A reconfigurable adaptive multi-purpose modular unmanned aerial vehicle system

Abstract

The present application relates to the technical field of modular unmanned aerial vehicle, in particular to a reconfigurable adaptive multi-purpose modular unmanned aerial vehicle system, the system comprises: configuration identification module, semantic mapping module, channel conversion module, control quantity distribution module and output driving module.In the present application, the flight configuration is automatically identified and numbered by collecting the pin level and comparing the configuration determination table, the assembly state is linked with the control link, the remote control channel input is combined with the direction response configuration to complete the semantic mapping, the control logic under different configurations is ensured to be consistent, the remote control amplitude is normalized and proportionally adjusted, the standard instruction required by the current configuration is adapted, the error interference is avoided, the control quantity is distributed to the execution interface according to the configuration index, the response efficiency and control accuracy are improved, the output signal is subjected to feedback detection for deviation evaluation, the overall process is integrated into a closed loop of identification, self-adaption and verification, and the multi-configuration adaptability and system stability are enhanced.The embodiment realizes the rapid reconfigurable conversion of the "fixed wing - multi-rotor - compound wing" configuration on the basis of the same unmanned aerial vehicle platform.

Description

Technical Field

[0001] This invention relates to the field of modular unmanned aerial vehicle (UAV) technology, and in particular to a reconfigurable adaptive multi-purpose modular UAV system. Background Technology

[0002] Modular UAV technology refers to a collection of technologies focused on the design and application of UAV platforms in terms of structural configuration combinations and system adaptation. Its core aspects include the selection and conversion of UAV airframe configurations, the assembly relationship between flight components and airframe components, takeoff and landing methods and control configurations under different flight modes, and the adaptation methods of flight control and remote control systems in different configurations. This technology systematically designs the UAV rotor unit, wing assembly, power unit, and connection structure, enabling the UAV to combine and use various configurations within a given structural framework. In contrast, traditional UAV systems refer to aircraft systems designed based on a single aerodynamic layout, with their configuration determined at the factory. These systems focus on flight time controllability and takeoff and landing... Technical issues such as configuration are addressed using either multi-rotor, fixed-wing, or VTOL compound configurations. Flight control and operation are achieved by configuring corresponding airframe structures, rotor arrangements, wing forms, flight control parameters, and remote control equipment for different configurations. The reconfigurable adaptive multi-purpose modular UAV system addresses the problem that a single hardware package cannot fully cover multiple aerodynamic layouts and that it is difficult to share a unified flight control and remote control system between different configurations. It adopts a combination of detachable rotor components, fixed-wing components, and fuselage connection structures. Based on the same flight control hardware and remote control hardware, flight control configurations are preset for different configurations, thus forming a UAV system that can be reconfigured between multi-rotor fixed-wing and VTOL compound configurations.

[0003] Traditional UAV systems rely on fixed designs for configuration recognition, lacking the ability to instantly identify the actual assembly state. This makes it difficult to quickly adjust the control system after structural changes, easily leading to recognition lag or incorrect flight control parameter adaptation. At the control logic level, the mapping between remote control signal channels and flight control targets is a fixed configuration, unable to be dynamically adjusted according to current configuration changes. This forces multi-configuration aircraft to have separate control modes for different forms, increasing operational burden and error probability. In remote control input processing, different configurations have different requirements for remote control signal amplitude. Existing technologies mostly use preset standards, failing to achieve adaptive adjustment for different control needs, affecting control accuracy. On the control output side, the binding relationship between execution components and output channels is rigid, lacking a dynamic allocation mechanism, which is not conducive to the rapid reconstruction and accurate issuance of control commands across multiple configurations. Taking a vertical takeoff and landing (VTOL) compound wing as an example, the system structure combines rotor and fixed-wing components. If control quantities cannot be dynamically allocated, problems such as control response deviation or control failure will occur. The lack of a feedback verification mechanism in the output stage makes it impossible to effectively judge whether the control signal is accurately delivered, which can easily lead to risks such as decreased flight stability and limit the system's adaptability in complex environments. Summary of the Invention

[0004] To address the technical problems existing in the prior art, embodiments of the present invention provide a reconfigurable, adaptive, multi-purpose modular unmanned aerial vehicle (UAV) system. The technical solution is as follows:

[0005] On the one hand, a reconfigurable, adaptive, multi-purpose modular unmanned aerial vehicle system is provided, comprising:

[0006] The configuration recognition module reads the splicing detection pin level, collects the configuration recognition signal, compares it with the configuration judgment table, identifies whether it includes fixed-wing, vertical take-off and landing or multi-rotor components, verifies that the configuration is consistent with the flight control number, extracts the configuration index, and generates a reconfigurable configuration recognition result.

[0007] The semantic mapping module calls the channel mapping table based on the reconfigurable configuration recognition result, collects remote control channel input, identifies the control channel, determines the semantic direction through the direction response configuration, reconstructs the channel and control target mapping, and generates a configuration-consistent semantic instruction set.

[0008] The channel conversion module reads flight control parameters based on the configuration-consistent semantic instruction set, performs amplitude conversion and normalization on the remote control input, adjusts the output in combination with the proportional coefficient, and generates a configuration-matched remote control amplitude set.

[0009] Based on the configuration matching remote control amplitude set, the control quantity allocation module calls the PWM channel configuration table, filters the output channel matching the configuration, allocates the control quantity to the rotor, control surface or thrust device, establishes the mapping relationship between control and execution components, and generates a reconfiguration configuration control output mapping table.

[0010] The output drive module calls the reconfiguration configuration control output mapping table, configures the PWM channel frequency and pulse width, outputs control signals, detects the deviation between the feedback and the set value, makes a state determination based on the deviation, and generates a multi-configuration output signal drive state set.

[0011] As a further aspect of the present invention, the reconfigurable configuration identification result includes the fixed-wing component inclusion state, the vertical take-off and landing unit inclusion state, and the multi-rotor component inclusion state; the configuration-consistent semantic instruction set includes control channel semantic tags, control target semantic identifiers, and direction semantic pointers; the configuration-matching remote control amplitude set includes normalized control amplitude, median reference offset, and proportional coefficient correction amplitude; the reconfigurable configuration control output mapping table includes a PWM output channel number set, an execution component interface identifier set, and a control quantity type allocation item; and the multi-configuration output signal drive state set includes channel frequency setting state, pulse width range output state, amplitude deviation judgment item, and control satisfaction judgment result.

[0012] As a further aspect of the present invention, the configuration recognition module includes:

[0013] The signal acquisition submodule acquires the level status of the splicing detection pin, acquires the signal groups output by the configuration recognition component, records the information of the signal and the corresponding pin number, determines whether the signal duration meets the recognition time requirement, selects the signal groups that meet the conditions as the basis for analysis, and generates a signal combination sequence.

[0014] The state comparison submodule, based on the signal combination sequence, calls the standard configuration state in the configuration index judgment table, compares the level of the signal with the corresponding level of the standard state, determines whether the matching ratio reaches the recognition and comparison threshold, filters the configuration state numbers that meet the conditions, and obtains the matching configuration number group.

[0015] The index generation submodule calls the configuration number of the flight control system according to the matching configuration number group, determines whether the number is consistent with the current configuration number, and if there is no consistent item, it filters the number with the highest matching ratio and verifies the number information to obtain the reconfigurable configuration recognition result.

[0016] As a further aspect of the present invention, the process of determining whether the signal duration meets the recognition time requirement is specifically limited to continuously sampling the level state corresponding to the same pin number, and the number of samplings that continuously maintain the same level state is not less than a preset sampling number threshold, and the corresponding time length is not less than a preset recognition time parameter, in order to determine that the signal duration meets the recognition condition. The process of determining whether the matching ratio reaches the identification and comparison threshold is specifically defined as follows: taking the number of effective level bits participating in the comparison in the signal combination sequence as the comparison base, counting the number of level bits that are completely consistent with the standard configuration state in the configuration index determination table, and calculating the ratio of the number of consistent level bits to the comparison base. When the ratio is not less than the preset identification and comparison threshold, the corresponding configuration state number is determined as a valid matching number. The process of selecting the number with the highest matching ratio and verifying the number information is specifically limited to comparing the matching ratio values ​​corresponding to each configuration state number in the matching configuration number group, selecting the configuration state number with the largest matching ratio value, and verifying the consistency between the configuration identification component number information corresponding to the configuration state number and the configuration number information stored in the flight control system. After the verification is passed, it is taken as the reconfigurable configuration identification result.

[0017] As a further aspect of the present invention, the semantic mapping module includes:

[0018] Based on the reconfigurable configuration recognition result, the amplitude acquisition submodule calls the channel control mapping table to acquire the real-time input amplitude of the remote control channel, identifies the channel number corresponding to the input amplitude, records the correspondence between the channel and the input amplitude, eliminates signal loss channels and interference input channels, filters the data of signal stable channels, and generates a channel input amplitude sequence.

[0019] The direction discrimination submodule calls the direction response configuration parameters based on the channel input amplitude sequence, distinguishes and judges the channel input amplitude and direction response threshold, marks the response direction attribute of the channel under the current configuration, filters the channel data with clear direction attributes in the marking results, and generates a direction pointing identification sequence.

[0020] The instruction mapping output submodule calls the direction pointing identifier sequence, combines it with the channel mapping control rules under the current configuration, establishes the correspondence structure between the direction attribute and the target control channel, filters the combination relationship with consistent channel target pointing, constructs the consistent mapping relationship between channel instructions and target control channels, and generates a configuration-consistent semantic instruction set.

[0021] As a further aspect of the present invention, the channel conversion module includes:

[0022] The interval reading submodule reads the input signal amplitude range and median reference value from the flight control parameters based on the configuration-consistent semantic instruction set, marks the minimum amplitude, maximum amplitude and median amplitude of the corresponding channel, filters the channel records with complete amplitude range information, and generates the channel amplitude benchmark dataset.

[0023] The amplitude adjustment submodule calls the channel input values ​​from the configuration-consistent semantic instruction set based on the channel amplitude reference dataset, performs amplitude conversion and normalization on the input values, performs amplitude range conversion in combination with the proportional adjustment coefficient, filters the channel data whose amplitude falls within the preset range in the processing results, and generates a channel conversion amplitude sequence.

[0024] The standard generation submodule calls the channel conversion amplitude sequence to map the standard amplitude of the command required for the current configuration, filters the conversion channel amplitude that meets the preset mapping requirements, constructs the correspondence between channel commands and standard amplitudes, and generates a configuration-matched remote control amplitude set.

[0025] As a further aspect of the present invention, the process of reading the input signal amplitude range and median reference value in the flight control parameters is specifically defined as follows: reading the minimum amplitude parameter, maximum amplitude parameter, and median reference value parameter of the input signal corresponding to each channel from the flight control parameter storage area, and establishing a correspondence between the minimum amplitude parameter, maximum amplitude parameter, and median reference value parameter of the input signal according to the channel number, as the minimum amplitude, maximum amplitude, and median amplitude corresponding to the channel; The process of amplitude conversion and normalization of the input value is specifically defined as follows: based on the minimum amplitude, maximum amplitude and median amplitude, the channel input value in the configuration consistent semantic instruction set is linearly mapped to a uniform normalized amplitude range, and the amplitude range is converted in combination with a preset proportional adjustment coefficient within the normalized amplitude range. The process of filtering channel data whose amplitude falls within a preset range is specifically defined as follows: mapping the normalized conversion result after proportional adjustment to the corresponding physical amplitude, comparing the physical amplitude with the minimum amplitude and the maximum amplitude, retaining only the channel data whose mapped physical amplitude is between the minimum amplitude and the maximum amplitude, and forming the channel conversion amplitude sequence according to the channel number order.

[0026] As a further aspect of the present invention, the control quantity allocation module includes:

[0027] The channel filtering submodule, based on the configuration matching remote control amplitude set, calls the PWM output channel configuration table, extracts all output channel information corresponding to the current configuration index, compares the enable status of the channel with the channel requirements of the configuration index, filters the channel data with matching status, and generates a configuration output channel set.

[0028] The control pointing submodule outputs the channel set according to the configuration, reads the function type information required for the control quantity, compares the channel data with the function categories of rotor control, control surface execution, and thrust vector interface, establishes the allocation correspondence of the control quantity in the function category, filters the allocation structure control quantity records, and generates a function control quantity lookup table.

[0029] The output channel mapping submodule calls the function control variable lookup table, extracts the target output channel in the function category, establishes the mapping relationship between the output port and the physical execution unit, eliminates records with redundant bindings or undefined channels, organizes the channel items with complete mapping structure, and generates a reconfiguration configuration control output mapping table.

[0030] As a further aspect of the present invention, the output driving module includes:

[0031] The channel configuration submodule calls the reconfiguration configuration control output mapping table, configures the PWM signal output channel, sets the output frequency range and pulse width interval parameters corresponding to the channel, records the correspondence between the channel number and the parameter configuration, filters the channel records with complete configuration information, and establishes the output channel parameter set.

[0032] Based on the output channel parameter set, the signal transmission submodule loads the control signal to the corresponding output channel, sends the PWM signal to the target port, monitors the real-time output status of the channel during the transmission process, collects the signal amplitude data returned by the port, and forms an output signal feedback set.

[0033] The state determination submodule compares the channel set amplitude range with the feedback signal amplitude based on the output signal feedback set, calculates the amplitude deviation, determines whether the channel deviation is within the deviation threshold range, summarizes the channel determination results, and generates a multi-configuration output signal drive state set.

[0034] As a further aspect of the present invention, the process of setting the output frequency range and pulse width interval parameters corresponding to the channel is specifically limited to configuring the minimum output frequency parameter, the maximum output frequency parameter, the minimum pulse width parameter, and the maximum pulse width parameter for each channel according to the channel type identifier pre-stored in the reconfiguration control output mapping table, and establishing a correspondence between the minimum output frequency parameter, the maximum output frequency parameter, the minimum pulse width parameter, and the maximum pulse width parameter and the channel number to form the output channel parameter set; The process of monitoring the real-time output status of the channel during transmission and collecting the signal amplitude data returned by the port is specifically defined as follows: during the PWM signal transmission period, the actual pulse width value of each channel output port is sampled multiple times according to a fixed sampling period, the pulse width amplitude value corresponding to each sampling time is recorded, and the output signal feedback set is summarized. The process of determining whether the channel deviation is within the deviation threshold range is specifically defined as follows: the actual pulse width amplitude of each channel in the output signal feedback set is compared with the minimum pulse width parameter and the maximum pulse width parameter, and the actual pulse width amplitude is determined based on the set pulse width deviation threshold to determine whether it is within the allowable deviation range. The determination result is used to generate the multi-configuration output signal drive state set.

[0035] Compared with the prior art, the advantages and positive effects of the present invention are as follows: 1. In this invention, by collecting pin levels and comparing them with a configuration determination table, automatic identification and number extraction of flight configurations are achieved, enabling efficient linkage between assembly status and control link. The remote control channel input, combined with directional response configuration, completes semantic mapping, ensuring consistent control logic across different configurations. The remote control amplitude is normalized and proportionally adjusted to adapt to the standard commands required for the current configuration, avoiding error interference. Control quantities are precisely allocated to the execution interface based on the configuration index, improving response efficiency and control accuracy. The output signal undergoes feedback detection for deviation evaluation. The overall process constructs an integrated closed loop of identification, self-adaptation, and verification, enhancing multi-configuration adaptability and system stability.

[0036] 2. This invention further enables rapid switching between fixed-wing and multi-rotor modes on the same UAV platform through a modular structure for quick assembly and disassembly of the wing and rotor components. This allows the UAV to select "long-endurance cruise" or "vertical take-off and landing hovering" capabilities according to the mission type, solving the problems of equipment redundancy and low mission adaptability caused by the need to configure fixed-wing and multi-rotor platforms separately in traditional UAVs, thereby improving the system's versatility and engineering application value. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 This is a system flowchart of the present invention; Figure 2 This is a system block diagram of the present invention; Figure 3 This is a flowchart of the configuration recognition module in this invention; Figure 4 This is a flowchart of the semantic mapping module in this invention; Figure 5 This is a flowchart of the channel conversion module in this invention; Figure 6 This is a flowchart of the control quantity allocation module in this invention; Figure 7 This is a flowchart of the output driving module in this invention; Figure 8 This is a schematic diagram of the unmanned aerial vehicle (UAV) structure of the present invention.

[0039] In the diagram: 1. Nose; 2. Fuselage; 3. Wing; 4. Tail; 5. Arm; 6. Vertical propeller; 7. Propulsion propeller. Detailed Implementation

[0040] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0041] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0042] This invention provides a reconfigurable, adaptive, multi-purpose modular unmanned aerial vehicle (UAV) system, such as... Figure 1-2 The diagram shows a reconfigurable, adaptive, multi-purpose modular unmanned aerial vehicle (UAV) system, which includes:

[0043] The configuration recognition module acquires the level status of the splicing detection pin, collects the signal combination output by the configuration recognition component, compares it with the standard configuration status in the configuration index judgment table, identifies whether the current assembly of the UAV includes fixed-wing components, vertical take-off and landing units or multi-rotor components, verifies whether the configuration status is consistent with the configuration number of the flight control system, extracts the configuration index number as the input basis for the flight control link, and generates reconfigurable configuration recognition results. Based on the reconfigurable configuration recognition results, the semantic mapping module calls the channel control mapping table, collects the real-time input amplitude of the remote control channel, identifies the control channel it represents, and judges the semantic direction in combination with the direction response configuration. Under the current configuration, it completes the mapping reconstruction between the channel and the control target, unifies the control logic structure, and generates a configuration-consistent semantic instruction set. The channel conversion module reads the input signal amplitude range and median reference value from the flight control parameters according to the configuration-consistent semantic instruction set, performs amplitude conversion and normalization operations on the remote control input value, adjusts the amplitude range in combination with the proportional adjustment coefficient, converts it into the standard amplitude value of the control command required for the current configuration, and generates a configuration-matching remote control amplitude set. The control quantity allocation module, based on the configuration matching remote control amplitude set, calls the PWM output channel configuration table, filters the control output channel corresponding to the current configuration index, allocates the control quantity to the rotor control, control surface execution or thrust vector device interface, establishes the mapping relationship between the output port and the physical execution component, and generates the reconfiguration configuration control output mapping table. The output drive module calls the reconfiguration configuration control output mapping table, configures the PWM signal output channel, sets the required frequency and pulse width range of the channel, sends the control signal to the target port, detects the actual signal output status, compares the amplitude deviation between the set value and the feedback signal, determines whether the signal output meets the control requirements, and generates a multi-configuration output signal drive state set.

[0044] The reconfigurable configuration identification results include the fixed-wing component inclusion status, the vertical take-off and landing unit inclusion status, and the multi-rotor component inclusion status. The configuration-consistent semantic instruction set includes control channel semantic tags, control target semantic identifiers, and direction semantic pointers. The configuration-matching remote control amplitude set includes normalized control amplitude, median reference offset, and proportional coefficient correction amplitude. The reconfigurable configuration control output mapping table includes the PWM output channel number set, the actuator interface identifier set, and the control quantity type allocation item. The multi-configuration output signal drive state set includes the channel frequency setting state, pulse width range output state, amplitude deviation judgment item, and control satisfaction judgment result.

[0045] Table 1. Configuration Recognition Level Acquisition and Judgment Parameters Table 1 lists the parameter values ​​and ranges used in this embodiment for acquiring the level of the splicing detection pin, determining the continuous state, and setting the comparison threshold. The low-level and high-level judgment boundaries are set with reference to the common digital input level judgment range. The pulse width modulation related update frequency and the remote control channel pulse width range are given in subsequent tables.

[0046] Specifically, such as Figure 2 , 3 As shown, the configuration recognition module includes:

[0047] The signal acquisition submodule acquires the level status of the splicing detection pin, acquires the signal groups output by the configuration recognition component, records the information of the signal and the corresponding pin number, determines whether the signal duration meets the recognition time requirement, selects the signal groups that meet the conditions as the basis for analysis, and generates a signal combination sequence.

[0048] The signal level of the splicing detection pins is acquired according to the acquisition cycle listed in Table 1. The instantaneous voltage value of each detection pin is continuously read, and the instantaneous voltage value is divided into low-level state and high-level state according to the upper limit of low-level judgment and the lower limit of high-level judgment. Voltage values ​​in between are marked as uncertain states and trigger repeated sampling. The number of repeated samplings is executed according to the anti-jitter counting window listed in Table 1. When the proportion of uncertain states in the anti-jitter window exceeds 0.20, the signal level of the pin in this window is marked as invalid and the reason for invalidity is recorded. The acquisition action performs the same level quantization processing on the signal group output by the configuration recognition component. Each round of acquisition is written into a record entry. The entry field consists of acquisition timestamp, pin number, level state, and anti-jitter count result. The signal duration determination is performed according to the recognition duration requirement. The duration of the same pin number appearing in the same level state in the continuous acquisition entries is accumulated. When the accumulated duration reaches 300 milliseconds and there are no invalid markings during the period, the level state of the pin is confirmed as a stable state. The signal loss determination time is triggered when no valid level is acquired within a continuous sampling period. When a pin continuously outputs a valid level, the signal loss determination is not performed. The signal group filtering action is based on a stable state. The stable level values ​​of all pins within the same time window are concatenated in pin number order to form a signal group encoding sequence, and a time window number is inserted into the sequence to obtain a signal combination sequence. In this example, eight detection pins are set up, and the stable level values ​​within a certain time window are high, high, low, low, high, low, high, low, low. One signal group code is formed according to the above rules and written into the sequence. After obtaining 10 consecutive time windows, a signal combination sequence of length 10 is formed for subsequent comparison.

[0049] The state comparison submodule is based on the signal combination sequence. It calls the standard configuration state in the configuration index judgment table, compares the level of the signal with the corresponding level of the standard state, determines whether the matching ratio reaches the recognition and comparison threshold, filters the configuration state numbers that meet the conditions, and obtains the matching configuration number group.

[0050] After reading the signal combination sequence, the standard configuration state entries in the configuration index judgment table are called one by one. Each standard configuration state entry consists of a standard level bit sequence and a configuration state number. The comparison action is performed at the level bit granularity. The level bits of each signal group code in the signal combination sequence are compared with the level bits at the same position in the standard level bit sequence. The consistency count is recorded as a hit when the level states are the same and a miss when the level states are different. Uncertain state entries are not included in the hit and miss counts and the total number of valid bits for that comparison is reduced synchronously. The matching ratio is calculated using literal logic. The matching ratio is the number of hit bits divided by the total number of valid bits. When the matching ratio reaches the identification comparison threshold of 0.85 listed in Table 1, the standard configuration state number is recorded as a candidate and added to the candidate list. At the same time, the matching ratio and the total number of valid bits corresponding to the number are recorded. To ensure consistency of the comparison in the time dimension, the 10 signal group codes of the signal combination sequence are compared with the same standard configuration state, and the average of the 10 matching ratios is taken as the matching ratio of the standard configuration state. In this example, a standard configuration state has an average number of hit bits of 7.2 over 10 time windows, with a total of 8 valid bits. Substituting 7.2 and 8 into the matching ratio calculation logic yields a matching ratio of 0.90, which meets the threshold requirement and is added to the candidate list. After sorting the candidate list from high to low according to the matching ratio, matching configuration number groups are obtained. Each number in the group retains its average matching ratio, the number of time windows involved in the comparison, and the proportion of uncertain bits, which serve as the basis for subsequent index generation verification.

[0051] The index generation submodule calls the configuration number of the flight control system according to the matching configuration number group, and determines whether the number is consistent with the current configuration number. If there is no consistent item, it filters the number with the highest matching ratio and verifies the number information to obtain the reconfigurable configuration recognition result.

[0052] After reading the matching configuration number group, the configuration number of the flight control system is used as the current configuration number. Consistency checks are then performed on each of the current configuration number and the matching configuration number group. The consistency check uses complete string matching as the criterion, without performing fuzzy matching. When a matching item exists, the configuration index number corresponding to that item is directly selected and written to the output buffer. Simultaneously, the matching ratio and the total number of valid bits for that number are written to the verification record. When no matching item exists, the number with the highest matching ratio is selected as the primary candidate. The primary candidate's number information is then verified. The verification field consists of the number length, verification bit rules, and the mapping reachability of the decision table. The number length verification uses the fixed length constraint for this type of configuration in the decision table. The verification bit rules use the last digit parity check or fixed value check in the decision table. The mapping reachability check uses the existence of a unique mapping entry for the primary candidate number in the decision table as the condition. In this example, the current configuration number is 2103. The matching configuration number groups are 2101, 2105, and 2110, but none of them are consistent. The primary candidate is 2105, and the matching ratio is the highest at 0.92. After length verification and mapping reachability verification, 2105 is confirmed as the configuration index number and output. The experimental verification of the threshold and verification rules was carried out using a platform configuration switching test. The test performed 30 configuration switching tests, and the recognition and comparison threshold increased from 0.70 to 0.95, with 10 tests at each level. The number of correct recognitions and the number of misrecognitions were recorded. The results showed that when the threshold was 0.85, the number of correct recognitions was 9 and the number of misrecognitions was 0; when the threshold was 0.80, the number of correct recognitions was 9 but the number of misrecognitions was 1; and when the threshold was 0.90, the number of correct recognitions was 8 and there were no misrecognitions. A comprehensive selection of 0.85 was chosen as the recognition and comparison threshold and recorded in Table 1.

[0053] Table 2 Remote Control Channel Input Amplitude and Direction Threshold Parameters Table 2 provides the commonly used pulse width range and median reference value settings for remote control channels. It also provides the directional response threshold range and stability judgment window configuration. The pulse width range values ​​are referenced from the pulse width ranges of common remote control receivers and servo control systems.

[0054] Specifically, such as Figure 2 , 4 As shown, the semantic mapping module includes:

[0055] The amplitude acquisition submodule, based on the reconfigurable configuration recognition result, calls the channel control mapping table to acquire the real-time input amplitude of the remote control channel, identifies the channel number corresponding to the input amplitude, records the correspondence between the channel and the input amplitude, eliminates signal loss channels and interference input channels, filters the data of channel with stable signal, and generates a channel input amplitude sequence.

[0056] After obtaining the reconfigurable configuration recognition result, the channel control mapping table is invoked and the remote control channel list corresponding to the current configuration index is extracted. Real-time input amplitude acquisition is performed on each channel in the list. The signal type for acquisition is selected as pulse width modulation input pulse width value, and the sampling period is 10 milliseconds as listed in Table 2. Continuous sampling forms a channel input amplitude sequence, and each record consists of a sampling timestamp, channel number, and input pulse width value. The channel number identifier uses the logical channel number of the mapping table, and the physical input port number is written synchronously when writing the record to ensure consistency in subsequent data backtracking. Channels with lost signals are removed according to the signal loss judgment duration listed in Table 1. When a channel fails to acquire a valid pulse width or the pulse width value exceeds the minimum and maximum pulse width range listed in Table 2 within 200 consecutive milliseconds, the channel is marked as lost and removed from the stable channel candidate set. Interference input channel removal employs a deviation removal rule. For each channel, the absolute deviation of the input pulse width value relative to the median pulse width within a 500-millisecond window is calculated. Sampling points with an absolute deviation exceeding 80 microseconds are considered interference points. When the proportion of interference points exceeds 0.30, the channel is marked as interference and removed. In the example, channel 3 is sampled 50 times within a 500-millisecond window, with a median pulse width of 1498 microseconds. 17 sampling points have a deviation exceeding 80 microseconds. Substituting 17 and 50 into the proportion calculation logic, the proportion of interference points is calculated to be 0.34, and channel 3 is removed. The remaining channels form the signal-stabilized channel data, and the channel input amplitude sequence is generated based on this.

[0057] The direction discrimination submodule calls the direction response configuration parameters based on the channel input amplitude sequence, distinguishes and judges the channel input amplitude and direction response threshold, marks the response direction attribute of the channel under the current configuration, filters the channel data with clear direction attributes in the marking results, and generates a direction pointing identification sequence.

[0058] After reading the channel input amplitude sequence, the direction response configuration parameters are invoked, and direction attribute labeling is performed on each channel. The discrimination action uses the lower and upper bounds of the direction response thresholds listed in Table 2 as boundaries to distinguish between the average input pulse width and the threshold interval within a 500-millisecond window. An average input pulse width below 1450 microseconds is labeled as a reverse response, an average input pulse width above 1550 microseconds is labeled as a forward response, and an average input pulse width between 1450 and 1550 microseconds is labeled as median hold and considered to have an ambiguous direction attribute. To avoid transient bias, the pulse width fluctuation range within this window is simultaneously read. The fluctuation range is calculated by subtracting the minimum pulse width from the maximum pulse width within the window. If the fluctuation range exceeds 120 microseconds, the window is labeled as unstable, and no direction attribute is output. In the example, if the average input pulse width in channel 1 is 1620 microseconds and the fluctuation range is 60 microseconds, it is directly determined as a positive response and a direction indicator is written. If the average input pulse width in channel 2 is 1510 microseconds, it is marked as median hold and discarded. Channel data with clear directional attributes are organized into a direction indicator sequence. The sequence elements consist of channel number, directional attribute, average input pulse width, and fluctuation range, and retain the index association with the channel input amplitude sequence in the previous paragraph.

[0059] The instruction mapping output submodule calls the direction pointing identifier sequence, combines it with the channel mapping control rules under the current configuration, establishes the correspondence structure between the direction attribute and the target control channel, filters the combination relationship with consistent channel target pointing, constructs a consistent mapping relationship between channel instructions and target control channels, and generates a configuration-consistent semantic instruction set;

[0060] After reading the direction pointing identifier sequence, a correspondence structure between direction attributes and target control channels is established based on the channel mapping control rules under the current configuration. The correspondence structure construction process first reads the direction semantic constraints of each target control channel in the mapping rules. These semantic constraints are described by two fields: direction attribute and action category. Then, consistency filtering is performed on each channel in the direction pointing identifier sequence and its semantic constraints. The consistency condition is that the direction attributes are the same and the channel number belongs to the allowed set of the mapping rules. When multiple candidate input channels appear for the same target control channel, channel target pointing consistency filtering is performed. The filtering condition prioritizes channels with a larger deviation between the average input pulse width and the median pulse width of 1500 microseconds. A larger deviation indicates that the channel's pointing is more precise in the current control action. In this example, the target control channel is pitch control, the semantic constraint is positive response, and the candidate input channels are channel 1 and channel 4. Channel 1 has an average input pulse width of 1620 microseconds, and channel 4 has an average input pulse width of 1580 microseconds. Therefore, the deviation of channel 1 is 120 microseconds, and the deviation of channel 4 is 80 microseconds. Channel 1 is selected as the mapping input. After the filtering is completed, the channel instruction logic is organized and the relationship is unified. The unified relationship record field consists of the input channel number, the target control channel number, the direction attribute, and the amplitude source window number. A configuration-consistent semantic instruction set is generated and output for subsequent channel conversion.

[0061] Specifically, such as Figure 2 , 5 As shown, the channel conversion module includes:

[0062] The interval reading submodule reads the amplitude range and median reference value of the input signal from the flight control parameters based on the configuration-consistent semantic instruction set, marks the minimum amplitude, maximum amplitude and median amplitude of the corresponding channel, filters the channel records with complete amplitude range information, and generates the channel amplitude benchmark dataset.

[0063] Based on the configuration-consistent semantic instruction set, the amplitude range and median reference value of the input signal in the flight control system parameters are read, and additional amplitude reference information is recorded for each instruction. The reading action performs parameter retrieval for each target control channel, searching for the minimum amplitude, maximum amplitude, and median amplitude. The amplitude unit is uniformly in microseconds, and the value range is 1000 to 2000 microseconds as listed in Table 2, serving as the initial consistency verification boundary. If the read minimum amplitude is less than 900 microseconds or the maximum amplitude is greater than 2100 microseconds, the channel parameter is marked as abnormal and entered into the review queue. The integrity screening of amplitude range information uses the condition that all three fields exist and the minimum amplitude is less than the median amplitude and the median amplitude is less than the maximum amplitude. In the example, the throttle control channel reads a minimum amplitude of 990 microseconds, a maximum amplitude of 2010 microseconds, and a median amplitude of 1500 microseconds, satisfying the integrity condition and being written into the channel amplitude reference dataset; a control surface channel with a missing maximum amplitude is removed. The record fields of the channel amplitude benchmark dataset consist of the target control channel number, minimum amplitude, maximum amplitude, median amplitude, and parameter reading timestamp. It also retains the corresponding index of the semantic instruction set consistent with the configuration, ensuring that subsequent conversions are performed based on the same instruction.

[0064] The amplitude adjustment submodule calls the channel input values ​​from the configuration-consistent semantic instruction set based on the channel amplitude reference dataset, performs amplitude conversion and normalization on the input values, combines the proportional adjustment coefficient to convert the amplitude range, filters the channel data whose amplitude falls within the preset range in the processing results, and generates a channel converted amplitude sequence.

[0065] The system reads the channel input values ​​from the channel amplitude reference dataset and the configuration-consistent semantic instruction set, performs amplitude conversion and normalization, and combines the proportional adjustment coefficient to complete the amplitude range conversion. The conversion logic is described in words: first, the input pulse width value is truncated, with the minimum and maximum amplitude values ​​as the truncation boundaries; then, the offset of the input pulse width value relative to the median amplitude is calculated; and finally, the offset is normalized according to either the forward or reverse interval length to obtain the normalized control value. The proportional adjustment coefficient is set according to bench calibration tests. Under a fixed configuration, three fixed control levels are applied to each target control channel, with input pulse widths of 1300 microseconds, 1500 microseconds, and 1700 microseconds. The changes in angle or speed fed back by the actuator are recorded, and the ratio of the target change to the actual change is used as a candidate coefficient. The median of 10 repeated tests is taken as the proportional adjustment coefficient for that channel. In this example, the target change of a certain control surface channel is 15 degrees, and the median change over 10 tests is 12 degrees. Substituting 15 and 12 into the ratio calculation logic yields a proportional adjustment coefficient of 1.25. This coefficient is then used in the amplitude range conversion logic to amplify or reduce the normalized control quantity, generating a channel conversion amplitude sequence. The reasonable range selection is based on the condition that the converted pulse width back-calculated value still falls within the minimum and maximum amplitude range. Records exceeding this range are marked as out of bounds and reverted to the range boundary value, while an out-of-bounds count is written. If the out-of-bounds count exceeds twice within five consecutive windows, the channel is marked as an input anomaly.

[0066] The standard generation submodule calls the channel conversion amplitude sequence, maps the standard amplitude of the command required for the current configuration, filters the conversion channel amplitude that meets the preset mapping requirements, constructs the correspondence between channel commands and standard amplitudes, and generates a configuration-matched remote control amplitude set.

[0067] After reading the channel conversion amplitude sequence, the corresponding mapping is performed on the standard amplitude of the command required for the current configuration, and the configuration-matched remote control amplitude set is output. The mapping action first reads the standard amplitude table, which provides the standard minimum, standard median, and standard maximum values ​​for the target control channel. Then, the converted normalized control quantity is mapped to the standard amplitude range according to a linear correspondence. When the normalized control quantity is negative, it is mapped to the range from the standard minimum to the standard median; when the normalized control quantity is positive, it is mapped to the range from the standard median to the standard maximum. In the example, the standard minimum value of a target control channel is 1100 microseconds, the standard median value is 1500 microseconds, and the standard maximum value is 1900 microseconds. The normalized control quantity obtained in the previous paragraph is 0.60, and it is still 0.60 after proportional adjustment. Substituting 0.60 into the mapping logic, the standard output pulse width is 1740 microseconds. The mapping requirement is to use the standard output pulse width falling within the range of 1100 microseconds to 1900 microseconds and the out-of-bounds count being 0 as the condition. The set of configuration matching remote control amplitude is generated. The set record fields consist of the target control channel number, standard output pulse width, source input channel number, proportional adjustment coefficient, and nearest window number.

[0068] Specifically, such as Figure 2 , 6 As shown, the control quantity allocation module includes:

[0069] The channel filtering submodule is based on the configuration matching remote control amplitude set. It calls the PWM output channel configuration table, extracts all output channel information corresponding to the current configuration index, compares the enable status of the channel with the channel requirements of the configuration index, filters the channel data with matching status, and generates the configuration output channel set.

[0070] After reading the configuration matching remote control amplitude set, the pulse width modulation output channel configuration table is called and all output channel information corresponding to the current configuration index is extracted. The filtering process first reads the enable status, port number, supported update frequency upper limit, and pulse width range constraint of each output channel. Then, these fields are matched against the channel requirements of the configuration index. The matching condition is that the enable status is enabled and the pulse width range covers the standard output pulse width of the target control channel. If multiple available output channels are mapped to the same target control channel, conflict resolution is performed. The resolution rule prioritizes the consistency between the port number and the physical execution component binding table, followed by prioritizing the channel with the higher supported update frequency. A reasonable update frequency range is selected based on common servo update rate configurations, with 50 Hz to 400 Hz as an optional range. In this example, the configuration index is 2105, and the corresponding output channel list is channel 9 to channel 12. Channel 10 is disabled and has been removed. Channel 9 supports an update frequency limit of 400 Hz and is reserved for the range of 1100 microseconds to 1900 microseconds. The configuration output channel set is generated, and the record fields consist of the output channel number, port number, supported update frequency limit, pulse width range constraint, and bound execution component number.

[0071] The control pointing submodule outputs the channel set according to the configuration, reads the function type information required for the control quantity, compares the channel data with the function categories of rotor control, control surface execution, and thrust vector interface, establishes the allocation correspondence of control quantity in the function category, filters and allocates structural control quantity records, and generates a function control quantity lookup table.

[0072] The system reads the required function type information for the control input based on the configuration output channel set and compares the channel data with the function categories of rotor control, control surface actuation, and thrust vector interface. The comparison process first reads the function category label of each output channel in the configuration table, then reads the action category label of each target control channel in the configuration-matched remote control amplitude set, and subsequently establishes an allocation correspondence based on label consistency. For channel records with missing labels, the system completes the information based on the component type table corresponding to the port-bound execution component number. If the component type is an ESC, it is classified as rotor control; if the component type is a servo, it is classified as control surface actuation; and if the component type is a vector mechanism actuator, it is classified as thrust vector interface. In this example, output channel 9 is bound to a servo component, so it is classified as control surface actuation and a correspondence is established with the target control channel pitch control. The screening action performs an integrity check on the allocation structure. The check condition is that each target control channel has a unique output channel and each output channel is only assigned to one target control channel. Records that violate the condition are written to the conflict list and trigger reallocation. Reallocation prioritizes unassigned channels, and then selects channels with larger pulse width range coverage within the same functional category, generating a functional control quantity comparison table.

[0073] The output channel mapping submodule calls the function control variable lookup table, extracts the target output channel in the function category, establishes the mapping relationship between the output port and the physical execution unit, eliminates records with redundant binding or undefined channels, organizes the channel items with complete mapping structure, and generates the reconfiguration control output mapping table.

[0074] The process involves reading the function control variable lookup table and extracting the target output channel from the function category to establish a mapping relationship between the output port and the physical execution component. The mapping construction process first reads the port number and bound execution component number for each lookup record, then reads the execution component interface definition table. The interface fields consist of interface type, power supply limitations, allowed update frequency range, and allowed pulse width range. Subsequently, the port parameters and interface constraints are checked for consistency. Redundant bindings are eliminated if the same execution component number appears more than once in the mapping relationship. Undefined channels are eliminated if the port number does not exist in the interface definition table. In the example, a thrust vector actuator is bound to both channel 11 and channel 12 simultaneously. The allocation record with the higher matching ratio is retained. The matching ratio is taken from the matching ratio in the previous configuration identification stage and written along the index into the current stage record, achieving cross-stage data association. After processing, a reconstructed configuration control output mapping table is output, with fields consisting of configuration index number, function category, target control channel number, output channel number, port number, execution component number, and allowed update frequency range.

[0075] Specifically, such as Figure 2 , 7 As shown, the output driver module includes:

[0076] The channel configuration submodule calls the reconfiguration configuration control output mapping table, configures the PWM signal output channel, sets the output frequency range and pulse width interval parameters corresponding to the channel, records the correspondence between the channel number and the parameter configuration, filters the channel records with complete configuration information, and establishes the output channel parameter set.

[0077] The reconfiguration configuration control output mapping table is invoked to configure parameters for the pulse width modulation (PWM) signal output channel. The configuration fields consist of the output frequency range and pulse width interval parameters. The output frequency range is selected from 50 Hz to 400 Hz. The selection rule uses the smaller of the allowed update frequency upper limit from the actuator interface definition table and the global update frequency upper limit of the flight control system as the upper limit. Within this upper limit, the closest setting to 200 Hz is selected as the default value. If the actuator is a simulated servo, 50 Hz is forcibly selected. The pulse width interval parameter is taken from the standard minimum and maximum values ​​generated during the standard generation phase, and the intersection with the allowed pulse width interval of the interface is used to obtain the configuration interval. In this example, the allowed update frequency upper limit for the control surface servo is 333 Hz, and the global upper limit is 400 Hz. Therefore, 333 Hz is selected as the upper limit, and 200 Hz is selected as the configuration frequency from the candidate settings. The pulse width interval intersection is 1100 microseconds to 1900 microseconds, which is written into the output channel parameter set. The configuration information integrity screening uses the condition that each record simultaneously has the output channel number, configuration frequency, minimum pulse width, maximum pulse width, port number and execution unit number. Records that do not meet the conditions are added to the missing list and subsequent transmission actions of that channel are blocked.

[0078] The signal transmission submodule loads the control signal to the corresponding output channel based on the output channel parameter set, sends the PWM signal to the target port, monitors the real-time output status of the channel during the transmission process, collects the signal amplitude data returned by the port, and forms an output signal feedback set.

[0079] Based on the output channel parameter set, the control signal is loaded onto the corresponding output channel, and a pulse width modulation (PWM) signal is sent to the target port. The loading action first reads the standard output PWM value bound to the output channel from the configuration-matched remote control amplitude set, and performs a pre-transmission verification on the standard output PWM. The verification condition uses a time constraint that the PWM falls within the configured PWM range and the most recent window number does not exceed two windows. Records exceeding the time constraint trigger resampling and return to the amplitude acquisition stage to obtain a new input amplitude. The transmission action executes PWM output at the configured frequency and monitors the real-time output status during transmission. The monitored fields consist of the actual output PWM, output period jitter, and port voltage drop. The signal amplitude data returned by the acquisition port is processed using a readback mechanism or an external measurement channel sampling mechanism. The external measurement channel sampling period is 10 milliseconds, consistent with the input amplitude sampling period for alignment. In the example, output channel 9 sends a standard output pulse width of 1740 microseconds and is configured with a frequency of 200 Hz. The actual median output pulse width is detected to be 1732 microseconds and the period jitter is 3 microseconds. The signal is written into the output signal feedback set. The feedback set record fields consist of the output channel number, target pulse width, actual median pulse width, period jitter, and sampling timestamp.

[0080] The state determination submodule compares the channel set amplitude range with the feedback signal amplitude based on the output signal feedback set, calculates the amplitude deviation, determines whether the channel deviation is within the deviation threshold range, summarizes the channel determination results, and generates a multi-configuration output signal drive state set.

[0081] The output signal feedback set is read and compared with the configured pulse width range and the feedback signal amplitude. The amplitude deviation is calculated, and it is determined whether the deviation is within the deviation threshold range. The deviation threshold range is obtained through experimental calibration. In the calibration test, 10 pulse width commands are continuously sent to each output channel on a test bench. The command pulse width covers 1100 microseconds to 1900 microseconds and is incremented in 80-microsecond steps. The absolute value of the deviation between the target pulse width and the median of the actual pulse width is recorded using the same measurement link. The 90th percentile of the absolute value of the deviation is used as the deviation threshold for that channel. In the example, the 90th percentile of the absolute value of the deviation for channel 9 in the 10-level test is 12 microseconds, so 12 microseconds is written into the deviation threshold. During operation, if the current feedback deviation absolute value of this channel is 8 microseconds, then 8 and 12 are substituted into the comparison logic, determined to be within the threshold range, and marked as normal. Multi-channel aggregation is performed by aggregating by configuration index number, and the judgment results of all output channels are organized into a multi-configuration output signal driven state set. The state set fields consist of configuration index number, output channel number, normal label, absolute value of deviation, deviation threshold, period jitter, and nearest window number.

[0082] Table 3. Comparison of Key Thresholds and Recognition Output: Experimental Results Table 3 lists the bench test results for different configuration recognition threshold values, and also provides the output deviation compliance rate, which is the percentage of sampling points where the absolute value of the deviation of all output channels falls within the deviation threshold during the 30-second stable period. The experimental results show that when the threshold is 0.85, compared to 0.80, the number of false recognitions is reduced by 100%, the output deviation compliance rate is increased by 4.3%, and the average matching ratio remains at 0.90. Furthermore, the matching configuration number groups converge more stably to a single configuration index number, thereby reducing the number of subsequent channel mapping conflict entries.

[0083] like Figure 8 As shown, the UAV airframe structure in this embodiment adopts a modular and reconfigurable design. The airframe consists of a nose 1, fuselage 2, wings 3, tail 4, arms 5, vertical propeller 6, and propeller 7. The wings 3 are detachably connected to the arms 5 and the vertical propeller 6. Similarly, the propeller 7 is quickly detached from the fuselage 2, enabling the UAV to rapidly switch configurations and reuse its platform under different mission requirements. The wings 3 adopt a high-wing configuration, reducing aerodynamic interference between the wings and the vertical propeller 6 to improve lift characteristics.

[0084] Specifically, this embodiment achieves at least three configuration reconfiguration methods through rapid assembly and disassembly of components:

[0085] 1) Fixed-wing configuration switching method: When long-endurance, long-distance cruise missions are required, the vertical propeller 6 and its corresponding arm 5 can be quickly disassembled, leaving the wings 3 and propeller 7 intact, thus forming a fixed-wing unmanned aircraft configuration. In this configuration, propeller 7 provides forward thrust, and wings 3 provide lift, making it suitable for mission scenarios such as long-endurance cruise, surveying, and wide-area reconnaissance.

[0086] 2) Multi-rotor configuration switching method: When tasks such as hovering, vertical takeoff and landing, and fixed-point operations are required, the nose 1, wings 3, tail 4, and propeller 7 can be quickly disassembled without removing the vertical rotor 6, transforming the aircraft into a multi-rotor unmanned structure. In this configuration, the vertical rotor 6 provides lift and attitude control, enabling vertical takeoff and landing and stable hovering. It is suitable for mission scenarios such as takeoff and landing in confined spaces, fixed-point inspection, and low-altitude operations.

[0087] 3) Compound Wing Configuration Switching: When performing missions requiring both long-endurance cruise and vertical takeoff and landing (VTOL) capabilities, the airframe can be configured as a compound wing. In this configuration, the vertical rotor 6 and the propeller 7 are retained, while the wing 3 can be adjusted to a suitable angle for compound wing flight, allowing the UAV to freely switch between VTOL and cruise flight. Specifically, by adjusting the operating mode of the vertical rotor 6 and the lift characteristics of the wing 3, the vertical rotor 6 provides sufficient lift during VTOL; during cruise flight, the wing 3 provides lift, and the propeller 7 provides forward thrust.

[0088] Through the above structural design, this embodiment achieves rapid reconfigurable conversion between fixed-wing, multi-rotor, and compound-wing configurations on the same UAV platform. This significantly improves the adaptability and efficiency of the UAV system under different application tasks and reduces the cost of repetitive configuration across multiple platforms. The introduction of the compound-wing configuration enables the UAV system to operate not only in vertical takeoff and landing and stable hovering modes, but also to operate efficiently in missions requiring long endurance and long-distance cruising. The compound-wing configuration is particularly suitable for use in urban air travel, disaster search and rescue, and maritime patrol scenarios. It exhibits extremely high flexibility and adaptability in various complex environments, enhancing the UAV's performance capabilities in various complex tasks and further expanding the application areas of UAVs.

[0089] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A reconfigurable, adaptive, multi-purpose modular unmanned aerial vehicle system, characterized in that: The system includes: The configuration recognition module reads the splicing detection pin level, collects the configuration recognition signal, compares it with the configuration judgment table, identifies whether it includes fixed-wing, vertical take-off and landing or multi-rotor components, verifies that the configuration is consistent with the flight control number, extracts the configuration index, and generates a reconfigurable configuration recognition result. The semantic mapping module calls the channel mapping table based on the reconfigurable configuration recognition result, collects remote control channel input, identifies the control channel, determines the semantic direction through the direction response configuration, reconstructs the channel and control target mapping, and generates a configuration-consistent semantic instruction set. The channel conversion module reads flight control parameters based on the configuration-consistent semantic instruction set, performs amplitude conversion and normalization on the remote control input, adjusts the output in combination with a preset proportional coefficient, and generates a configuration-matched remote control amplitude set. Based on the configuration matching remote control amplitude set, the control quantity allocation module calls the PWM channel configuration table, filters the output channel matching the configuration, allocates the control quantity to the rotor, control surface or thrust device, establishes the mapping relationship between control and execution components, and generates a reconfiguration configuration control output mapping table. The output drive module calls the reconfiguration configuration control output mapping table, configures the PWM channel frequency and pulse width, outputs control signals, detects the deviation between the feedback and the set value, makes a state determination based on the deviation, and generates a multi-configuration output signal drive state set.

2. The reconfigurable adaptive multi-purpose modular unmanned aerial vehicle system according to claim 1, characterized in that: The reconfigurable configuration identification results include the fixed-wing component inclusion status, the vertical take-off and landing unit inclusion status, and the multi-rotor component inclusion status. The configuration-consistent semantic instruction set includes control channel semantic tags, control target semantic identifiers, and direction semantic pointers. The configuration-matching remote control amplitude set includes normalized control amplitude, median reference offset, and proportional coefficient correction amplitude. The reconfigurable configuration control output mapping table includes a PWM output channel number set, an execution component interface identifier set, and a control quantity type allocation item. The multi-configuration output signal drive state set includes channel frequency setting state, pulse width range output state, amplitude deviation judgment item, and control satisfaction judgment result.

3. The reconfigurable adaptive multi-purpose modular unmanned aerial vehicle system according to claim 1, characterized in that, The configuration recognition module includes: The signal acquisition submodule acquires the level status of the splicing detection pin, acquires the signal groups output by the configuration recognition component, records the information of the signal and the corresponding pin number, determines whether the signal duration meets the recognition time requirement, selects the signal groups that meet the recognition time requirement as the basis for analysis, and generates a signal combination sequence. The state comparison submodule, based on the signal combination sequence, calls the standard configuration state in the configuration index judgment table, compares the level of the signal with the corresponding level of the standard state, determines whether the matching ratio reaches the preset recognition comparison threshold, filters the configuration state numbers that meet the conditions, and obtains the matching configuration number group. The index generation submodule calls the configuration number of the flight control system according to the matching configuration number group, determines whether the number is consistent with the current configuration number, and if there is no consistent item, it filters the number with the highest matching ratio and verifies the number information to obtain the reconfigurable configuration recognition result.

4. A reconfigurable adaptive multi-purpose modular unmanned aerial vehicle system according to claim 3, characterized in that: The process of determining whether the signal duration meets the recognition time requirement is specifically defined as follows: continuously sampling the level state corresponding to the same pin number, the number of samplings that continuously maintain the same level state is not less than a preset sampling number threshold, and the corresponding time length is not less than a preset recognition time parameter, in order to determine that the signal duration meets the recognition condition. The process of determining whether the matching ratio reaches the identification and comparison threshold is specifically defined as follows: taking the number of effective level bits participating in the comparison in the signal combination sequence as the comparison base, counting the number of level bits that are completely consistent with the standard configuration state in the configuration index determination table, and calculating the ratio of the number of consistent level bits to the comparison base. When the ratio is not less than the preset identification and comparison threshold, the corresponding configuration state number is determined as a valid matching number. The process of selecting the number with the highest matching ratio and verifying the number information is specifically limited to comparing the matching ratio values ​​corresponding to each configuration state number in the matching configuration number group, selecting the configuration state number with the largest matching ratio value, and verifying the consistency between the configuration identification component number information corresponding to the configuration state number and the configuration number information stored in the flight control system. After the verification is passed, it is taken as the reconfigurable configuration identification result.

5. A reconfigurable adaptive multi-purpose modular unmanned aerial vehicle system according to claim 3, characterized in that, The semantic mapping module includes: Based on the reconfigurable configuration recognition result, the amplitude acquisition submodule calls the channel control mapping table to acquire the real-time input amplitude of the remote control channel, identifies the channel number corresponding to the input amplitude, records the correspondence between the channel and the input amplitude, eliminates signal loss channels and interference input channels, filters the data of signal stable channels, and generates a channel input amplitude sequence. The direction discrimination submodule calls the direction response configuration parameters based on the channel input amplitude sequence, distinguishes and judges the channel input amplitude and direction response threshold, marks the response direction attribute of the channel under the current configuration, filters the channel data with clear direction attributes in the marking results, and generates a direction pointing identification sequence. The instruction mapping output submodule calls the direction pointing identifier sequence, combines it with the channel mapping control rules under the current configuration, establishes the correspondence structure between the direction attribute and the target control channel, filters the combination relationship with consistent channel target pointing, constructs the consistent mapping relationship between channel instructions and target control channels, and generates a configuration-consistent semantic instruction set.

6. A reconfigurable adaptive multi-purpose modular unmanned aerial vehicle system according to claim 5, characterized in that, The channel conversion module includes: The interval reading submodule reads the input signal amplitude interval and median reference value from the flight control parameters based on the configuration-consistent semantic instruction set, marks the minimum amplitude, maximum amplitude and median amplitude corresponding to the channel, filters the channel records with amplitude interval information, and generates a channel amplitude benchmark dataset. The amplitude adjustment submodule calls the channel input values ​​from the configuration-consistent semantic instruction set based on the channel amplitude reference dataset, performs amplitude conversion and normalization on the input values, performs amplitude range conversion in combination with the proportional adjustment coefficient, filters the channel data whose amplitude falls within the preset range in the processing results, and generates a channel conversion amplitude sequence. The standard generation submodule calls the channel conversion amplitude sequence to map the standard amplitude of the command required for the current configuration, filters the conversion channel amplitude that meets the preset mapping requirements, constructs the correspondence between channel commands and standard amplitudes, and generates a configuration-matched remote control amplitude set.

7. A reconfigurable adaptive multi-purpose modular unmanned aerial vehicle system according to claim 6, characterized in that: The process of reading the input signal amplitude range and median reference value in the flight control parameters is specifically defined as follows: reading the minimum amplitude parameter, maximum amplitude parameter, and median reference value parameter of the input signal corresponding to each channel from the flight control parameter storage area, and establishing a correspondence between the minimum amplitude parameter, maximum amplitude parameter, and median reference value parameter of the input signal according to the channel number, as the minimum amplitude, maximum amplitude, and median amplitude corresponding to the channel; The process of amplitude conversion and normalization of the input value is specifically defined as follows: based on the minimum amplitude, maximum amplitude and median amplitude, the channel input value in the configuration consistent semantic instruction set is linearly mapped to a uniform normalized amplitude range, and the amplitude range is converted in combination with a preset proportional adjustment coefficient within the normalized amplitude range. The process of filtering channel data whose amplitude falls within a preset range is specifically defined as follows: mapping the normalized conversion result after proportional adjustment to the corresponding physical amplitude, comparing the physical amplitude with the minimum amplitude and the maximum amplitude, retaining only the channel data whose mapped physical amplitude is between the minimum amplitude and the maximum amplitude, and forming the channel conversion amplitude sequence according to the channel number order.

8. A reconfigurable adaptive multi-purpose modular unmanned aerial vehicle system according to claim 6, characterized in that, The control quantity allocation module includes: The channel filtering submodule, based on the configuration matching remote control amplitude set, calls the PWM output channel configuration table, extracts all output channel information corresponding to the current configuration index, compares the enable status of the channel with the channel requirements of the configuration index, filters the channel data with matching status, and generates a configuration output channel set. The control pointing submodule outputs the channel set according to the configuration, reads the function type information required for the control quantity, compares the channel data with the function categories of rotor control, control surface execution, and thrust vector interface, establishes the allocation correspondence of the control quantity in the function category, filters the allocation structure control quantity records, and generates a function control quantity lookup table. The output channel mapping submodule calls the function control variable lookup table, extracts the target output channel in the function category, establishes the mapping relationship between the output port and the physical execution unit, eliminates records with redundant bindings or undefined channels, organizes the channel items with complete mapping structure, and generates a reconfiguration configuration control output mapping table.

9. A reconfigurable adaptive multi-purpose modular unmanned aerial vehicle system according to claim 8, characterized in that, The output driver module includes: The channel configuration submodule calls the reconfiguration configuration control output mapping table, configures the PWM signal output channel, sets the output frequency range and pulse width interval parameters corresponding to the channel, records the correspondence between the channel number and the parameter configuration, filters the channel records with complete configuration information, and establishes the output channel parameter set. Based on the output channel parameter set, the signal transmission submodule loads the control signal to the corresponding output channel, sends the PWM signal to the target port, monitors the real-time output status of the channel during the transmission process, collects the signal amplitude data returned by the port, and forms an output signal feedback set. The state determination submodule compares the channel set amplitude range with the feedback signal amplitude based on the output signal feedback set, calculates the amplitude deviation, determines whether the channel deviation is within the preset allowable deviation threshold range, summarizes the channel determination results, and generates a multi-configuration output signal drive state set.

10. A reconfigurable adaptive multi-purpose modular unmanned aerial vehicle system according to claim 9, characterized in that: The process of setting the output frequency range and pulse width interval parameters corresponding to the channel is specifically defined as follows: based on the channel type identifier pre-stored in the reconfiguration control output mapping table, configure the minimum output frequency parameter, maximum output frequency parameter, minimum pulse width parameter and maximum pulse width parameter for each channel, and establish a correspondence between the minimum output frequency parameter, maximum output frequency parameter, minimum pulse width parameter and maximum pulse width parameter and the channel number to form the output channel parameter set; The process of monitoring the real-time output status of the channel during transmission and collecting the signal amplitude data returned by the port is specifically defined as follows: during the PWM signal transmission period, the actual pulse width value of each channel output port is sampled multiple times according to a fixed sampling period, the pulse width value corresponding to each sampling time is recorded, and the output signal feedback set is formed by summarizing the data. The process of determining whether the channel deviation is within the deviation threshold range is specifically defined as follows: the actual pulse width amplitude of each channel in the output signal feedback set is compared with the minimum pulse width parameter and the maximum pulse width parameter, and the actual pulse width amplitude is determined based on the set pulse width deviation threshold to determine whether it is within the allowable deviation range. The determination result is used to generate the multi-configuration output signal drive state set.

Images