Pose control method, device and equipment for reconfigurable unmanned vehicle docking mechanism

Abstract

The invention provides a pose control method, device and equipment for a reconfigurable unmanned vehicle docking mechanism, and relates to the technical field of pose control, and the method comprises the steps: inputting an initial pose and a target pose of a reconfigurable unmanned vehicle into a pre-trained deep learning model, and obtaining a first adjustment scheme of each first unmanned vehicle module; correcting each first adjustment scheme according to the working terrain to obtain a second adjustment scheme of the corresponding first unmanned vehicle module, and determining a first pose control scheme of each first docking mechanism; and controlling each first docking mechanism of the reconfigurable unmanned vehicle to perform pose adjustment according to the first pose control scheme, and performing real-time adjustment on the first pose control scheme by using real-time data acquired by the state acquisition device to obtain a real-time pose control scheme of each first docking mechanism. According to the invention, when pose control is carried out on the reconfigurable unmanned vehicle docking mechanism, the robustness of pose control can be ensured, so that successful docking of all unmanned vehicle modules can be ensured.

Description

Technical Field

[0001] This invention relates to the field of posture control technology, and in particular to a posture control method, device and equipment for a reconfigurable unmanned vehicle docking mechanism. Background Technology

[0002] Reconfigurable unmanned vehicles (UAVs) are a new type of unmanned system that breaks through the limitations of traditional fixed configurations. Their core lies in the dynamic reconfiguration of function, topology, and form through the autonomous assembly and disassembly of modules. Composed of multiple modules, each linked to others via docking mechanisms, reconfigurable UAVs can be combined into equipment with different functions such as cargo transport and rescue, depending on mission requirements. When faced with battlefield damage or complex road conditions, they can adjust the orientation of the connecting mechanisms to detach damaged modules and autonomously escape, significantly improving adaptability to complex environments. They have the potential to transform the application modes of existing unmanned equipment in fields such as military transportation, intelligent transportation, and emergency rescue.

[0003] Existing pose control methods for reconfigurable unmanned vehicle docking mechanisms require first constructing an integrated mathematical model of the servo motor and mechanical transmission of the electric cylinder, clarifying the mapping relationship between input and output, and then compensating for the disturbance of end-effector pose by the friction force of the Hooke joint through a joint friction dynamics model. Next, a single-cylinder active disturbance rejection controller is used to suppress external disturbances such as load fluctuations and road bumps. Finally, the multi-agent consensus theory is used to achieve six-cylinder coordination, while combining vision and laser multi-sensor fusion to ensure pose estimation accuracy, ultimately achieving high-precision pose control of the docking mechanism.

[0004] Existing pose control methods are highly dependent on mathematical models, lack robustness under complex dynamic disturbances, and are difficult to fully adapt to complex ground environments. Summary of the Invention

[0005] This invention provides a pose control method, device, and equipment for a reconfigurable unmanned vehicle docking mechanism to solve the problems of existing pose control methods being highly dependent on mathematical models, lacking robustness under complex dynamic disturbances, and being difficult to fully adapt to complex ground environments.

[0006] In a first aspect, embodiments of the present invention provide a pose control method for a reconfigurable unmanned vehicle docking mechanism, comprising: The initial pose, target pose, and application scenario of the reconfigurable autonomous vehicle are input into a pre-trained deep learning model to obtain a first adjustment scheme for each first autonomous vehicle module of the reconfigurable autonomous vehicle; wherein, the reconfigurable autonomous vehicle includes multiple first autonomous vehicle modules and one second autonomous vehicle module; each first autonomous vehicle module includes a first docking mechanism; the second autonomous vehicle module includes multiple second docking mechanisms; Each first adjustment scheme is modified according to the working terrain of the reconfigurable unmanned vehicle to obtain the corresponding second adjustment scheme of the first unmanned vehicle module; Based on the second adjustment scheme of each first unmanned vehicle module of the reconfigurable unmanned vehicle, the first pose control scheme of the corresponding first docking mechanism of each first unmanned vehicle module is determined; The first docking mechanism of each first unmanned vehicle module of the reconfigurable unmanned vehicle is controlled, and the posture is adjusted according to the corresponding first posture control scheme. Based on the real-time data collected by the state acquisition device pre-set on each first docking mechanism, the corresponding first posture control scheme is adjusted in real time to obtain the real-time posture control scheme of each first docking mechanism.

[0007] Secondly, embodiments of the present invention provide a pose control device for a reconfigurable unmanned vehicle docking mechanism, comprising: The learning module is used to input the initial pose, target pose, and application scenario of the reconfigurable autonomous vehicle into a pre-trained deep learning model to obtain a first adjustment scheme for each first autonomous vehicle module of the reconfigurable autonomous vehicle; wherein, the reconfigurable autonomous vehicle includes multiple first autonomous vehicle modules and one second autonomous vehicle module; each first autonomous vehicle module includes a first docking mechanism; the second autonomous vehicle module includes multiple second docking mechanisms; The correction module is used to correct each first adjustment scheme according to the working terrain of the reconfigurable unmanned vehicle, so as to obtain the corresponding second adjustment scheme of the first unmanned vehicle module. The determination module is used to determine the first pose control scheme of the corresponding first docking mechanism of each first autonomous vehicle module based on the adjustment scheme of each first autonomous vehicle module of the reconfigurable autonomous vehicle; The adjustment module is used to control the corresponding first docking mechanism of each first unmanned vehicle module of the reconfigurable unmanned vehicle, adjust the posture according to the corresponding first posture control scheme, and adjust the corresponding first posture control scheme in real time based on the real-time data collected by the state acquisition device pre-set on each first docking mechanism to obtain the real-time posture control scheme of each first docking mechanism.

[0008] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect or any possible implementation thereof.

[0009] In this embodiment of the invention, the initial pose, target pose, and application scenario of the reconfigurable unmanned vehicle (UAV) are input into a deep learning model to obtain a first adjustment scheme for each first UAV module. This reduces the dependence of the pose control method on the mathematical model. Each first adjustment scheme is then corrected based on the working terrain of the UAV to obtain a second adjustment scheme for each first UAV module. This determines the first pose control scheme for each first docking mechanism. During pose control, the influence of complex ground environments is considered. Real-time data collected by a pre-set state acquisition device on the first docking mechanism is used to adjust each first pose control scheme in real time, resulting in a real-time pose control scheme for each first docking mechanism of the reconfigurable UAV. This ensures robustness of pose control under complex dynamic disturbances, improves the accuracy of the pose control scheme, and guarantees successful docking of each UAV module of the reconfigurable UAV. Attached Figure Description

[0010] Figure 1 This is a flowchart illustrating the implementation of the pose control method for the reconfigurable unmanned vehicle docking mechanism provided in this embodiment of the invention. Figure 2 This is a side view of the first unmanned vehicle module of the pose control method for the reconfigurable unmanned vehicle docking mechanism provided in an embodiment of the present invention. Figure 3 This is a flowchart illustrating the implementation of step S120 of the pose control method for the reconfigurable unmanned vehicle docking mechanism provided in this embodiment of the invention. Figure 4 This is a schematic diagram of the docking surface of the first docking mechanism in the pose control method of the reconfigurable unmanned vehicle docking mechanism provided in an embodiment of the present invention; Figure 5 This is a flowchart illustrating the implementation of step S140 of the pose control method for the reconfigurable unmanned vehicle docking mechanism provided in this embodiment of the invention. Figure 6 This is a schematic diagram of the posture control device for the reconfigurable unmanned vehicle docking mechanism provided in an embodiment of the present invention; Figure 7 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0011] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0012] See Figure 1 The diagram illustrates the implementation flowchart of the pose control method for the reconfigurable unmanned vehicle docking mechanism provided in this embodiment of the invention, which is described in detail below: Step S110: Input the initial pose, target pose, and application scenario of the reconfigurable autonomous vehicle into the pre-trained deep learning model to obtain the first adjustment scheme of each first autonomous vehicle module of the reconfigurable autonomous vehicle.

[0013] In some embodiments, the reconfigurable autonomous vehicle includes multiple first autonomous vehicle modules and one second autonomous vehicle module; each first autonomous vehicle module includes a first docking mechanism; the second autonomous vehicle module includes multiple second docking mechanisms, and the first autonomous vehicle module consists of a vehicle body and the first docking mechanisms. A side view of the first autonomous vehicle module is shown below. Figure 2 As shown. Reconfigurable unmanned vehicles (UAVs) refer to a new type of unmanned system that breaks through the limitations of traditional fixed configurations. Their core feature is the dynamic adjustment of function, topology, and form through the autonomous assembly and disassembly of multiple modules. They can be combined into different functional equipment according to mission requirements. For example, in military transport scenarios, they can be assembled into a heavy-duty cargo configuration; when a module is damaged on the battlefield, it can detach and autonomously escape; in emergency rescue scenarios, the module combination can be adjusted to adapt to narrow rescue passages. Initial pose refers to the combined spatial position and attitude of the first UAV module before docking adjustments begin. Target pose is the preset spatial position and attitude that the reconfigurable UAV needs to achieve after re-docking the first and second UAV modules to meet functional requirements. Application scenario refers to the environmental type or mission type in which the reconfigurable UAV actually performs its tasks. Different scenarios will affect the module combination requirements and docking strategies of the UAV.

[0014] It should be noted that the pre-trained deep learning model is trained using historical pose adjustment data from the reconfigurable autonomous vehicle (RV). During training, the pre-trained deep learning model needs to utilize historical pose adjustment data recorded by the RV's modules under different scenarios and terrains. It must be capable of receiving initial pose, target pose, and application scenario information, and outputting an adjustment plan for the first RV module. The first RV module is a component of the reconfigurable RV, consisting of multiple modules, primarily responsible for interfacing with second RV modules to achieve functional expansion. Each first RV module is equipped with a first docking mechanism for docking. The first adjustment plan refers to the preliminary adjustment plan generated for each first RV module by the pre-trained deep learning model based on the input initial pose, target pose, and application scenario. The first adjustment plan only considers the input information and does not take into account the influence of the actual working terrain. The second RV module is the core component of the reconfigurable RV, equipped with multiple second docking mechanisms for docking with multiple first RV modules. It typically undertakes key functions such as central control and core power. The first docking mechanism refers to the component installed on each first autonomous vehicle module, used to cooperate with the second docking mechanism of the second autonomous vehicle module to achieve physical connection and power transmission between the modules. The second docking mechanism refers to the component installed on the second autonomous vehicle module, used to cooperate with the first docking mechanism of the first autonomous vehicle module for docking. The number of second docking mechanisms matches the number of first autonomous vehicle modules.

[0015] Step S120: Modify each first adjustment scheme according to the working terrain of the reconfigurable unmanned vehicle to obtain the corresponding second adjustment scheme of the first unmanned vehicle module.

[0016] In some embodiments, the working terrain refers to the ground environment in which the reconfigurable autonomous vehicle (RV) performs its task, including features such as the direction and angle of inclination of the ground. These features affect the adjustment accuracy of the first RV module. For example, the RV may be working on "mountainous terrain with a steep slope," where the ground is tilted to the left at an angle of 15°. The second adjustment scheme refers to an adjustment plan obtained by modifying the first adjustment scheme based on the working terrain features of the reconfigurable RV. This modification aims to offset the impact of the terrain on the adjustment of the first RV module.

[0017] See Figure 3 The specific implementation of step S120 above includes steps S1201 and S1202, the details of which are as follows: Step S1201: Determine the tilt angle as the basic position deviation angle of the second unmanned vehicle module, and determine the tilt direction as the basic position deviation direction of the second unmanned vehicle module.

[0018] It should be noted that the working terrain information includes the tilt angle and tilt direction. The tilt angle refers to the degree to which the ground is tilted relative to a horizontal state in the working terrain of the reconfigurable unmanned vehicle. The tilt angle of the working terrain is directly determined as the basic position deviation angle of the second unmanned vehicle module. The basic position deviation angle is used to quantify the degree to which the second unmanned vehicle module deviates from the ideal docking reference position due to the terrain tilt. Since the second unmanned vehicle module is the docking core, its position deviation will directly affect the docking accuracy of the first unmanned vehicle module. Therefore, the terrain tilt angle needs to be converted into the deviation angle of the second module as the basis for correcting the first adjustment scheme. The tilt direction refers to the specific orientation of the ground tilt in the working terrain of the reconfigurable unmanned vehicle. It is a key parameter describing the slope orientation of the working terrain and determines the specific direction of the position deviation of the second unmanned vehicle module. For example, when the reconfigurable unmanned vehicle performs a task on a road surface with damage on one side, the road surface tilts to the right (the side closer to the roadside ditch). This "to the right" orientation is the tilt direction of the working terrain; this direction will cause the position of the second unmanned vehicle module to shift to the right. The slope direction of the working terrain is directly determined as the basic position deviation direction of the second unmanned vehicle module. This basic position deviation direction is used to clarify the specific orientation of the second unmanned vehicle module due to the terrain slope, which deviates from the ideal docking reference position. Since the deviation direction of the second unmanned vehicle module will directly cause directional deviation when the first unmanned vehicle module docks according to the original first adjustment scheme, it is necessary to convert the terrain slope direction into the deviation direction of the second module to provide a directional basis for correcting the first adjustment scheme.

[0019] Step S1202: Based on the basic position deviation angle and the basic position deviation direction, each first adjustment scheme is corrected to obtain the corresponding second adjustment scheme for the first unmanned vehicle module.

[0020] In some embodiments, the first adjustment scheme includes the corresponding movement direction, movement distance, adjustment time, rotation direction, and rotation angle for each first unmanned vehicle module.

[0021] In one possible implementation, step S1202 is specifically processed as follows: superimposing the basic position deviation direction with the movement direction in the corresponding first adjustment scheme of each first unmanned vehicle module to obtain the final movement direction in the corresponding second adjustment scheme; calculating the final movement distance of the corresponding second adjustment scheme of each first unmanned vehicle module based on the basic position deviation angle and the movement distance in the first adjustment scheme of each first unmanned vehicle module; replacing the movement direction in each first adjustment scheme with the corresponding final movement direction, and replacing the movement distance in each first adjustment scheme with the corresponding final movement distance to obtain the corresponding second adjustment scheme.

[0022] In some embodiments, the movement direction in the first adjustment scheme is a preliminary movement orientation set only based on the initial pose, target pose, and application scenario when the pre-trained deep learning model generates the first adjustment scheme for the first autonomous vehicle module. The operation of combining the basic position deviation direction of the second autonomous vehicle module (i.e., the deviation direction caused by terrain tilt) with the movement direction in the first adjustment scheme of the first autonomous vehicle module using spatial direction synthesis logic is essentially to offset the position deviation of the second autonomous vehicle module, ensuring that the movement direction of the first autonomous vehicle module aligns with the docking position after the actual deviation. The final movement direction in the second adjustment scheme refers to the corrected movement orientation obtained by superimposing the basic position deviation direction and the movement direction in the first adjustment scheme. This direction has been adapted to the position deviation of the second autonomous vehicle module, ensuring that the first autonomous vehicle module moves towards the actual docking position. The movement distance in the first adjustment scheme refers to the preliminary movement length set only based on the horizontal straight-line distance between the initial pose and the target pose of the first autonomous vehicle module in the first adjustment scheme, without considering the actual path lengthening or shortening caused by terrain tilt. The process of calculating the corrected movement length based on the basic position deviation angle of the second unmanned vehicle module and the movement distance in the first adjustment scheme, through the conversion between slope distance and horizontal distance, aims to ensure that the first unmanned vehicle module accurately reaches the docking preparation position after moving along the final movement direction. For example, if the basic position deviation angle is the angle between the slope and the horizontal plane, and the movement distance in the first adjustment scheme is the horizontal distance, the length required for the first unmanned vehicle module to move along the slope is obtained through geometric calculation of "horizontal distance ÷ cos(inclination angle)". The final movement distance in the second adjustment scheme refers to the corrected movement length obtained through the above calculation for the second adjustment scheme. This length takes into account the path changes caused by terrain inclination, ensuring that the first unmanned vehicle module can reach the docking position with the second unmanned vehicle module exactly after moving along the final movement direction. For example, the movement length along the slope obtained through the above geometric calculation is the final movement distance of the first unmanned vehicle module in the second adjustment scheme. Moving at this distance can avoid failure to reach the docking position due to an increased slope path or excessive movement due to a shortened path.

[0023] It should be noted that updating the original movement direction and the original movement distance in the first adjustment scheme to the final movement direction and distance respectively yields the second adjustment scheme. For example, in the first adjustment scheme of a certain first unmanned vehicle module, the movement direction is due east and the movement distance is d1, the final movement direction is 25° east of south and the final movement distance is d2. Changing the movement direction to 25° east of south and the movement distance to d2, the second adjustment scheme, obtained by replacing the movement direction and distance of the first adjustment scheme with the final movement direction and distance respectively, is the direct basis for determining the first attitude control scheme of the first docking mechanism.

[0024] Step S130: Based on the second adjustment scheme of each first unmanned vehicle module of the reconfigurable unmanned vehicle, determine the first pose control scheme of the corresponding first docking mechanism of each first unmanned vehicle module.

[0025] In some embodiments, the first posture control scheme refers to a specific posture adjustment scheme for the first docking mechanism, further refined from the second adjustment scheme of the first unmanned vehicle module. This scheme includes key parameters such as the displacement direction, displacement speed, displacement distance, and rotation angles around the X and Y axes of the first docking mechanism. For example, according to the second adjustment scheme, if a certain first docking mechanism needs to move 2 meters southeast at a moderate speed, rotate 5 degrees clockwise around the X-axis, and rotate 3 degrees counterclockwise around the Y-axis, this scheme containing specific adjustment parameters is the first posture control scheme. The adjustment scheme includes the corresponding final movement direction, final movement distance, adjustment time, rotation direction, and rotation angle for each first unmanned vehicle module.

[0026] In one possible implementation, step S130 is specifically processed as follows: Based on the final moving direction, final moving distance, rotation direction, and rotation angle of each first unmanned vehicle module, determine the initial coordinate position, final coordinate position, and attitude change information of the first docking mechanism of that first unmanned vehicle module; determine the difference between the initial and final coordinate positions of each first docking mechanism as the displacement direction and displacement distance of the corresponding first docking mechanism; calculate the displacement velocity of the corresponding first docking mechanism based on the adjustment time of each first unmanned vehicle module, the preset docking time of the corresponding first docking mechanism, and the displacement distance of the corresponding first docking mechanism; extract the rotation angle of the corresponding first docking mechanism on the X-axis and Y-axis from the attitude change information of the corresponding docking mechanism of each first unmanned vehicle module; and form the first attitude control scheme of the corresponding first docking mechanism of the first unmanned vehicle module based on the displacement direction, displacement distance, displacement velocity, and rotation angle on the X-axis and Y-axis of the corresponding first docking mechanism of each first unmanned vehicle module.

[0027] In some embodiments, the final movement direction refers to the final movement direction determined for the first unmanned vehicle module in the second adjustment scheme after working terrain correction. This direction has offset interference from terrain tilt, uneven road surface, etc., ensuring that the module can stably approach the docking target. For example, the first adjustment scheme for a certain first unmanned vehicle module is "moving towards the positive direction of the second unmanned vehicle module," but the working terrain tilts to the south. After correction, it is determined to "move slightly north of the direction of the second unmanned vehicle module" to offset the offset caused by the terrain. This corrected orientation is the final movement direction. The final movement distance refers to the total length that the first unmanned vehicle module needs to move from its current position to the docking preparation position, determined for the first unmanned vehicle module in the second adjustment scheme after working terrain correction. The rotation direction refers to the direction in which the first unmanned vehicle module rotates around the reference axis when adjusting its own attitude, determined for the second adjustment scheme. It is usually divided into clockwise and counterclockwise, with the purpose of ensuring that the first docking mechanism of the module is accurately aligned with the docking position of the second docking mechanism. The rotation angle refers to the magnitude of the rotation along the specified rotation direction, determined for the first unmanned vehicle module in the second adjustment scheme. It is used to accurately correct the module's attitude and ensure that the interface, latches, and other structures of the first docking mechanism are fully matched with the second docking mechanism. For example, in order to make the data interface of the first docking mechanism align with the interface of the second docking mechanism, the determining module needs to rotate counterclockwise by a certain amount, and this amount is the rotation angle.

[0028] It should be noted that the initial coordinate position refers to the specific coordinate point in the preset three-dimensional spatial coordinate system where the first docking mechanism of the first unmanned vehicle module is located before the start of pose adjustment. This accurately reflects the initial spatial position of the docking mechanism. The preset three-dimensional spatial coordinate system uses the second unmanned vehicle module as the reference origin. For example, with the geometric center of the second unmanned vehicle module as the origin, the (X, Y, Z) coordinate point corresponding to the first docking mechanism in this coordinate system before adjustment is the initial coordinate position. The final coordinate position refers to the target coordinate point that the first docking mechanism should reach in the three-dimensional spatial coordinate system after adjustment, determined based on the final movement direction, final movement distance, and rotation angle of the first unmanned vehicle module. This coordinate is perfectly matched with the docking position of the second docking mechanism. For example, to achieve precise latch engagement, the first docking mechanism needs to move to a specific coordinate point in the coordinate system with the second unmanned vehicle module as the origin; this point is the final coordinate position. Attitude change information refers to the set of all attitude-related change data during the process of the first docking mechanism of the first unmanned vehicle module adjusting from the initial attitude to the target attitude. This typically includes key information such as rotation direction, rotation angle, and tilt state changes. The displacement direction refers to the specific orientation of the docking mechanism in three-dimensional space, determined by calculating the difference between the initial and final coordinate positions of the first docking mechanism. It accurately reflects the path direction of the docking mechanism from its initial position to its target position. The displacement distance refers to the actual distance the docking mechanism needs to travel from its initial position to its target position, obtained by calculating the straight-line distance between the initial and final coordinate positions. For example, if the initial and final coordinate positions form a straight line in three-dimensional space, the length of this line is the displacement distance.

[0029] In some embodiments, the adjustment time refers to the total time set for the first unmanned vehicle module in the second adjustment scheme, from the start of movement and rotation adjustment to the first docking mechanism reaching its final coordinate position and completing attitude adjustment, covering the entire process of module movement and attitude correction. For example, the adjustment time is the total time required for the first unmanned vehicle module to move from its current position to the docking preparation position and complete attitude rotation. The preset docking time refers to the fixed time set in advance for the first docking mechanism to complete physical docking operations such as latching, interface connection, and sealing confirmation after reaching the final coordinate position. It is usually determined based on the mechanical movement speed and sensor response time of the docking mechanism. For example, the preset docking time is the time set in advance for the first docking mechanism to complete latching and locking based on the extension and retraction speed of the latch. The displacement speed refers to the average speed of the docking mechanism's movement calculated by "displacement distance ÷ total time" based on the displacement distance of the first docking mechanism, combined with the adjustment time of the first unmanned vehicle module and the preset docking time. It is used to control the movement speed to ensure accurate arrival at the target position within a specified time. For example, if the displacement distance is a certain length and the total time is the sum of the adjustment time and the preset docking time, the velocity value obtained by dividing the displacement distance by the total time is the displacement velocity.

[0030] Step S140: Control the first docking mechanism of each first unmanned vehicle module of the reconfigurable unmanned vehicle, adjust the pose according to the corresponding first pose control scheme, and adjust the corresponding first pose control scheme in real time based on the real-time data collected by the state acquisition device pre-set on each first docking mechanism to obtain the real-time pose control scheme of each first docking mechanism.

[0031] In some embodiments, the status acquisition device refers to a device pre-installed on each first docking mechanism for real-time acquisition of relevant data during the docking process. It includes a first distance sensor installed at the top of the docking surface of each first docking mechanism, a second distance sensor installed at the bottom of the docking surface of each first docking mechanism, a third distance sensor installed at the left end of the docking surface of each first docking mechanism, a fourth distance sensor installed at the right end of the docking surface of each first docking mechanism, and an image acquisition device. The docking surface of the first docking mechanism can be circular or other shapes. When the docking surface of the first docking mechanism is circular, such as... Figure 4As shown, distance measuring sensors are installed at the top, bottom, left, and right ends of the docking surface of the first docking mechanism, and a camera is installed in the center of the docking surface to capture the docking area. These sensors and the camera together constitute the state acquisition device. Real-time data refers to the data related to the docking state collected in real time by the state acquisition device during the posture adjustment process of the first docking mechanism, including distance data and image data. The real-time posture control scheme refers to the posture adjustment scheme obtained by dynamically adjusting the first posture control scheme based on the real-time data collected by the state acquisition device, which can adapt to the current docking state and ensure accurate docking of the first docking mechanism.

[0032] See Figure 5 The specific implementation of step S140 above includes steps S1401-S1407, the details of which are as follows: Step S1401: For each first docking mechanism, calculate the first real-time distance difference between the first docking mechanism and the corresponding second docking mechanism as collected by the first distance sensor of the first docking mechanism and the second real-time distance between the first docking mechanism and the corresponding second docking mechanism as collected by the second distance sensor.

[0033] In some embodiments, the first real-time distance is the distance between the first docking mechanism and the corresponding second docking mechanism, which is collected in real time during the docking process by a first distance sensor installed at the top of the docking surface of the first docking mechanism. The second real-time distance is the distance between the first docking mechanism and the corresponding second docking mechanism, which is collected in real time during the docking process by a second distance sensor installed at the bottom of the docking surface of the first docking mechanism. The difference between the first real-time distance and the second real-time distance is the first real-time distance difference, which is obtained by calculating the difference between the two values ​​and is used to reflect the distance difference between the top and bottom of the first docking mechanism relative to the second docking mechanism. The first real-time distance difference can be a positive or negative value.

[0034] Step S1402: For each first docking mechanism, the arctangent value of the ratio of the first real-time distance difference to the height of the docking surface of the first docking mechanism is determined as the first inclination of the bottom and top of the first docking mechanism.

[0035] In some embodiments, the length of the plane on which the first docking mechanism contacts and docks with the second docking mechanism in the vertical direction, i.e., the vertical distance from the top edge to the bottom edge of the plane, is the height of the docking surface. The first inclination is a parameter obtained by calculating the ratio of the first real-time distance difference to the height of the docking surface and then taking the arctangent of that ratio. It is used to quantify the degree of inclination between the bottom and top of the first docking mechanism, reflecting the inclination state of the docking surface in the vertical direction. For example, when the ratio of the first real-time distance difference to the height of the docking surface is a certain value, the angle obtained by taking the arctangent of that ratio is the first inclination. If the angle is positive, it means that the top of the first docking mechanism is tilted upward relative to the bottom, resulting in a greater distance between the top and the second docking mechanism.

[0036] Step S1403: For each first docking mechanism, calculate the second real-time distance difference between the third real-time distance between the first docking mechanism and the corresponding second docking mechanism collected by the third distance sensor of the first docking mechanism and the fourth real-time distance between the first docking mechanism and the corresponding second docking mechanism collected by the fourth distance sensor.

[0037] In some embodiments, the third real-time distance refers to the distance value between the first docking mechanism and the corresponding second docking mechanism, which is collected in real time by a third distance sensor installed at the left end of the docking surface of the first docking mechanism during the docking process. For example, the third distance sensor at the left end of the first docking mechanism continuously monitors the spatial interval between itself and the second docking mechanism, and the interval value recorded at a certain moment is the third real-time distance at that moment. The fourth distance sensor installed at the right end of the docking surface of the first docking mechanism collects the distance value between the first docking mechanism and the corresponding second docking mechanism in real time during the docking process, which is the fourth real-time distance. The difference between the third real-time distance and the fourth real-time distance is the second real-time distance difference, which is obtained by calculating the difference between the two values ​​and is used to reflect the distance difference between the left and right ends of the first docking mechanism relative to the second docking mechanism. For example, if the third real-time distance is C and the fourth real-time distance is D at a certain moment, then the difference between C and D is the second real-time distance difference. If the difference is negative, it means that the distance between the right end of the first docking mechanism and the second docking mechanism is greater than the distance between the left end and the right end.

[0038] Step S1404: For each first docking mechanism, the arctangent of the ratio of the second real-time distance difference to the width of the docking surface of the first docking mechanism is determined as the second inclination of the left and right ends of the first docking mechanism.

[0039] In some embodiments, the length of the plane on which the first docking mechanism contacts and docks with the second docking mechanism in the left-right direction, i.e., the horizontal distance from the left edge to the right edge of the plane, is the width of the docking surface of the first docking mechanism. The parameter obtained by calculating the ratio of the second real-time distance difference to the width of the docking surface and then taking the arctangent of this ratio is the second tilt angle. The second tilt angle is used to quantify the degree of tilt between the left and right ends of the first docking mechanism, reflecting the tilt state of the docking surface in the left-right direction. For example, when the ratio of the second real-time distance difference to the width of the docking surface is a certain value, the angle obtained by taking its arctangent is the second tilt angle; if this angle is positive, it indicates that the left end of the first docking mechanism is tilted upwards relative to the right end, resulting in a greater distance between the left end and the second docking mechanism.

[0040] Step S1405: For each first docking mechanism, determine the directional deviation of the first docking mechanism based on the real-time image acquired by the image acquisition device of the first docking mechanism.

[0041] It should be noted that the orientation deviation is determined by comparing the actual edge contour features of the first docking mechanism with the standard edge contour features in the real-time image, which is the deviation between the actual orientation of the first docking mechanism and the preset standard orientation.

[0042] In one possible implementation, step S1405 is specifically processed as follows: image feature extraction is performed on the real-time image to obtain the edge contour features of the first docking mechanism; the standard edge contour features of the first docking mechanism are determined according to the first pose control scheme; the edge contour features are compared with the standard edge contour features to obtain the directional deviation of the first docking mechanism.

[0043] In some embodiments, edge contour features refer to the shape boundary lines of the first docking mechanism as presented in a real-time image after image feature extraction. These features directly reflect the actual shape, current orientation, and spatial position of the first docking mechanism in the image, serving as the core visual basis for determining whether the docking mechanism deviates from the standard orientation. Standard edge contour features refer to the edge contour features that the docking mechanism should present in an ideal docking state, predetermined based on the preset orientation, preset position, and other preset target poses of the first docking mechanism in the first pose control scheme. These features are used for comparison with the actually extracted edge contour features. Based on the comparison results, the angle between the extracted edge contour features and the standard edge contour features represents the directional deviation of the first docking mechanism.

[0044] Step S1406: For each first docking mechanism, the absolute values ​​of the first tilt and the second tilt are compared with the first preset threshold to obtain the first comparison result, and the average value of the absolute values ​​of the first tilt and the second tilt is compared with the second threshold to obtain the second comparison result.

[0045] In some embodiments, the first preset threshold refers to a reference value set before pose adjustment to determine whether the first tilt and the second tilt are within the allowable range. If the absolute value of the tilt exceeds this value, it indicates that the tilt of the mating surface may affect the mating accuracy and adjustment is required. For example, to avoid excessive tilt of the mating surface leading to mating failure, an upper limit of the tilt angle is set in advance as the first preset threshold. When the absolute value of the first tilt exceeds this threshold, the attitude of the first mating mechanism needs to be corrected. The first comparison result refers to the judgment result obtained after comparing the absolute values ​​of the first tilt and the second tilt with the first preset threshold, respectively, regarding whether both exceed the allowable range. The second threshold is a reference value set before pose adjustment to determine whether the average of the absolute values ​​of the first tilt and the second tilt is within the allowable range. Unlike the first preset threshold, it focuses more on the overall average tilt of the mating surface. The second comparison result refers to the judgment result obtained after comparing the average of the absolute values ​​of the first tilt and the second tilt with the second threshold, regarding whether the average exceeds the allowable range.

[0046] Step S1407: For each first docking mechanism, the first position control scheme of the first docking mechanism is adjusted in real time according to the first comparison result, the second comparison result and the direction deviation, so as to obtain the real-time position control scheme of the first docking mechanism.

[0047] In some embodiments, real-time adjustment of the first posture scheme is to prevent deviations in the position and attitude adjustment of the first unmanned vehicle module due to control errors or bumps during movement. Therefore, real-time adjustment of the first posture control scheme is necessary to obtain a real-time posture control scheme, ensuring that the first docking mechanism can successfully complete docking. The first posture control scheme includes the displacement direction, displacement speed, displacement distance, and rotation angles of the first docking mechanism on the X and Y axes.

[0048] In one possible implementation, step S1407 specifically includes: if, in the first comparison result, the first tilt angle or the second tilt angle exceeds a first preset threshold, then the difference between the first tilt angle and the preset allowance is determined as the deviation rotation angle of the first docking mechanism on the X-axis, and the difference between the second tilt angle and the preset allowance is determined as the deviation rotation angle of the first docking mechanism on the Y-axis; if, in the second comparison result, the average of the absolute values ​​of the first tilt angle and the second tilt angle exceeds a second threshold, then the sine value of the average value is calculated and the longest side length or radius of the docking surface of the docking mechanism is calculated. The product of the distance deviation and the displacement distance of the first docking mechanism on the Z-axis in the first attitude control scheme is determined as the velocity deviation percentage of the first docking mechanism. Based on the directional deviation, the deviation rotation angle on the X-axis, the deviation rotation angle on the Y-axis, and the velocity deviation percentage of the first docking mechanism, the displacement direction, rotation angle on the X-axis, rotation angle on the Y-axis, and displacement velocity of the corresponding first attitude control scheme are adjusted to obtain the real-time attitude control scheme of the first docking mechanism.

[0049] In some embodiments, the preset margin refers to a safety compensation amount pre-set before calculating the deviation rotation angle to offset possible errors or fluctuations during tilt adjustment. Its purpose is to avoid overcorrection or docking deviation caused by direct tilt adjustment. When determining the pose of the first docking mechanism, a coordinate system needs to be set for the docking surface of each first docking mechanism. The coordinate system of the first docking surface adopts a right-handed coordinate system, with the origin located at the center of the docking surface of the first docking mechanism. The positive direction of the Z-axis is the movement direction of the first unmanned vehicle module corresponding to the first docking mechanism. The deviation rotation angle on the Y-axis refers to the angle parameter used to adjust the rotation of the first docking mechanism around the Y-axis, obtained by calculating the difference between the second tilt and the preset margin when the second tilt exceeds the first preset threshold in the first comparison result. Its function is to correct the tilt deviation between the left and right ends of the first docking mechanism. The sine value of the average value refers to the value obtained by taking the sine function of the average of the absolute values ​​of the first tilt and the second tilt. This value is used to calculate the actual position deviation in conjunction with the docking surface dimensions. The average of the absolute values ​​of the first tilt and the second tilt is the average angle used to reflect the overall tilt degree of the docking surface. If the docking surface of the first docking mechanism is a polygon, the longest side among all the sides of the polygon is the longest side length of the docking surface. If the docking surface is an irregular polygon, the longest straight-line distance that covers its edge is taken. If the docking surface of the first docking mechanism is circular or near-circular, the maximum distance from the geometric center of the docking surface to its edge is taken as the radius. When the average tilt in the second comparison result exceeds the second threshold, the deviation length obtained by multiplying the sine of the average value by the longest side length or radius of the docking surface is the distance deviation value, which is used to determine the magnitude of adjustment required for the displacement distance.

[0050] It should be noted that the displacement distance on the Z-axis refers to the total length of the first docking mechanism's movement along the Z-axis, pre-set in the first attitude control scheme. This is a core parameter for adjusting the height or vertical position of the docking mechanism during docking. The velocity deviation percentage, calculated as the ratio of the distance deviation to the displacement distance on the Z-axis, reflects the required adjustment ratio of the first docking mechanism's displacement velocity. Its function is to dynamically correct the displacement velocity based on the position deviation, ensuring docking accuracy. The second attitude control scheme refers to the final attitude control scheme that adapts to the actual docking state and ensures accurate docking, obtained by specifically adjusting the displacement direction, X-axis rotation angle, Y-axis rotation angle, and displacement velocity in the first attitude control scheme based on the first docking mechanism's directional deviation, X-axis deviation rotation angle, Y-axis deviation rotation angle, and velocity deviation percentage. For example, if the displacement direction is corrected based on the directional deviation, the rotation angle is adjusted based on the X / Y-axis deviation rotation angle, and the displacement velocity is optimized based on the velocity deviation percentage, the resulting scheme, directly usable for docking execution, is the second attitude control scheme.

[0051] By inputting the initial pose, target pose, and application scenario of the reconfigurable autonomous vehicle (RV) into a deep learning model trained using historical pose change data, a first adjustment scheme for each first RV module is obtained, reducing the dependence of the pose control method on the mathematical model. Each first adjustment scheme is then modified using the RV's working terrain to obtain a second adjustment scheme for each first RV module, thereby determining the first pose control scheme for each first docking mechanism. This allows for consideration of the impact of complex ground environments during pose control, improving docking accuracy. Real-time adjustments to each first pose control scheme are made in real-time using pre-set distance sensors and image acquisition devices on the first docking mechanism, resulting in a real-time pose control scheme for each first docking mechanism. This ensures robustness of pose control under complex dynamic disturbances, improves the accuracy of the pose control scheme, and guarantees successful docking of each RV module.

[0052] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0053] The following are device embodiments of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0054] Figure 6 The diagram shows a schematic representation of the pose control device for a reconfigurable unmanned vehicle docking mechanism according to an embodiment of the present invention. For ease of explanation, only the parts relevant to the embodiment of the present invention are shown, and are described in detail below: like Figure 6 As shown, the pose control device 6 of the reconfigurable unmanned vehicle docking mechanism includes: Learning module 61 is used to input the initial pose, target pose, and application scenario of the reconfigurable autonomous vehicle into a pre-trained deep learning model to obtain a first adjustment scheme for each first autonomous vehicle module of the reconfigurable autonomous vehicle; wherein, the reconfigurable autonomous vehicle includes multiple first autonomous vehicle modules and one second autonomous vehicle module; each first autonomous vehicle module includes a first docking mechanism; the second autonomous vehicle module includes multiple second docking mechanisms; The correction module 62 is used to correct each first adjustment scheme according to the working terrain of the reconfigurable unmanned vehicle to obtain the corresponding second adjustment scheme of the first unmanned vehicle module. Module 63 is used to determine the first pose control scheme of the first docking mechanism corresponding to each first unmanned vehicle module based on the adjustment scheme of each first unmanned vehicle module of the reconfigurable unmanned vehicle; The adjustment module 64 is used to control the corresponding first docking mechanism of each first unmanned vehicle module of the reconfigurable unmanned vehicle, adjust the posture according to the corresponding first posture control scheme, and adjust the corresponding first posture control scheme in real time based on the real-time data collected by the state acquisition device pre-set on each first docking mechanism to obtain the real-time posture control scheme of each first docking mechanism.

[0055] In one possible implementation, the correction module 62 specifically includes: determining the tilt angle as the basic position deviation angle of the second unmanned vehicle module, and determining the tilt direction as the basic position deviation direction of the second unmanned vehicle module; and correcting each first adjustment scheme based on the basic position deviation angle and the basic position deviation direction to obtain the corresponding second adjustment scheme of the first unmanned vehicle module.

[0056] In one possible implementation, the correction module 62 further includes: superimposing the basic position deviation direction with the movement direction in the corresponding first adjustment scheme of each first unmanned vehicle module to obtain the final movement direction in the corresponding second adjustment scheme; calculating the final movement distance of the corresponding second adjustment scheme of each first unmanned vehicle module based on the basic position deviation angle and the movement distance in the first adjustment scheme of each first unmanned vehicle module; replacing the movement direction in each first adjustment scheme with the corresponding final movement direction, and replacing the movement distance in each first adjustment scheme with the corresponding final movement distance to obtain the corresponding second adjustment scheme.

[0057] In one possible implementation, the determining module 63 specifically includes: determining the initial coordinate position, final coordinate position, and attitude change information of the first docking mechanism of each first unmanned vehicle module based on the final movement direction, final movement distance, rotation direction, and rotation angle; determining the difference between the initial coordinate position and the final coordinate position of each first docking mechanism as the displacement direction and displacement distance of the corresponding first docking mechanism; calculating the displacement velocity of the corresponding first docking mechanism based on the adjustment time of each first unmanned vehicle module, the preset docking time of the corresponding first docking mechanism, and the displacement distance of the corresponding first docking mechanism; extracting the rotation angle of the corresponding first docking mechanism on the X-axis and Y-axis from the attitude change information of the corresponding docking mechanism of each first unmanned vehicle module; and forming the first attitude control scheme of the corresponding first docking mechanism of the first unmanned vehicle module based on the displacement direction, displacement distance, displacement velocity, and rotation angle on the X-axis and Y-axis of the corresponding first docking mechanism of each first unmanned vehicle module.

[0058] In one possible implementation, the adjustment module 64 specifically includes: for each first docking mechanism, performing the following steps: calculating the first real-time distance difference between the first docking mechanism and the corresponding second docking mechanism as collected by the first distance sensor of the first docking mechanism and the second real-time distance between the first docking mechanism and the corresponding second docking mechanism as collected by the second distance sensor; determining the arctangent of the ratio of the first real-time distance difference to the height of the docking surface of the first docking mechanism as the first inclination of the bottom and top of the first docking mechanism; calculating the third real-time distance between the first docking mechanism and the corresponding second docking mechanism as collected by the third distance sensor of the first docking mechanism and the third real-time distance between the first docking mechanism and the corresponding second docking mechanism as collected by the fourth distance sensor. The second real-time distance difference between the fourth real-time distances of the docking mechanism; the arctangent of the ratio of the second real-time distance difference to the width of the docking surface of the first docking mechanism is determined as the second tilt of the left and right ends of the first docking mechanism; the directional deviation of the first docking mechanism is determined based on the real-time image acquired by the image acquisition device of the first docking mechanism; the absolute values ​​of the first tilt and the second tilt are compared with a first preset threshold to obtain a first comparison result, and the average of the absolute values ​​of the first tilt and the second tilt is compared with a second threshold to obtain a second comparison result; based on the first comparison result, the second comparison result, and the directional deviation, the first pose control scheme of the first docking mechanism is adjusted in real time to obtain the real-time pose control scheme of the first docking mechanism.

[0059] In one possible implementation, the adjustment module 64 further includes: if, in the first comparison result, the first tilt angle or the second tilt angle exceeds a first preset threshold, then the difference between the first tilt angle and the preset margin is determined as the deviation rotation angle of the first docking mechanism on the X-axis, and the difference between the second tilt angle and the preset margin is determined as the deviation rotation angle of the first docking mechanism on the Y-axis; if, in the second comparison result, the average of the absolute values ​​of the first tilt angle and the second tilt angle exceeds a second threshold, then the sine value of the average value is calculated and the product of the longest side length or radius of the docking surface of the docking mechanism is determined, and the product is determined as the distance deviation value of the first docking mechanism; the ratio of the distance deviation value to the displacement distance of the first docking mechanism on the Z-axis in the first posture control scheme is determined as the speed deviation percentage of the first docking mechanism; based on the direction deviation of the first docking mechanism, the deviation rotation angle on the X-axis, the deviation rotation angle on the Y-axis, and the speed deviation percentage, the displacement direction, rotation angle on the X-axis, rotation angle on the Y-axis, and displacement speed of the corresponding first posture control scheme are adjusted respectively to obtain the real-time posture control scheme of the first docking mechanism.

[0060] In one possible implementation, the adjustment module 64 further includes: extracting image features from the real-time image to obtain the edge contour features of the first docking mechanism; determining the standard edge contour features of the first docking mechanism according to the first pose control scheme; and comparing the edge contour features with the standard edge contour features to obtain the directional deviation of the first docking mechanism.

[0061] In one possible implementation, the pre-trained deep learning model is trained based on the historical pose adjustment data of the reconfigurable autonomous vehicle.

[0062] Figure 7 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. For example... Figure 7 As shown, the electronic device 7 of this embodiment includes a processor 70 and a memory 71. The memory 71 stores a computer program 72. When the processor 70 executes the computer program 72, it implements the steps in the various method embodiments described above. Alternatively, when the processor 70 executes the computer program 72, it implements the functions of each module / unit in the various device embodiments described above.

[0063] For example, computer program 72 may be divided into one or more modules / units, which are stored in memory 71 and executed by processor 70 to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of computer program 72 in electronic device 7.

[0064] Electronic device 7 may include, but is not limited to, processor 70 and memory 71. Those skilled in the art will understand that... Figure 7 This is merely an example of electronic device 7 and does not constitute a limitation on electronic device 7. It may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device 7 may also include input / output devices, network access devices, buses, etc.

[0065] For the sake of simplicity and clarity, only the above-described functional modules / units are used as examples. In practical applications, the functions described above can be assigned to different functional modules / units as needed. These modules / units can be implemented in hardware, software, or a combination of both.

[0066] In the above embodiments, the descriptions of each embodiment have their own emphasis. Parts not detailed or described in a particular embodiment can be referred to in the relevant descriptions of other embodiments. Unless otherwise specified or in conflict with logic, the terminology and / or descriptions between different embodiments are consistent and can be referenced interchangeably. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0067] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A pose control method for a reconfigurable unmanned vehicle docking mechanism, characterized in that, include: The initial pose, target pose, and application scenario of the reconfigurable autonomous vehicle are input into a pre-trained deep learning model to obtain a first adjustment scheme for each first autonomous vehicle module of the reconfigurable autonomous vehicle; wherein, the reconfigurable autonomous vehicle includes multiple first autonomous vehicle modules and one second autonomous vehicle module; each first autonomous vehicle module includes a first docking mechanism; the second autonomous vehicle module includes multiple second docking mechanisms; Each first adjustment scheme is modified according to the working terrain of the reconfigurable unmanned vehicle to obtain the corresponding second adjustment scheme of the first unmanned vehicle module; Based on the second adjustment scheme of each first unmanned vehicle module of the reconfigurable unmanned vehicle, the first pose control scheme of the corresponding first docking mechanism of each first unmanned vehicle module is determined; The first docking mechanism of each first unmanned vehicle module of the reconfigurable unmanned vehicle is controlled to adjust its posture according to the corresponding first posture control scheme. Based on the real-time data collected by the state acquisition device pre-set on each first docking mechanism, the corresponding first posture control scheme is adjusted in real time to obtain the real-time posture control scheme of each first docking mechanism.

2. The pose control method for the reconfigurable unmanned vehicle docking mechanism according to claim 1, characterized in that, The status acquisition device includes a first distance sensor installed at the top of the docking surface of each first docking mechanism, a second distance sensor installed at the bottom of the docking surface of each first docking mechanism, a third distance sensor installed at the left end of the docking surface of each first docking mechanism, a fourth distance sensor installed at the right end of the docking surface of each first docking mechanism, and an image acquisition device. The real-time position and posture control scheme for each first docking mechanism is obtained by adjusting the corresponding first posture control scheme in real time based on the real-time data collected by the pre-set state acquisition device on each first docking mechanism, including: For each first docking institution, perform the following steps: Calculate the first real-time distance difference between the first docking mechanism and the corresponding second docking mechanism, as collected by the first distance sensor of the first docking mechanism and the second real-time distance between the first docking mechanism and the corresponding second docking mechanism, and the second real-time distance between the first docking mechanism and the corresponding second docking mechanism, as collected by the second distance sensor. The arctangent of the ratio of the first real-time distance difference to the height of the docking surface of the first docking mechanism is determined as the first inclination of the bottom and top of the first docking mechanism. Calculate the second real-time distance difference between the third real-time distance between the first docking mechanism and the corresponding second docking mechanism, which is collected by the third distance sensor of the first docking mechanism, and the fourth real-time distance between the first docking mechanism and the corresponding second docking mechanism, which is collected by the fourth distance sensor; The arctangent of the ratio of the second real-time distance difference to the width of the docking surface of the first docking mechanism is determined as the second inclination of the left and right ends of the first docking mechanism. Based on the real-time images acquired by the image acquisition device of the first docking mechanism, the directional deviation of the first docking mechanism is determined; The absolute values ​​of the first tilt and the second tilt are compared with a first preset threshold to obtain a first comparison result. The average value of the absolute values ​​of the first tilt and the second tilt is compared with a second threshold to obtain a second comparison result. Based on the first comparison result, the second comparison result, and the directional deviation, the first position control scheme of the first docking mechanism is adjusted in real time to obtain the real-time position control scheme of the first docking mechanism.

3. The pose control method for the reconfigurable unmanned vehicle docking mechanism according to claim 2, characterized in that, The first posture control scheme includes the displacement direction, displacement speed, displacement distance, and rotation angle of the first docking mechanism on the X-axis and Y-axis; The step of adjusting the first pose control scheme of the first docking mechanism in real time based on the first comparison result, the second comparison result, and the direction deviation to obtain the real-time pose control scheme of the first docking mechanism includes: If, in the first comparison result, the first tilt or the second tilt exceeds the first preset threshold, then the difference between the first tilt and the preset margin is determined as the deviation rotation angle of the first docking mechanism on the X-axis, and the difference between the second tilt and the preset margin is determined as the deviation rotation angle of the first docking mechanism on the Y-axis. If the average of the absolute values ​​of the first tilt and the second tilt in the second comparison result exceeds the second threshold, then the sine of the average value is calculated and the product of the longest side length or radius of the docking surface of the docking mechanism is determined as the distance deviation value of the first docking mechanism. The ratio of the distance deviation value to the displacement distance of the first docking mechanism on the Z-axis in the first pose control scheme is determined as the speed deviation percentage of the first docking mechanism. Based on the directional deviation of the first docking mechanism, the deviation rotation angle on the X-axis, the deviation rotation angle on the Y-axis, and the percentage of velocity deviation, the displacement direction, rotation angle on the X-axis, rotation angle on the Y-axis, and displacement velocity of the corresponding first posture control scheme are adjusted to obtain the real-time posture control scheme of the first docking mechanism.

4. The pose control method for the reconfigurable unmanned vehicle docking mechanism according to claim 2, characterized in that, The determination of the directional deviation of the first docking mechanism based on the real-time image acquired by the image acquisition device based on the first docking mechanism includes: Image feature extraction is performed on the real-time image to obtain the edge contour features of the first docking mechanism; Based on the first pose control scheme, the standard edge contour features of the first docking mechanism are determined; The directional deviation of the first docking mechanism is obtained by comparing the edge contour feature with the standard edge contour feature.

5. The pose control method for the reconfigurable unmanned vehicle docking mechanism according to claim 3, characterized in that, The adjustment scheme includes the final movement direction, final movement distance, adjustment time, rotation direction, and rotation angle for each first unmanned vehicle module; based on the adjustment scheme of each first unmanned vehicle module of the reconfigurable unmanned vehicle, the first pose control scheme of the corresponding first docking mechanism for each first unmanned vehicle module is determined, including: Based on the final moving direction, final moving distance, rotation direction and rotation angle of each first unmanned vehicle module, determine the initial coordinate position, final coordinate position and attitude change information of the first docking mechanism of the first unmanned vehicle module; The difference between the initial and final coordinate positions of each first docking mechanism is determined as the displacement direction and displacement distance of the corresponding first docking mechanism. Based on the adjustment time of each first unmanned vehicle module, the preset docking time of the corresponding first docking mechanism, and the displacement distance of the corresponding first docking mechanism, calculate the displacement speed of the corresponding first docking mechanism; From the attitude change information of the docking mechanism of each first unmanned vehicle module, the rotation angle of the corresponding first docking mechanism on the X and Y axes is extracted; The displacement direction, displacement distance, displacement speed, and rotation angle on the X and Y axes of the corresponding first docking mechanism of each first unmanned vehicle module constitute the first posture control scheme of the corresponding first docking mechanism of the first unmanned vehicle module.

6. The pose control method for the reconfigurable unmanned vehicle docking mechanism according to claim 1, characterized in that, The working terrain information includes the tilt angle and tilt direction; the step of correcting each first adjustment scheme based on the working terrain information of the reconfigurable unmanned vehicle to obtain a corresponding second adjustment scheme for the first unmanned vehicle module includes: The tilt angle is determined as the basic position deviation angle of the second unmanned vehicle module, and the tilt direction is determined as the basic position deviation direction of the second unmanned vehicle module; Based on the basic position deviation angle and the basic position deviation direction, each first adjustment scheme is corrected to obtain the corresponding second adjustment scheme for the first unmanned vehicle module.

7. The pose control method for the reconfigurable unmanned vehicle docking mechanism according to claim 6, characterized in that, The first adjustment scheme includes the corresponding movement direction, movement distance, adjustment time, rotation direction, and rotation angle for each first unmanned vehicle module; the second adjustment scheme for each first adjustment scheme is obtained by correcting the basic position deviation angle and the basic position deviation direction based on the basic position deviation angle and the basic position deviation direction, including: The basic position deviation direction is superimposed with the movement direction in the corresponding first adjustment scheme of each first unmanned vehicle module to obtain the final movement direction in the corresponding second adjustment scheme. Based on the basic position deviation angle and the movement distance in the first adjustment scheme of each first unmanned vehicle module, the final movement distance in the corresponding second adjustment scheme of each first unmanned vehicle module is calculated. Replace the movement direction in each first adjustment scheme with the corresponding final movement direction, and replace the movement distance in each first adjustment scheme with the corresponding final movement distance to obtain the corresponding second adjustment scheme.

8. The pose control method for the reconfigurable unmanned vehicle docking mechanism according to claim 1, characterized in that, The pre-trained deep learning model is trained based on the historical pose adjustment data of the reconfigurable unmanned vehicle.

9. A pose control device for a reconfigurable unmanned vehicle docking mechanism, characterized in that, include: The learning module is used to input the initial pose, target pose, and application scenario of the reconfigurable autonomous vehicle into a pre-trained deep learning model to obtain a first adjustment scheme for each first autonomous vehicle module of the reconfigurable autonomous vehicle; wherein, the reconfigurable autonomous vehicle includes multiple first autonomous vehicle modules and one second autonomous vehicle module; each first autonomous vehicle module includes a first docking mechanism; the second autonomous vehicle module includes multiple second docking mechanisms; The correction module is used to correct each first adjustment scheme according to the working terrain of the reconfigurable unmanned vehicle, so as to obtain the corresponding second adjustment scheme of the first unmanned vehicle module; The determination module is used to determine the first pose control scheme of the corresponding first docking mechanism of each first unmanned vehicle module based on the adjustment scheme of each first unmanned vehicle module of the reconfigurable unmanned vehicle; The adjustment module is used to control the corresponding first docking mechanism of each first unmanned vehicle module of the reconfigurable unmanned vehicle, adjust the posture according to the corresponding first posture control scheme, and adjust the corresponding first posture control scheme in real time based on the real-time data collected by the state acquisition device pre-set on each first docking mechanism to obtain the real-time posture control scheme of each first docking mechanism.

10. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method as described in any one of claims 1 to 8.

Images