An environmental adaptive inflation control method and system based on attitude analysis

Abstract

This invention relates to the field of air pump technology and discloses an environmentally adaptive inflation control method and system based on attitude analysis. By establishing a communication network between a main air pump and several slave air pumps, and utilizing the sensing components integrated into the communication network and the air pumps, the main air pump can determine the imbalance pressure replenishment area based on its own collected inflation pressure data and attitude data received externally through the communication network. Then, using imbalance quantification parameters to construct an input state vector, an improved DDPG model trained on it is used to infer the replenishment pressure value of each imbalance replenishment air pump. This generates inflation replenishment commands and drives the corresponding slave air pumps to perform replenishment inflation actions, achieving dynamic attitude maintenance of the inflatable tent. This improves the automation and intelligence of inflatable tent attitude maintenance in scenarios involving steep slopes and gusts of wind, enhancing the accuracy and rationality of air pump replenishment while ensuring the safety of inflatable tent use and reducing interference with outdoor operations.

Description

Technical Field

[0001] This invention relates to the field of air pump technology, and in particular to an environmental adaptive inflation control method and system based on attitude analysis. Background Technology

[0002] In outdoor operations such as exploration, scientific research, and special operations, flat campsites are extremely scarce. Core operational areas in mountains, forests, and suburbs often require tents to be pitched on steep slopes with a gradient of ≥15°. Pitching tents on steep slopes not only effectively avoids low-lying, waterlogged, and damp areas, preventing tent damage from moisture, but also utilizes the terrain's elevation difference to achieve natural ventilation and improve the internal living environment, making it a common choice for tent pitching in outdoor operations. However, such scenarios place stringent requirements on maintaining the posture of inflatable tents. Steep slope environments inherently present a natural imbalance of forces, and when combined with sudden gusts of wind (instantaneous wind speed ≥10m / s), they can have a significant negative impact on the tent: the steep slope causes the tent's center of gravity to shift along the slope direction, with the lower area continuously compressed by gravity and the upper area stretched; sudden gusts of wind also exert lateral impact forces, further exacerbating the uneven stress on different areas of the tent, causing a sudden drop in pressure on the windward side and the lower part of the steep slope, while the pressure on the leeward side and the upper part of the steep slope abnormally increases. Under this dual effect, tents are prone to continuous tilting, insufficient support stability, and even local fabric tearing or collapse due to excessive stretching or improper compression, directly interfering with the normal conduct of outdoor operations.

[0003] Existing solutions for maintaining the attitude of inflatable tents using air pumps have significant technical shortcomings and are ill-suited to the requirements of the aforementioned scenarios: First, air pumps are typically only equipped with pressure sensors, lacking the ability to sense the overall tilt and attitude changes of the tent, and thus unable to determine the direction of tilt and degree of imbalance under conditions of steep slopes and gusts. Second, air pumps at different locations passively inflate only according to fixed pressure thresholds, unable to adjust the pressure replenishment strategy in real time based on attitude changes caused by external forces, easily leading to insufficient or excessive pressure replenishment. Third, each air pump operates independently, unable to share status data, and can only respond to its own pressure signals in isolation, failing to form a global perspective and making it difficult to accurately locate key pressure replenishment areas such as the area below steep slopes and the windward side of gusts. Fourth, existing air pumps cannot accurately distinguish between normal pressure decay and pressure surges caused by attitude imbalance, and are prone to accidental start-up or stoppage under instantaneous pressure fluctuations caused by gusts, further compromising the tent's attitude stability.

[0004] Therefore, improving the automation and intelligence of inflatable tent posture maintenance in scenarios involving steep slopes and gusts of wind, while enhancing the accuracy and rationality of air pump pressurization, ensuring tent safety and reducing interference with outdoor operations, is a technical problem that urgently needs to be solved. Summary of the Invention

[0005] This invention provides an environment-adaptive inflation control method and system based on attitude analysis, aiming to solve at least one of the above-mentioned technical problems.

[0006] To achieve the above objectives, the present invention provides an environment-adaptive inflation control method based on attitude analysis, the method comprising the following steps:

[0007] A communication network is established for several air pumps; wherein, the several air pumps are configured to be installed at several air inlets of the target inflatable tent, and each air pump integrates a pressure sensor for collecting inflation pressure data and an attitude sensor for collecting attitude data.

[0008] Each air pump sends collected inflation pressure and attitude data to the main air pump. The main air pump then builds a real-time global dataset based on the collected and received inflation pressure and attitude data.

[0009] The main air pump uses the attitude data in the real-time global dataset to perform attitude angle calculation of the target inflatable tent. Based on the attitude angle calculation result and the inflation pressure data in the real-time global dataset, it determines whether the target inflatable tent is unbalanced. If so, it determines the unbalanced pressure replenishment area and the unbalanced pressure replenishment air pump.

[0010] The main air pump constructs the imbalance quantification parameters of the imbalance replenishment air pump as an input state vector, and uses the pre-built and trained improved DDPG model to infer the target replenishment pressure value, and converts the target replenishment pressure value into an air replenishment command and sends it to each slave air pump.

[0011] Each air pump performs supplementary inflation of the corresponding unbalanced area according to the inflation and pressurization control parameters in the inflation and pressurization command, in order to adjust the posture of the target inflatable tent.

[0012] Optionally, the steps for establishing a communication network for several air pumps include:

[0013] After the air pumps located at several air inlets complete the inflation of the target inflatable tent, the main air pump sends a network establishment request to several slave air pumps.

[0014] After receiving a network establishment request from the main air pump, several slave air pumps send network response information back to the main air pump. Based on the network response information, the main air pump establishes a communication network with the several slave air pumps.

[0015] Optionally, each air pump sends the collected inflation pressure data and attitude data to the main air pump, and the main air pump establishes a real-time global dataset based on the collected and received inflation pressure data and attitude data. This process specifically includes:

[0016] Each air pump uses an integrated pressure sensor and attitude sensor to collect inflation pressure data and attitude data of the corresponding air inlet at a preset sampling frequency; wherein, the attitude data includes triaxial acceleration data and triaxial angular velocity data.

[0017] Each air pump sends collected inflation pressure and attitude data to the main air pump. The main air pump then builds a real-time global dataset based on the collected and received inflation pressure and attitude data.

[0018] The specific expression for the real-time global dataset is as follows:

[0019]

[0020] In the formula, Represents the global dataset. This represents the inflation pressure and attitude data from each air pump. This indicates the number of air pumps. This indicates the inflation pressure and attitude data collected by the main air pump itself. To obtain triaxial acceleration data from the air pump, To obtain the triaxial angular velocity data from the air pump, To obtain inflation pressure data from the air pump, The inflation pressure data for the main air pump. Indicates the main air pump. This indicates a number of sampling times corresponding to the preset sampling frequency.

[0021] Optionally, the main air pump uses attitude data from the real-time global dataset to calculate the attitude angle of the target inflatable tent. Based on the attitude angle calculation results and the inflation pressure data from the real-time global dataset, it determines whether the target inflatable tent is unbalanced. If so, it determines the unbalanced pressurization area and the unbalanced pressurization pump step, specifically including:

[0022] The main air pump uses attitude data from a real-time global dataset to calculate the attitude angles of each air pump in the target inflatable tent using the quaternion method; wherein, the attitude angles include pitch angle and roll angle.

[0023] The main air pump uses its own attitude angle as a reference to calculate the pitch angle deviation and roll angle deviation between each slave air pump and the main air pump. Combined with the current inflation pressure data of each slave air pump and the reference pressure when the inflation action is completed, it determines whether the target inflatable tent is unbalanced.

[0024] If so, output several unbalanced pressurization zones and unbalanced pressurization pumps of the target inflatable tent, and calculate the quantification coefficient of the degree of imbalance in each unbalanced pressurization zone.

[0025] Optionally, the main air pump utilizes attitude data from the real-time global dataset to calculate the attitude angle of each air pump in the target inflatable tent using the quaternion method. Specifically, this includes:

[0026] Define quaternions ,satisfy ;in, For the real part, This represents the imaginary part corresponding to rotation along the x, y, and z axes;

[0027] Using attitude data from the real-time global dataset, the quaternion is updated through the quaternion update equation, and the pitch and roll angles of the main air pump and each slave air pump are calculated through the attitude angle transformation equation.

[0028] The specific expressions for the quaternion update equation and attitude angle transformation equation are as follows:

[0029]

[0030]

[0031]

[0032] In the formula, This is the preprocessed angular velocity data. This is an antisymmetric matrix of angular velocity, used to characterize the effect of angular velocity on the update of quaternions, pitch angle. Corresponding to the longitudinal inclination of the steep slope, the roll angle Corresponding to the lateral impact of gusts.

[0033] Optionally, the main air pump uses its own attitude angle as a reference to calculate the pitch and roll angle deviations between each slave air pump and the main air pump. Combining this with the current inflation pressure data of each slave air pump and the reference pressure at the completion of the inflation action, a step is taken to determine whether the target inflatable tent is unbalanced. This step specifically includes:

[0034] The pitch angle of the main air pump and roll angle Based on this, the attitude angle deviation of each air pump is calculated; wherein, the expression for the attitude angle deviation is specifically:

[0035]

[0036]

[0037] In the formula, To reflect the difference in inclination along a steep slope by the pitch angle deviation between the air pump and the main air pump, The difference in the direction of gusts is reflected by the roll angle deviation between the air pump and the main air pump.

[0038] Preset pitch imbalance threshold and roll angle imbalance threshold If the air pump meets the imbalance condition, then the preset area of ​​the air pump is determined as the imbalance pressure replenishment area, and the air pump is determined as the imbalance pressure replenishment air pump.

[0039] The imbalance condition is configured as follows: or ,and , To establish the baseline pressure when the inflation process is completed.

[0040] Optionally, the expression for calculating the quantification coefficient of the imbalance degree in each imbalance compensation zone is as follows:

[0041]

[0042] In the formula, A coefficient quantifying the degree of imbalance. This indicates the pitch angle deviation from the air pump. This indicates the roll angle deviation from the air pump. This indicates the preset pitch angle imbalance threshold. This indicates the roll angle imbalance threshold.

[0043] Optionally, the main air pump constructs the imbalance parameters of the unbalanced pressurization pump as an input state vector, and uses a pre-built and trained improved DDPG model to infer the target pressurization pressure value. This process specifically includes:

[0044] The pitch angle deviation, roll angle deviation, imbalance degree quantification coefficient, inflation pressure data, and baseline pressure at the completion of the inflation operation of the unbalanced pressurization pump are integrated into an input state vector; the specific expression of the input state vector is as follows:

[0045]

[0046] In the formula, Represents the input state vector. This indicates the pitch angle deviation from the air pump. This indicates the roll angle deviation from the air pump. A quantification coefficient representing the degree of imbalance. This indicates the inflation pressure data from the air pump. This indicates the reference pressure at which the inflation process is completed;

[0047] The input state vector is input into an improved DDPG model Actor network pre-trained with simulation data of steep slope and gust wind scenarios. After being mapped by a multi-layer neural network of the Actor network, the initial target pressure compensation value is output. Specifically, the expression output by the improved DDPG model Actor network is as follows:

[0048]

[0049] In the formula, This indicates the initial target pressure value. Here are the weight matrix and bias vector from the input layer to hidden layer 1. Here are the weight matrix and bias vector from hidden layer 2 to the output layer. Use the Sigmoid activation function;

[0050] The expression for the contextualized reward function during training of the improved DDPG model is as follows:

[0051]

[0052] In the formula, This represents a contextualized reward function. To achieve the final target pressure value, , , These are the weighting coefficients, This is an indicator function.

[0053] Optionally, the target pressurization pressure value is converted into an inflation pressurization command and sent to each step of the inflation pump, specifically including:

[0054] The target pressure compensation value output by the Actor network is inversely normalized to convert it into an actual pressure value; wherein, the expression for the actual pressure value is as follows:

[0055]

[0056] In the formula, This represents the actual pressure value. , ;

[0057] The actual pressure value is calibrated by boundary calibration to obtain the final target pressure compensation value; wherein, the expression for the final target pressure compensation value is as follows:

[0058]

[0059] In the formula, This indicates the final target pressure compensation value. Then the calibration is ,like Then the calibration is ;

[0060] Based on the final target replenishment pressure value, the replenishment time is estimated, and the estimated replenishment time is sent as an inflation replenishment command to each corresponding air pump; wherein, the expression for the replenishment time is specifically:

[0061]

[0062] To determine the volume of the tent area corresponding to the air pump s, To achieve the final target pressure value, To obtain inflation pressure data from the air pump, This is the maximum inflation flow rate of the air pump. To improve the inflation efficiency of the air pump, When the air pump s is pressurized as estimated.

[0063] Furthermore, to achieve the above objectives, the present invention also provides an environment-adaptive inflation control system based on attitude analysis, comprising:

[0064] A module is established to establish a communication network for several air pumps; wherein, the several air pumps are configured to be installed at several air inlets of the target inflatable tent, and each air pump integrates a pressure sensor for collecting inflation pressure data and an attitude sensor for collecting attitude data.

[0065] The sending module is used to send the collected inflation pressure data and attitude data from each air pump to the main air pump. The main air pump establishes a real-time global dataset based on the collected and received inflation pressure data and attitude data.

[0066] The judgment module is used to calculate the attitude angle of the target inflatable tent by the main air pump using the attitude data in the real-time global dataset. Based on the attitude angle calculation result and the inflation pressure data in the real-time global dataset, it determines whether the target inflatable tent is unbalanced. If so, it determines the unbalanced pressure replenishment area and the unbalanced pressure replenishment air pump.

[0067] The inference module is used by the main air pump to construct the imbalance quantification parameters of the imbalance replenishment air pump into an input state vector, and use the pre-built and trained improved DDPG model to infer the target replenishment pressure value, and convert the target replenishment pressure value into an air replenishment command and send it to each slave air pump.

[0068] The adjustment module is used to perform supplementary inflation of the corresponding unbalanced pressure area from each air pump according to the inflation and pressure control parameters in the inflation and pressure replenishment command, so as to adjust the posture of the target inflatable tent.

[0069] The beneficial effects of this invention are as follows: It proposes an environment-adaptive inflation control method and system based on attitude analysis. By establishing a communication network between a main inflation pump and several slave inflation pumps, each inflation pump collects inflation pressure and attitude data at the corresponding inflation port position on the target inflatable tent. The main inflation pump performs attitude angle calculation and imbalance pressure replenishment area judgment based on the inflation pressure and attitude data it collects and receives from the outside via the communication network. Then, it constructs an input state vector using imbalance quantification parameters. The improved DDPG model, after training, infers the pressure replenishment value of each imbalance pressure replenishment inflation pump. This generates inflation pressure replenishment commands to drive the corresponding slave inflation pumps to perform pressure replenishment and inflation actions, thereby dynamically maintaining the attitude of the target inflatable tent. This improves the automation and intelligence of inflatable tent attitude maintenance in scenarios with steep slopes and gusts. While improving the accuracy and rationality of inflation pump pressure replenishment, it ensures the safety of inflatable tent use and reduces interference with outdoor operations. Attached Figure Description

[0070] Figure 1 This is a flowchart illustrating the environmental adaptive inflation control method based on attitude analysis of the present invention.

[0071] Figure 2 This is a schematic diagram of the environmental adaptive inflation control system based on attitude analysis of the present invention. Detailed Implementation

[0072] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0073] This invention provides an environment-adaptive inflation control method based on attitude analysis, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the environment-adaptive inflation control method based on attitude analysis according to an embodiment of the present invention.

[0074] In this embodiment, an environment-adaptive inflation control method based on attitude analysis is provided, the method comprising the following steps:

[0075] S1: Establish a communication network for several air pumps; wherein, the several air pumps are configured to be installed at several air inlets of the target inflatable tent, and each air pump integrates a pressure sensor for collecting inflation pressure data and an attitude sensor for collecting attitude data.

[0076] S2: Each air pump sends the collected inflation pressure data and attitude data to the main air pump. The main air pump establishes a real-time global dataset based on the collected and received inflation pressure data and attitude data.

[0077] S3: The main air pump uses the attitude data in the real-time global dataset to perform attitude angle calculation of the target inflatable tent. Based on the attitude angle calculation result and the inflation pressure data in the real-time global dataset, it determines whether the target inflatable tent is unbalanced. If so, it determines the unbalanced pressure replenishment area and the unbalanced pressure replenishment air pump.

[0078] S4: The main air pump constructs the imbalance quantification parameters of the imbalance replenishment air pump into an input state vector, uses the pre-built and trained improved DDPG model to infer the target replenishment pressure value, and converts the target replenishment pressure value into an air replenishment command and sends it to each slave air pump.

[0079] S5: Each air pump performs pressurization and inflation of the corresponding unbalanced pressurization area according to the pressurization and inflation control parameters in the pressurization and inflation command, so as to adjust the posture of the target inflatable tent.

[0080] It should be noted that existing solutions for maintaining the attitude of inflatable tents using air pumps have significant technical shortcomings and are difficult to meet the high stability requirements of tent attitude in outdoor operations: First, air pumps are usually only equipped with pressure sensors, lacking the ability to sense the overall tilt and attitude changes of the tent, and cannot determine the tilt direction and degree of imbalance of the tent under the combined conditions of steep slopes and gusts of wind; Second, air pumps in different locations passively inflate according to fixed pressure thresholds, and cannot adjust the pressure replenishment strategy in real time according to attitude changes caused by external forces, which is prone to insufficient or excessive pressure replenishment; Third, each air pump works independently and cannot share status data, and can only respond to its own pressure signal in isolation, which cannot form a global perspective and makes it difficult to accurately locate key pressure replenishment areas such as the lower part of steep slopes and the windward side of gusts; Fourth, existing air pumps cannot accurately distinguish between normal pressure decay and pressure changes caused by attitude imbalance, and are prone to accidental start-up or stoppage under instantaneous pressure fluctuations caused by gusts, further compromising the attitude stability of the tent.

[0081] To address the aforementioned issues, this embodiment establishes a communication network between a main air pump and several slave air pumps. Each air pump collects inflation pressure and attitude data from the corresponding inflation port on the target inflatable tent. The main air pump, based on its own collected inflation pressure and attitude data and data received from the network, performs attitude angle calculation and imbalance pressure replenishment area judgment. Then, using imbalance quantification parameters to construct an input state vector, an improved DDPG model is trained to infer the replenishment pressure value of each imbalance pressure replenishment air pump. This generates inflation pressure replenishment commands to drive the corresponding slave air pumps to perform replenishment inflation actions, dynamically maintaining the attitude of the target inflatable tent. This improves the automation and intelligence of inflatable tent attitude maintenance in scenarios involving steep slopes and gusts of wind. While enhancing the accuracy and rationality of air pump replenishment, it ensures the safety of inflatable tent use and reduces interference with outdoor operations.

[0082] In a preferred embodiment, the step of establishing a communication network for several air pumps specifically includes:

[0083] S11: After the air pumps set at several air ports complete the inflation action of the target inflatable tent, the main air pump sends a network establishment request to several slave air pumps.

[0084] S12: After receiving the network establishment request sent by the main air pump, several slave air pumps send network response information back to the main air pump. Based on the network response information, the main air pump establishes a communication network between the main air pump and several slave air pumps.

[0085] After the tent is erected and inflated, the main air pump actively sends a network establishment request to all slave air pumps. Upon receiving the request, each slave air pump responds with its own device ID and communication signal strength. After verifying the integrity of the response, the main air pump establishes a stable point-to-point communication link, forming a master-slave collaborative communication network. This network provides the communication foundation for global data acquisition, ensuring the main air pump can obtain status data from all slave air pumps, breaking down data silos. This allows the main air pump to perform joint analysis and reasoning based on data collected from several slave air pumps, outputting inflation and pressurization commands adapted to the current real-world environment, thus improving the accuracy and rationality of the air pump's pressurization.

[0086] It should be noted that the present invention only starts networking after the tent is set up and inflated, which can avoid energy waste during non-pressurization periods, so that the main air pump only starts to perform attitude maintenance inflation and pressurization control after the tent is set up.

[0087] In a preferred embodiment, each air pump sends collected inflation pressure data and attitude data to the main air pump, and the main air pump establishes a real-time global dataset based on the collected and received inflation pressure data and attitude data. This process specifically includes:

[0088] S21: Each air pump uses an integrated pressure sensor and attitude sensor to collect inflation pressure data and attitude data of the corresponding air inlet according to a preset sampling frequency; wherein, the attitude data includes triaxial acceleration data and triaxial angular velocity data;

[0089] S22: Each air pump sends the collected inflation pressure data and attitude data to the main air pump. The main air pump establishes a real-time global dataset based on the collected and received inflation pressure data and attitude data.

[0090] In this embodiment, each air pump synchronously collects attitude data and pressure data at a preset sampling frequency (e.g., 50Hz). The data from the air pump is packaged in a unified format and sent to the main air pump. After receiving all the data, the main air pump uses its own sampling timestamp as a reference to perform time-series alignment on the data from the air pumps, completes the deviation data, and finally constructs a real-time global dataset containing the status of all pumps.

[0091] In practical applications, the expression for the real-time global dataset is specifically as follows:

[0092]

[0093] In the formula, Represents the global dataset. This represents the inflation pressure and attitude data from each air pump. This indicates the number of air pumps. This indicates the inflation pressure and attitude data collected by the main air pump itself. To obtain triaxial acceleration data from the air pump, To obtain the triaxial angular velocity data from the air pump, To obtain inflation pressure data from the air pump, The inflation pressure data for the main air pump. Indicates the main air pump. This indicates a number of sampling times corresponding to the preset sampling frequency.

[0094] Specifically, the attitude tilt caused by steep slopes is directly reflected in the acceleration data, while the vibration caused by gusts is reflected in the angular velocity data. The main air pump is usually installed at the center of the tent, and its data serves as the benchmark for attitude analysis. It is used to compare the attitude deviation from the air pump (the center of the tent is relatively evenly affected by steep slopes and gusts, and its data can more accurately determine the imbalance state of the edge areas). The form of this real-time global dataset can adapt to the characteristics of synchronous changes in the state of various areas of the tent under steep slopes and gusts, ensuring that the main air pump can simultaneously acquire the attitude and pressure changes of all areas, avoiding analysis errors caused by data timing misalignment.

[0095] In a preferred embodiment, the main air pump uses attitude data from the real-time global dataset to calculate the attitude angle of the target inflatable tent. Based on the attitude angle calculation result and the inflation pressure data from the real-time global dataset, it determines whether the target inflatable tent is unbalanced. If so, it determines the unbalanced pressurization area and the unbalanced pressurization pump step, specifically including:

[0096] S31: The main air pump uses attitude data from the real-time global dataset to calculate the attitude angle of each air pump in the target inflatable tent according to the quaternion method; wherein, the attitude angle includes pitch angle and roll angle.

[0097] Specifically, the main air pump utilizes attitude data from a real-time global dataset to calculate the attitude angles of each air pump in the target inflatable tent using the quaternion method. This process includes:

[0098] S311: Define Quaternions ,satisfy ;in, For the real part, This represents the imaginary part corresponding to rotation along the x, y, and z axes;

[0099] S312: Using attitude data from the real-time global dataset, update the quaternion through the quaternion update equation, and solve the pitch and roll angles of the main air pump and each slave air pump through the attitude angle transformation equation.

[0100] In practical applications, the expressions for the quaternion update equation and attitude angle transformation equation are as follows:

[0101]

[0102]

[0103]

[0104] In the formula, This is the preprocessed angular velocity data. This is an antisymmetric matrix of angular velocity, used to characterize the effect of angular velocity on the update of quaternions, pitch angle. Corresponding to the longitudinal inclination of the steep slope, the roll angle Corresponding to the lateral impact of gusts.

[0105] In this embodiment, a quaternion vector is first defined, the quaternion is updated using angular velocity data, and then the quaternion is converted into a pitch angle representing the longitudinal tilt of the steep slope and a roll angle representing the lateral impact of the gust of wind through the attitude angle transformation equation, thus completing attitude quantization.

[0106] It should be noted that the introduction of the quaternion method allows for the description of three-dimensional spatial rotation using four-dimensional vectors, avoiding the gimbal lock problem of Euler angles during large-angle rotations. Furthermore, considering the longitudinal tilt of the tent caused by steep slopes (x-axis rotation) and the lateral tilt caused by gusts of wind (y-axis rotation), the imaginary part can be used... Precise capture lays the foundation for subsequent angle calculations. Dynamically updating quaternions based on angular velocity data can reflect real-time changes in the tent's attitude. Considering that instantaneous vibrations caused by gusts can lead to sudden changes in angular velocity data, the quaternion update equation can quickly update the quaternions, reflecting attitude changes in real time and avoiding analysis errors caused by update lag.

[0107] The formula for calculating the pitch angle uses a combination of the real and imaginary parts of quaternions to convert the rotation state into an angle value. Since the steeper the slope, the larger the absolute value of the pitch angle, this formula can accurately quantify the impact of steep slopes on the tent's attitude, providing a core indicator for imbalance judgment.

[0108] The formula for calculating the roll angle uses cross-operation of the real and imaginary parts of quaternions to calculate the lateral rotation angle. The greater the gust intensity, the larger the absolute value of the roll angle. This formula can capture the impact of gusts on the tent's attitude in real time, ensuring the timeliness of imbalance judgment.

[0109] Therefore, this embodiment uses the quaternion method to calculate the pitch and roll angles instead of the traditional Euler angle calculation, avoiding the gimbal lock problem when rotating at large angles, improving the accuracy of attitude angle calculation, and solving the problem of insufficient accuracy of traditional attitude calculation methods when the tent is tilted at large angles. This ensures that the tilt state of the tent can be accurately captured in steep slope and gust wind scenarios, providing reliable data support for imbalance judgment.

[0110] S32: The main air pump uses its own attitude angle as a reference to calculate the pitch angle deviation and roll angle deviation between each slave air pump and the main air pump. Combining the current inflation pressure data of each slave air pump with the reference pressure when the inflation action is completed, it determines whether the target inflatable tent is unbalanced.

[0111] Specifically, the main air pump uses its own attitude angle as a reference to calculate the pitch and roll angle deviations between each slave air pump and the main air pump. Combining this with the current inflation pressure data of each slave air pump and the reference pressure at the completion of the inflation process, it determines whether the target inflatable tent is unbalanced. This process includes:

[0112] S321: The pitch angle of the main air pump and roll angle Based on this, the attitude angle deviation of each air pump is calculated; wherein, the expression for the attitude angle deviation is specifically:

[0113]

[0114]

[0115] In the formula, To reflect the difference in inclination along a steep slope by the pitch angle deviation between the air pump and the main air pump, The difference in the direction of gusts is reflected by the roll angle deviation between the air pump and the main air pump.

[0116] S322: Preset pitch angle imbalance threshold and roll angle imbalance threshold If the air pump meets the imbalance condition, then the preset area of ​​the air pump is determined as the imbalance pressure replenishment area, and the air pump is determined as the imbalance pressure replenishment air pump.

[0117] The imbalance condition is configured as follows: or ,and , To establish the baseline pressure when the inflation process is completed.

[0118] In this embodiment, the pitch angle and roll angle of the main air pump are used as the reference to calculate the attitude angle deviation of each slave air pump. Then, by using a preset attitude angle imbalance threshold and combining the comparison between the current pressure of the slave air pump and the reference pressure, if the attitude angle deviation exceeds the standard and the pressure is lower than the reference, the area is determined to be an imbalance pressure replenishment area, and the corresponding slave air pump is an imbalance pressure replenishment air pump.

[0119] Specifically, absolute values ​​are used to calculate the pitch and roll angle differences between the auxiliary air pump and the main air pump, quantifying the tilt differences in the steep slope direction and the tilt differences caused by the lateral impact of gusts. This is because the pitch angle deviation of the auxiliary air pump below the steep slope will be significantly greater than in other areas. The attitude angle deviation formula can accurately capture this feature and provide a basis for locating the unbalanced area below the steep slope. The roll angle deviation of the auxiliary air pump on the windward side of the gust will be significantly increased. This formula can quickly locate the unbalanced area on the windward side and provide direction for targeted pressurization.

[0120] When setting the imbalance conditions, considering that tent imbalance will inevitably be accompanied by tilting and insufficient local pressure in steep slope and gusty wind scenarios, a dual check of abnormal posture and insufficient pressure is adopted to avoid misjudgment caused by a single posture deviation or a single insufficient pressure. This condition can accurately screen such areas and avoid false pressure supplementation to areas with normal posture but decreased pressure or tilted posture but sufficient pressure.

[0121] This embodiment proposes a deviation calculation method based on the attitude angle of the main air pump. It defines a dual imbalance condition of excessive attitude angle deviation and pressure below the benchmark, avoiding misjudgment caused by a single indicator. It solves the problem that traditional air pumps cannot distinguish between normal pressure decay and sudden pressure changes due to attitude imbalance, and achieves accurate screening of imbalance areas, ensuring that pressure replenishment is only applied to areas that truly need adjustment.

[0122] S33: If so, output several unbalanced pressurization areas and unbalanced pressurization pumps of the target inflatable tent, and calculate the quantification coefficient of the degree of imbalance of each unbalanced pressurization area.

[0123] Specifically, the expression for calculating the quantification coefficient of the imbalance degree in each imbalance compensation zone is as follows:

[0124]

[0125] In the formula, A coefficient quantifying the degree of imbalance. This indicates the pitch angle deviation from the air pump. This indicates the roll angle deviation from the air pump. This indicates the preset pitch angle imbalance threshold. This indicates the roll angle imbalance threshold.

[0126] Considering the dominance of imbalance on steep slopes, High proportion; when the gusts dominate and the balance is disrupted, With a high proportion, this embodiment quantifies the degree of imbalance by normalizing the pitch and roll angle deviations of the air pump with corresponding thresholds, and integrating the imbalance information from the two dimensions through square root operations to obtain a comprehensive imbalance degree quantification coefficient, thereby achieving the classification of the imbalance state.

[0127] This embodiment, by having both types of imbalance contribute during mixed imbalance, can comprehensively quantify different types of imbalance states, providing accurate imbalance degree input for the DDPG model. This solves the problem that traditional air pumps cannot quantify the degree of imbalance, enabling the pressurization strategy to be dynamically adjusted according to the severity of imbalance, avoiding insufficient or excessive pressurization.

[0128] In this embodiment, the main air pump uses attitude data from the global dataset to calculate the pitch and roll angles of each pump using the quaternion method; based on the attitude angle of the main air pump, the attitude angle deviation between the slave air pump and the main air pump is calculated; by comparing the current pressure of the slave air pump with the reference pressure, it is determined whether there is an imbalance; if there is an imbalance, the imbalance area and the corresponding slave air pump are output, and the imbalance degree quantification coefficient is calculated.

[0129] Therefore, by introducing the quaternion method into the tent attitude angle calculation and proposing a dual imbalance judgment mechanism of attitude angle deviation and pressure comparison, and combining the imbalance degree quantification coefficient, the imbalance area can be accurately located and classified, solving the problem that traditional air pumps cannot sense the tent attitude and cannot locate the imbalance area, providing a clear target for subsequent accurate pressure replenishment and avoiding blind pressure replenishment.

[0130] In a preferred embodiment, the main air pump constructs the imbalance quantification parameters of the imbalance replenishment air pump into an input state vector, and uses a pre-built and trained improved DDPG model to infer the target replenishment pressure value. This step specifically includes:

[0131] S41: Integrate the pitch angle deviation, roll angle deviation, imbalance degree quantification coefficient, inflation pressure data, and reference pressure at the completion of the inflation operation of the unbalanced pressurization pump into an input state vector; wherein, the specific expression of the input state vector is:

[0132]

[0133] In the formula, Represents the input state vector. This indicates the pitch angle deviation from the air pump. This indicates the roll angle deviation from the air pump. A quantification coefficient representing the degree of imbalance. This indicates the inflation pressure data from the air pump. This indicates the reference pressure at which the inflation process is completed;

[0134] S42: The input state vector is input into an improved DDPG model Actor network pre-trained with simulation data of steep slope and gust wind scenarios. After mapping by a multi-layer neural network of the Actor network, the initial target pressure compensation value is output; wherein, the expression output by the improved DDPG model Actor network is specifically as follows:

[0135]

[0136] In the formula, This indicates the initial target pressure value. Here are the weight matrix and bias vector from the input layer to hidden layer 1. Here are the weight matrix and bias vector from hidden layer 2 to the output layer. Use the Sigmoid activation function;

[0137] The expression for the contextualized reward function during training of the improved DDPG model is as follows:

[0138]

[0139] In the formula, This represents a contextualized reward function. To achieve the final target pressure value, , , These are the weighting coefficients, This is an indicator function.

[0140] In this embodiment, the pitch angle deviation, roll angle deviation, imbalance degree quantification coefficient, current pressure, and reference pressure of the unbalanced pressure replenishing air pump need to be integrated into an input state vector. Then, it is input into an improved DDPG model Actor network that has been pre-trained with scene data. Through multi-layer neural network mapping, a standardized preliminary target pressure replenishment value is output.

[0141] In this embodiment, the scenario data can be configured to cover scenario-based simulation data under conditions of steep slopes of 15°-25° and gusts of 10-18 m / s, combined with outdoor measured data, to construct an imbalance state-optimal pressure compensation sample. Using this scenario data, an Actor-Critic dual-network and target network architecture can be adopted. The state-action-reward-next state data is stored through empirical replay. After batch sampling, the Critic and Actor networks are updated using mean squared error and negative Q-value as loss functions, respectively. After training convergence, the model can accurately adapt to steep slope-dominated, gust-dominated, and mixed imbalance scenarios.

[0142] It should be noted that this embodiment integrates posture imbalance features and pressure features to form a comprehensive state description, ensuring that the model can capture the correlation between imbalance and pressure. The construction of the input state vector can completely describe the core state of the tent imbalance area under steep slopes and gusts of wind, enabling the model to generate differentiated pressure compensation for different situations such as steep slope-dominated imbalance, gust-dominated imbalance, and mixed imbalance. Subsequently, through the neural network structure of the Actor network with input layer-hidden layer-output layer, a nonlinear mapping from the state vector to the pressure compensation is realized. The ReLU activation function enhances the model's nonlinear fitting ability, and the Sigmoid activation function ensures that the output is in the [0,1] interval. This allows the model to learn the mapping rules between state and pressure compensation under steep slopes and gusts through scenario-based training, enabling it to quickly output a standardized pressure value adapted to the current imbalance state, with a response speed much faster than traditional fixed threshold decision-making.

[0143] For scenario-based reward functions, a penalty term guides the model's learning. The greater the imbalance and the higher the pressure bias, the lower the reward value, and the model will proactively optimize its strategy to increase the reward value. In practical applications, , , The first penalty is for the degree of imbalance; the more severe the imbalance, the lower the reward, guiding the model to prioritize solving severe imbalances. The second penalty is for pressure deviation, ensuring accurate pressure replenishment. The third penalty is for severe imbalance, incentivizing the model to respond quickly to severe imbalances.

[0144] Therefore, by designing a reward function to focus on penalizing severe imbalances and pressure deviations, the core needs of quickly correcting imbalances and accurately replenishing pressure in the scenario are met. This guides the model to prioritize solving severe imbalance problems caused by steep slopes and gusts of wind. By guiding network learning through a scenario-based reward function, the decision-making is ensured to adapt to the scenario requirements, thus solving the problem of rigidity in traditional air pump pressure replenishment strategies. Dynamic pressure replenishment decisions are achieved through a deep learning model, enabling the pressure replenishment to adapt to the changes in imbalance state under steep slopes and gusts of wind in real time.

[0145] In an optional embodiment, considering the differences in specifications, dimensions, layout, and inflation port locations of various inflatable tents, the appropriate improved DDPG model should be selected based on the actual conditions of the tent. In practice, a large number of improved DDPG models for different types of inflatable tents, pre-trained using simulation data of steep slopes and gusts, can be stored in the main air pump, a smart terminal (e.g., a mobile phone) connected to the main air pump, or even the cloud. When the user actually uses the model, they can input the required type, and the improved DDPG model matching the current tent condition can be retrieved from the model library through a matching method, allowing for direct use without retraining.

[0146] In a preferred embodiment, converting the target replenishment pressure value into an inflation replenishment command and sending it to each step of the inflation pump specifically includes:

[0147] S43: The target pressure compensation value output by the Actor network is inversely normalized to convert it into an actual pressure value; wherein, the expression for the actual pressure value is as follows:

[0148]

[0149] In the formula, This represents the actual pressure value. , .

[0150] The operating principle of the standardized formula is to convert the standardized initial target pressure ([0,1] interval) into the actual pressure unit (kPa), making it executable. , The settings are adapted to the tent material's tolerance, avoiding over- or under-pressure caused by abnormal model output, and ensuring the safety of pressure application in steep slope and gusty wind scenarios.

[0151] S44: The actual pressure value is calibrated through boundary calibration to obtain the final target pressure compensation value; wherein, the expression for the final target pressure compensation value is as follows:

[0152]

[0153] In the formula, This indicates the final target pressure compensation value. Then the calibration is ,like Then the calibration is .

[0154] Regarding the boundary calibration formula, in steep slope and gusty wind scenarios, the model may output extreme values ​​due to instantaneous strong disturbances. This formula effectively avoids such risks, protecting the tent fabric from excessive stretching, while ensuring that the pressure compensation reaches the minimum effective pressure. Through dual extreme value calculations, the inversely standardized pressure value is limited to […]. , This area forms the last line of defense for safety.

[0155] S45: Based on the final target replenishment pressure value, estimate the replenishment time, and send the estimated replenishment time as an inflation replenishment command to each corresponding air pump; wherein, the expression for the replenishment time is specifically:

[0156]

[0157] To determine the volume of the tent area corresponding to the air pump s, To achieve the final target pressure value, To obtain inflation pressure data from the air pump, This is the maximum inflation flow rate of the air pump. To improve the inflation efficiency of the air pump, When the air pump s is pressurized as estimated.

[0158] For the calculation of pressurization time, a simplified model based on the ideal gas law is used, taking into account the tent area volume, pressure difference, air pump flow rate and efficiency, to calculate the required pressurization time. This provides a time constraint for pressurization execution, avoids over-pressurization caused by continuous inflation, and allows the main air pump to predict the pressurization completion time, adapting to the time management requirements of dynamic pressurization in the scenario.

[0159] In this embodiment, the standardized initial target pressure output by the Actor network is first inversely normalized to convert it into an actual pressure value. Then, boundary calibration is used to limit the actual pressure value within a safe range, yielding the final target repressurization pressure. Finally, based on the tent area volume, the pressure difference between the target pressure and the current pressure, and the air pump flow rate and efficiency, the repressurization time is estimated, providing a time reference for repressurization execution. Therefore, a pressure processing flow based on inverse normalization and boundary calibration is proposed. This ensures that the target repressurization pressure remains within a safe range. Simultaneously, a repressurization time estimation formula based on tent volume, air flow rate, and efficiency provides a time constraint for the repressurization process. This solves the problems of traditional air pump repressurization pressure lacking safety constraints and being uncontrollable, ensuring that repressurization is effective without damaging the tent, while avoiding over-repressurization and improving the controllability of the repressurization process.

[0160] Reference Figure 2 , Figure 2 This is a schematic diagram of the environmental adaptive inflation control system based on attitude analysis, according to an embodiment of the present invention.

[0161] like Figure 2 As shown, the environmental adaptive inflation control system based on attitude analysis proposed in this embodiment of the invention includes:

[0162] Module 10 is established to establish a communication network for several air pumps; wherein, the several air pumps are configured to be installed at several air inlets of the target inflatable tent, and each air pump integrates a pressure sensor for collecting inflation pressure data and an attitude sensor for collecting attitude data.

[0163] The sending module 20 is used to send the collected inflation pressure data and attitude data from each air pump to the main air pump. The main air pump establishes a real-time global dataset based on the collected and received inflation pressure data and attitude data.

[0164] The judgment module 30 is used to perform attitude angle calculation of the target inflatable tent using the attitude data in the real-time global dataset by the main air pump. Based on the attitude angle calculation result and the inflation pressure data in the real-time global dataset, it determines whether the target inflatable tent is unbalanced. If so, it determines the unbalanced pressure replenishment area and the unbalanced pressure replenishment air pump.

[0165] The reasoning module 40 is used by the main air pump to construct the imbalance quantification parameters of the imbalance replenishment air pump into an input state vector, and use the pre-built and trained improved DDPG model to reason to obtain the target replenishment pressure value, and convert the target replenishment pressure value into an air replenishment command and send it to each slave air pump.

[0166] The adjustment module 50 is used to perform supplementary inflation of the corresponding unbalanced pressure area from each air pump according to the inflation and pressure control parameters in the inflation and pressure supplementation command, so as to adjust the posture of the target inflatable tent.

[0167] Other embodiments or specific implementations of the environmental adaptive inflation control system based on attitude analysis of the present invention can be found in the above-described method embodiments, and will not be repeated here.

[0168] It is understood that in the description of this specification, references to terms such as "one embodiment," "another embodiment," "other embodiments," or "first embodiment to Nth embodiment," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0169] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0170] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for environment adaptive inflation control based on posture analysis, characterized in that, The method comprises the following steps: A communication network of a plurality of air pumps is established; wherein the plurality of air pumps are configured to be arranged at a plurality of air inlets of a target inflatable tent, each air pump is integrated with a pressure sensor for collecting air inflation pressure data and an attitude sensor for collecting attitude data; Each slave air pump sends the collected air inflation pressure data and attitude data to the master air pump, and the master air pump establishes a real-time global data set based on the collected and received air inflation pressure data and attitude data; The master air pump uses the attitude data in the real-time global data set to perform attitude angle calculation of the target inflatable tent, and judges whether the target inflatable tent is unbalanced according to the attitude angle calculation result and the air inflation pressure data in the real-time global data set, if yes, determines the unbalanced pressure compensation area and the unbalanced pressure compensation air pump; specifically comprising: The master air pump uses the attitude data in the real-time global data set to calculate the attitude angle of each air pump of the target inflatable tent according to the quaternion method; wherein the attitude angle includes the pitch angle and the roll angle; The master air pump takes its own attitude angle as the reference to calculate the pitch angle deviation and the roll angle deviation of each slave air pump and the master air pump, and combines the current air inflation pressure data of each slave air pump and the reference pressure when the air inflation action is completed to judge whether the target inflatable tent is unbalanced; specifically comprising: The pitch angle of the main air pump and roll angle Based on this, the attitude angle deviation of each air pump is calculated; wherein, the expression for the attitude angle deviation is specifically: wherein is the deviation in pitch from the inflation pump to the main inflation pump, reflecting the difference in steep slope direction inclination, is the deviation in roll from the inflation pump to the main inflation pump, reflecting the difference in gust direction inclination; A preset pitch angle imbalance threshold and a roll angle imbalance threshold If the slave inflator satisfies the imbalance condition, the slave inflator is determined as an imbalance pressure compensation inflator, and the area to which the slave inflator belongs is determined as an imbalance pressure compensation area. wherein the unbalance condition is configured to: or , and , is the reference pressure at the time of building up the pneumatic action. If yes, output the plurality of unbalanced pressure compensation areas and the unbalanced pressure compensation air pump of the target inflatable tent, and calculate the unbalanced degree quantization coefficient of each unbalanced pressure compensation area; The master air pump constructs the unbalanced quantization parameter of the unbalanced pressure compensation air pump into an input state vector, uses the improved DDPG model which is constructed and trained in advance to infer to obtain a target pressure compensation value, and converts the target pressure compensation value into an air pressure compensation instruction and sends it to each slave air pump; Each slave air pump executes pressure compensation inflation of the corresponding unbalanced pressure compensation area according to the air pressure compensation control parameter in the air pressure compensation instruction to adjust the attitude of the target inflatable tent.

2. The method of claim 1, wherein, The step of establishing a communication network of a plurality of air pumps comprises: After the air pumps arranged at the plurality of air inlets complete the air inflation action of building the target inflatable tent, the master air pump sends a network establishment request to the plurality of slave air pumps; After receiving the network establishment request sent by the master air pump, the plurality of slave air pumps feed back network response information to the master air pump, and the master air pump establishes a communication network of the master air pump and the plurality of slave air pumps according to the network response information.

3. The method of claim 1, wherein, The step of each slave air pump sending the collected air inflation pressure data and attitude data to the master air pump, and the master air pump establishing a real-time global data set based on the collected and received air inflation pressure data and attitude data comprises: Each air pump uses the integrated pressure sensor and attitude sensor to collect air inflation pressure data and attitude data of the corresponding air inlet at a preset sampling frequency; wherein the attitude data includes three-axis acceleration data and three-axis angular velocity data; Each slave air pump sends the collected air inflation pressure data and attitude data to the master air pump, and the master air pump establishes a real-time global data set based on the collected and received air inflation pressure data and attitude data; The expression of the real-time global data set is specifically as follows: ； In the formula, represents the global data set, represents the inflation pressure data and attitude data of each slave inflation pump, represents the number of slave inflation pumps, represents the inflation pressure data and attitude data collected by the master inflation pump itself, is the three-axis acceleration data of the slave inflation pump, is the three-axis angular velocity data of the slave inflation pump, is the inflation pressure data of the slave inflation pump, is the inflation pressure data of the master inflation pump, represents the master inflation pump, represents a plurality of sampling times corresponding to a preset sampling frequency.

4. The method of claim 1, wherein, The main air pump utilizes attitude data from a real-time global dataset to calculate the attitude angles of each air pump in the target inflatable tent using the quaternion method. Specifically, this includes: Defining quaternions , satisfying ; wherein, is a real part, is an imaginary part corresponding to x, y, z axis rotations; Using attitude data from the real-time global dataset, the quaternion is updated through the quaternion update equation, and the pitch and roll angles of the main air pump and each slave air pump are calculated through the attitude angle transformation equation. The specific expressions for the quaternion update equation and attitude angle transformation equation are as follows: wherein is the pre-processed angular velocity data, is the angular velocity antisymmetric matrix, which is used to characterize the angular velocity's effect on the quaternion update, pitch angle corresponds to steep slope longitudinal inclination, roll angle corresponds to crosswind lateral impact.

5. The method of claim 4, wherein, The specific expression for calculating the quantification coefficient of the imbalance degree in each imbalance compensation zone is as follows: In the formula, is a quantization coefficient of the imbalance degree, represents a deviation from the pitch angle of the inflation pump, represents a deviation from the roll angle of the inflation pump, represents a preset pitch angle imbalance threshold value, represents a roll angle imbalance threshold value.

6. The method of claim 5, wherein, The main air pump constructs the imbalance parameters of the imbalance replenishment air pump into an input state vector, and uses a pre-built and trained improved DDPG model to infer the target replenishment pressure value. The specific steps include: The pitch angle deviation, roll angle deviation, imbalance degree quantification coefficient, inflation pressure data, and baseline pressure at the completion of the inflation operation of the unbalanced pressurization pump are integrated into an input state vector; the specific expression of the input state vector is as follows: ； wherein, represents an input state vector, represents a pitch angle deviation from the inflator pump, represents a roll angle deviation from the inflator pump, represents an unbalance degree quantification coefficient, represents the inflator pressure data from the inflator pump, represents a reference pressure at the time of completion of the erection inflation action; The input state vector is input into an improved DDPG model Actor network pre-trained with simulation data of steep slope and gust wind scenarios. After being mapped by a multi-layer neural network of the Actor network, the initial target pressure compensation value is output. Specifically, the expression output by the improved DDPG model Actor network is as follows: wherein, represents the preliminary target pressure compensation pressure value, is the weight matrix and bias vector from the input layer to the hidden layer 1, is the weight matrix and bias vector from the hidden layer 2 to the output layer, is the Sigmoid activation function; The expression for the contextualized reward function during training of the improved DDPG model is as follows: In the formula, represents a scenario reward function, is a final target compensation pressure value, , , are weight coefficients, respectively, is an indicator function.

7. The method of claim 1, wherein, The target pressurization pressure value is converted into an inflation pressurization command and sent to each step of the inflation pump, specifically including: The target pressure compensation value output by the Actor network is inversely normalized to convert it into an actual pressure value; wherein, the expression for the actual pressure value is as follows: In the formula, represents the actual pressure value, , ; The actual pressure value is calibrated by boundary calibration to obtain the final target pressure compensation value; wherein, the expression for the final target pressure compensation value is as follows: ； wherein represents the final target pressure compensation value, if then the calibration is if then the calibration is ; Based on the final target replenishment pressure value, the replenishment time is estimated, and the estimated replenishment time is sent as an inflation replenishment command to each corresponding air pump; wherein, the expression for the replenishment time is specifically: for a volume of a corresponding tent area from the air pump s, for a final target pressure value, for air pressure data from the air pump s, for a maximum inflation flow rate of the air pump s, for an inflation efficiency of the air pump s, for an estimated pressure makeup time from the air pump s.

8. An environment adaptive inflation control system based on posture analysis, characterized in that, include: A module is established to establish a communication network for several air pumps; wherein, the several air pumps are configured to be installed at several air inlets of the target inflatable tent, and each air pump integrates a pressure sensor for collecting inflation pressure data and an attitude sensor for collecting attitude data. The sending module is used to send the collected inflation pressure data and attitude data from each air pump to the main air pump. The main air pump establishes a real-time global dataset based on the collected and received inflation pressure data and attitude data. The judgment module is used by the main air pump to calculate the attitude angle of the target inflatable tent using attitude data from the real-time global dataset. Based on the attitude angle calculation results and the inflation pressure data from the real-time global dataset, it determines whether the target inflatable tent is unbalanced. If so, it identifies the unbalanced pressurization area and the unbalanced pressurization pump. Specifically, it includes: The main air pump uses attitude data from a real-time global dataset to calculate the attitude angles of each air pump in the target inflatable tent using the quaternion method; wherein, the attitude angles include pitch angle and roll angle. The master inflation pump calculates the pitch angle deviation and roll angle deviation of each slave inflation pump from the master inflation pump based on the attitude angle of the master inflation pump, and judges whether the target inflatable tent is unbalanced by combining the current inflation pressure data of each slave inflation pump and the reference pressure when the erection inflation action is completed; specifically comprising: The pitch angle of the main air pump and roll angle Based on this, the attitude angle deviation of each air pump is calculated; wherein, the expression for the attitude angle deviation is specifically: wherein is the deviation in pitch from the inflation pump to the main inflation pump, reflecting the difference in steep slope direction inclination, is the deviation in roll from the inflation pump to the main inflation pump, reflecting the difference in gust direction inclination; A preset pitch angle imbalance threshold and a roll angle imbalance threshold If the slave inflator satisfies the imbalance condition, the slave inflator is determined as an imbalance pressure compensation inflator, and the area to which the slave inflator belongs is determined as an imbalance pressure compensation area. wherein the unbalance condition is configured to: or , and , is the reference pressure at the time of building up the pneumatic action; If so, output several unbalance compensation pressure areas and unbalance compensation inflation pumps of the target inflatable tent, and calculate the unbalance degree quantification coefficient of each unbalance compensation pressure area; The reasoning module is used for the master inflation pump to construct the unbalance quantification parameters of the unbalance compensation inflation pump into an input state vector, use the improved DDPG model which is constructed and trained in advance, reason to obtain a target compensation pressure value, and convert the target compensation pressure value into an inflation compensation instruction and send it to each slave inflation pump; The adjustment module is used for each slave inflation pump to perform compensation inflation on the corresponding unbalance compensation pressure area according to the inflation compensation control parameters in the inflation compensation instruction, so as to adjust the attitude of the target inflatable tent.

Images