process, the ground driving mode is used to save energy, and the air flight mode is used to improve maneuverability when obstacles are encountered; (3) Establish a land and air amphibious unmanned platform model and analyze its transport performance boundary. The transport is achieved by utilizing the platform's normal pressure and friction. Construct a multi-platform force distribution optimization strategy that considers friction cone, maximum thrust, and attitude stability. Complex operating mechanisms such as suction cups, magnets, and grippers are not required. (4) A force / motion hybrid control strategy was designed, which adopts dynamic equilibrium point admittance control and position correction-based active control for ground driving and flight modes respectively, so as to realize the platform's coordinated tracking of the desired trajectory and contact force under mode switching and external disturbances, effectively improving the system's adaptability and robustness in unknown environments.

[0077] Based on the same inventive concept, this application also provides a collaborative transport system for amphibious unmanned platforms oriented towards non-specific targets. The solution provided by this system is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the collaborative transport system for amphibious unmanned platforms oriented towards non-specific targets provided below can be found in the limitations of the collaborative transport method for amphibious unmanned platforms oriented towards non-specific targets described above, and will not be repeated here.

[0078] In one exemplary embodiment, a collaborative transport system for amphibious unmanned platforms oriented towards non-specific targets is provided, comprising the following modules.

[0079] The model construction and transport performance boundary determination module is used to construct a land-air amphibious unmanned platform model and determine the transport performance boundary; the transport performance boundary includes the maximum thrust in ground driving mode and the maximum transport force in flight mode.

[0080] The estimation module is used to initially estimate the mass and centroid position of the transported target using a pre-trained convolutional neural network, thereby obtaining the initial mass and initial centroid position of the target.

[0081] The initial quantity and initial position distribution determination module is used to determine the initial quantity of the land and air amphibious unmanned platforms based on the target initial mass, and to determine the initial position distribution of the land and air amphibious unmanned platforms based on the platform initial distribution optimization model.

[0082] The physical characteristic parameter calculation module is used to perform the transport task on the amphibious unmanned platforms with the initial quantity and initial position distribution, and to calculate the physical characteristic parameters of the transport target based on the state information of each amphibious unmanned platform; the physical characteristic parameters include the center of mass angular velocity, moment of inertia, relative position of the center of mass, center of mass velocity, and mass.

[0083] The final quantity and final location distribution determination module is used to determine the final quantity of the amphibious unmanned platforms based on the mass of the transport target and the transport performance boundary, and to determine the final location distribution of the amphibious unmanned platforms based on a multi-objective optimization model.

[0084] The control strategy determination module is used to determine the control strategy of the amphibious unmanned platform for transporting the target based on the physical characteristic parameters of the target; the control strategy includes a hybrid control strategy in ground driving mode and a hybrid control strategy in flight mode.

[0085] The collaborative transport module is used to achieve collaborative transport of amphibious unmanned platforms based on the final quantity, the final location distribution, and the control strategy.

[0086] In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments. The computer device can be a server or a terminal. The computer device includes a processor, memory, an input / output interface (I / O), and a communication interface. The processor, memory, and I / O interface are connected via a system bus, and the communication interface is connected to the system bus via the I / O interface. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device stores data to be processed. The I / O interface of the computer device is used for exchanging information between the processor and external devices. The communication interface of the computer device is used for communicating with an external terminal via a network connection. When the computer program is executed by the processor, it implements the steps in the above-described method embodiments.

[0087] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0088] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0089] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0090] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0091] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A non-target-specific amphibious unmanned platform cooperative carrying method, characterized in that, The method comprises the following steps: constructing a model of the amphibious unmanned platform and determining a carrying performance boundary; the carrying performance boundary comprises a maximum thrust in a ground travel mode and a maximum carrying force in a flight mode; preliminarily estimating the mass and the mass center position of the carrying target by using a pre-trained convolutional neural network to obtain an initial mass of the target and an initial mass center position of the target; determining an initial number of the amphibious unmanned platform based on the initial mass of the target and determining an initial position distribution of the amphibious unmanned platform based on a platform initial distribution optimization model; carrying out a carrying task by the amphibious unmanned platform with the initial number and the initial position distribution and calculating physical characteristic parameters of the carrying target based on state information of each amphibious unmanned platform; the physical characteristic parameters comprise an angular velocity of the mass center, a moment of inertia, a relative position of the mass center, a velocity of the mass center and a mass; determining a final number of the amphibious unmanned platform based on the mass of the carrying target and the carrying performance boundary and determining a final position distribution of the amphibious unmanned platform based on a multi-objective optimization model; determining a control strategy of the amphibious unmanned platform for the carrying target based on the physical characteristic parameters of the carrying target; the control strategy comprises a hybrid control strategy in the ground travel mode and a hybrid control strategy in the flight mode; realizing cooperative carrying of the amphibious unmanned platform based on the final number, the final position distribution and the control strategy.

2. The non-target-specific amphibious unmanned platform cooperative handling method according to claim 1, characterized in that, The motion modes of the amphibious unmanned platform comprise a ground travel mode and a flight mode; the model of the amphibious unmanned platform comprises a kinematic model and a dynamic model; the kinematic model in the ground travel mode is: wherein a pose of the amphibious unmanned platform when driving on the ground a first derivative of the pose, a velocity of the amphibious unmanned platform when driving on the ground a Jacobian matrix in the ground driving mode; the kinematic model in the flight mode is: in, The speed of the amphibious unmanned platform during flight. The angular velocity of the amphibious unmanned platform during flight. This is the rotation matrix between the world coordinate system and the body coordinate system. Position of the amphibious unmanned platform during flight The first derivative, This is the transformation matrix between the Euler angle differential and angular velocity of the machine. The yaw attitude angle of an amphibious unmanned platform during flight The first derivative; the dynamic model in the ground travel mode is: wherein, is the mass-inertia matrix of the amphibious unmanned platform in ground driving mode, is the Coriolis and centripetal force matrix, is the ground damping matrix, is the thrust and torque provided by the wheel hub motors of the amphibious unmanned platform, is the external disturbance; the dynamic model in the flight mode is: wherein, is the mass of the amphibious unmanned platform, is the matrix of the moments of inertia of the amphibious unmanned platform, , are the resultant external force and the resultant external moment, respectively, to which the amphibious unmanned platform is subjected, is the first derivative of .

3. The non-target-specific amphibious unmanned platform cooperative handling method according to claim 2, characterized in that, the calculation formula of the initial number of the amphibious unmanned platform is: wherein is the initial number of amphibious unmanned platforms, is the initial mass of the transported object, is the acceleration of gravity, is the ground static friction coefficient, is the safety factor, is the maximum thrust in ground travel mode, denotes the ceiling function; the expression of the platform initial distribution optimization model is: S.t. wherein, is the direction vector of the thrust of the th amphibious unmanned platform, is a random fixed direction vector, is the direction vector of the thrust of the th amphibious unmanned platform, is the position of the contact point between the th amphibious unmanned platform and the target to be transported, is the minimum anti-collision distance between the amphibious unmanned platforms, is a weight coefficient.

4. The non-target-specific amphibious unmanned platform cooperative handling method according to claim 3, characterized in that, in the ground travel mode, the calculation formula of the final number of the amphibious unmanned platform is: wherein, is the final number of amphibious unmanned platforms in ground mode, is the mass of the object to be transported; in the ground travel mode, the expression of the multi-objective optimization model is: wherein, is the position of the contact point of the nthairborne amphibious unmanned platform with the transport target, is the position of the contact point of the nthairborne amphibious unmanned platform with the transport target, is the direction vector of the thrust of the nthairborne amphibious unmanned platform, is the direction vector of the thrust of the nthairborne amphibious unmanned platform, is the total number of airborne amphibious unmanned platforms, is the direction of the movement of the transport target, is the weight coefficient.

5. The non-target-specific amphibious unmanned platform cooperative handling method according to claim 4, characterized in that, in the flight mode, the calculation formula of the final number of the amphibious unmanned platform is: wherein, is the final number of amphibious unmanned platforms in flight mode; is the maximum carrying capacity in flight mode; in the flight mode, the expression of the multi-objective optimization model is: wherein, is the position of the contact point between the is the direction vector of the applied force in the flight mode of the is the position of the contact point between the is the direction vector of the applied force in the flight mode of the is the position of the contact point between the is the position of the contact point between the is the position of the contact point between the is the position of the contact point between the 6. The non-target-specific amphibious unmanned platform cooperative handling method according to claim 1, characterized in that, determining a control strategy of the amphibious unmanned platform for the carrying target based on the physical characteristic parameters of the carrying target, specifically comprising: determining an expected trajectory according to the carrying task; determining an expected contact force applied to the carrying target by the amphibious unmanned platform based on the expected trajectory and the physical characteristic parameters of the carrying target; determining the control strategy of the amphibious unmanned platform based on the expected contact force.

7. The non-target-specific amphibious unmanned platform cooperative handling method according to claim 6, characterized in that, The determination process of the hybrid control strategy in the ground travel mode is as follows: determining a virtual balance position based on the expected contact force; generating a travel expected trajectory based on the virtual balance position; calculating the total control force of the motor by using a combination of feedforward control and feedback control based on the travel expected trajectory.

8. The non-target-specific amphibious unmanned platform cooperative handling method according to claim 6, characterized in that, The determination process of the hybrid control strategy in the flight mode is as follows: determining a position adjustment amount based on the expected contact force; generating an adjusted expected position based on the position adjustment amount; Based on the adjusted desired position, a position control loop is employed to calculate a world coordinate system total thrust; Based on the world coordinate system total thrust, a desired attitude and a body coordinate system total thrust are solved; Based on the desired attitude, a three-axis control moment is calculated by an attitude control loop; Based on the body coordinate system total thrust and the three-axis control moment, control signals of each ducted propeller of the amphibious unmanned platform are calculated.

9. The non-target-specific amphibious unmanned platform cooperative handling method according to claim 6, characterized in that, After determining the desired trajectory according to the carrying task, further comprising: When an obstacle is detected, the expected trajectory is updated according to the formula wherein, is the updated expected trajectory, is the expected position of the carrying target at time t, is the obstacle avoidance amplitude vector, is the obstacle avoidance start time, is the control obstacle avoidance duration, is the obstacle avoidance direction unit vector.

10. A non-target-specific amphibious unmanned platform cooperative handling system, characterized in that, Comprising: A model construction and carrying performance boundary determination module is configured to construct an amphibious unmanned platform model and determine a carrying performance boundary; the carrying performance boundary comprises a maximum thrust in a ground driving mode and a maximum carrying force in a flight mode; An estimation module is configured to preliminarily estimate the mass and the mass center position of the carrying target by using a pre-trained convolutional neural network, to obtain a target initial mass and a target initial mass center position; An initial quantity and initial position distribution determination module is configured to determine an initial quantity of the amphibious unmanned platform based on the target initial mass, and determine an initial position distribution of the amphibious unmanned platform based on a platform initial distribution optimization model; A physical characteristic parameter calculation module is configured to execute the carrying task by the amphibious unmanned platform with the initial quantity and the initial position distribution, and calculate physical characteristic parameters of the carrying target based on state information of each amphibious unmanned platform; the physical characteristic parameters comprise a mass center angular velocity, a moment of inertia, a mass center relative position, a mass center velocity and a mass; A final quantity and final position distribution determination module is configured to determine a final quantity of the amphibious unmanned platform based on the mass of the carrying target and the carrying performance boundary, and determine a final position distribution of the amphibious unmanned platform based on a multi-objective optimization model; A control strategy determination module is configured to determine a control strategy of the amphibious unmanned platform for the carrying target based on the physical characteristic parameters of the carrying target; the control strategy comprises a hybrid control strategy in a ground driving mode and a hybrid control strategy in a flight mode; A cooperative carrying module is configured to implement cooperative carrying of the amphibious unmanned platform based on the final quantity, the final position distribution and the control strategy.

Images