Electric hydrofoil surfboard control method and device, computer equipment and storage medium

Abstract

The application belongs to the technical field of automatic control, and relates to a control method and device for an electric hydrofoil surfboard, computer equipment and a storage medium, which comprises the following steps: determining a current state prediction value of the surfboard according to a preset dynamic model; obtaining current sensing data of the surfboard, fusing the current sensing data and the current state prediction value to obtain a state fusion value; taking the state fusion value as an initial state of the surfboard, taking the dynamic model as a prediction benchmark of the state of the surfboard, and predicting a plurality of state sequences of the surfboard within a preset time period; screening the plurality of state sequences according to a preset flight boundary to obtain a target state sequence; determining a current control parameter corresponding to the target state sequence, and controlling the surfboard according to the current control parameter. The application improves the dynamic stability of the electric hydrofoil surfboard under multi-working-condition interference.

Description

Technical Field

[0001] This application relates to the field of automatic control technology, and in particular to a control method, device, computer equipment, and storage medium for an electric hydrofoil surfboard. Background Technology

[0002] With the rapid development of electrification and intelligent technology in water sports, electric hydrofoil surfboards, as a new type of water vehicle, are gradually gaining popularity among users. They use hydrofoils to generate lift, allowing the board to leave the water and achieve low-resistance gliding. However, real open water environments present various disturbances such as waves and currents, placing higher demands on the dynamic stability of the board.

[0003] Currently, most mainstream electric hydrofoil surfboards adopt a fixed parameter electronic control mode, which is combined with basic sensor equipment to collect data. During operation, the user mainly relies on adjusting their own center of gravity and using a simple control structure to maintain the board's posture and gliding height.

[0004] This type of control method lacks flexibility and struggles to adapt quickly to various conditions such as wave impact, user changes, and environmental disturbances. The board's pitch and tilt, as well as its height above the water, tend to fluctuate significantly, greatly reducing the gliding experience and increasing the risk of loss of control. Therefore, the lack of dynamic stability under multiple conditions is a key technical issue hindering the safety and user experience of electric hydrofoil surfboards.

[0005] In view of the above, this application is hereby submitted. Summary of the Invention

[0006] The purpose of this application is to provide a control method, device, computer equipment, and storage medium for an electric hydrofoil surfboard, in order to solve the technical problem of the lack of dynamic stability of the electric hydrofoil surfboard under multiple operating conditions.

[0007] To address the aforementioned technical problems, this application provides a control method for an electric hydrofoil surfboard, employing the following technical solution: The current state prediction value of the surfboard is determined based on a preset dynamic model; Obtain the current sensing data of the surfboard, and fuse the current sensing data with the current state prediction value to obtain a state fusion value; The state fusion value is used as the initial state of the surfboard, and the dynamic model is used as the prediction benchmark for the state of the surfboard to predict multiple state sequences of the surfboard within a preset time period. Based on the preset flight boundary, multiple state sequences are filtered to obtain the target state sequence; Determine the current control parameters corresponding to the target state sequence, and control the surfboard according to the current control parameters.

[0008] Furthermore, determining the predicted current state of the surfboard based on a preset dynamic model includes: Obtain the historical state fusion value and historical control parameters of the surfboard; The historical state fusion value and the historical control parameters are input into the dynamic model. The dynamic model then performs state recursion based on the historical state fusion value and the historical control parameters to output the current state prediction value.

[0009] Furthermore, the historical state fusion values ​​include historical height above water, historical pitch angle, historical forward velocity, and historical forward acceleration, and the historical control parameters include historical control surface angle and historical motor thrust. The step of using the dynamic model to perform state recursion based on the historical state fusion value and the historical control parameters, and outputting the current state prediction value, includes: The angle of attack of the surfboard hydrofoil is calculated based on the historical pitch angle, the historical forward velocity, and the historical control surface angle. The lift of the surfboard hydrofoil is calculated based on the angle of attack and the historical forward velocity. The vertical acceleration of the surfboard is calculated based on the hydrofoil lift and preset mass parameters. The pitch acceleration of the surfboard is calculated based on the hydrofoil lift, the historical motor thrust, and preset pitch motion parameters. The predicted current state value is calculated based on the historical lift-off height, vertical acceleration, historical pitch angle, pitch angle acceleration, historical forward velocity, historical forward acceleration, historical motor thrust, hydrofoil lift, and preset control cycle duration. The predicted current state value includes predicted lift-off height, predicted pitch angle, predicted forward velocity, and predicted forward acceleration.

[0010] Furthermore, the step of using the state fusion value as the initial state of the surfboard and the dynamic model as the prediction benchmark for the surfboard's state to predict multiple state sequences of the surfboard within a preset time period includes: The state fusion value is used as the initial state of the surfboard, and the dynamic model is used as the prediction benchmark for the surfboard state to generate multiple candidate control sequences within the preset time period. In chronological order, multiple candidate control sequences and the initial state are substituted into the dynamic model for recursion to obtain the predicted state sequence corresponding to each candidate control sequence.

[0011] Furthermore, the step of filtering multiple state sequences according to a preset flight boundary to obtain a target state sequence includes: Detect whether each state parameter in all the state sequences is within the flight boundary, and use the state sequences that are all within the flight boundary as candidate state sequences; If there are multiple candidate state sequences, the target state sequence is determined from the multiple candidate state sequences according to a preset optimization objective function.

[0012] Furthermore, determining the target state sequence from the multiple candidate state sequences according to a preset optimization objective function includes: Obtain the desired state sequence of the surfboard within the preset time period; Calculate the weighted deviation between each candidate state sequence and the desired state sequence according to the optimization objective function, and obtain the deviation cost corresponding to each candidate state sequence; The candidate state sequence with the smallest deviation cost is selected as the target state sequence.

[0013] Furthermore, fusing the current perceived data and the current state prediction value to obtain a state fusion value includes: Calculate the first covariance of the current state prediction value and the second covariance of the current sensed data, and determine the fusion weight of the current state prediction value and the current sensed data based on the first covariance and the second covariance; The current state prediction value and the current perceived data are weighted and fused according to the fusion weight to obtain the state fusion value.

[0014] To address the aforementioned technical problems, this application also provides an electric hydrofoil surfboard control device, which employs the following technical solution: An electric hydrofoil surfboard control device, comprising: The determination module is used to determine the predicted current state of the surfboard based on a preset dynamic model; The fusion module is used to acquire the current sensing data of the surfboard, fuse the current sensing data and the current state prediction value to obtain a state fusion value; The prediction module is used to use the state fusion value as the initial state of the surfboard, use the dynamic model as the prediction benchmark for the state of the surfboard, and predict multiple state sequences of the surfboard within a preset time period. The filtering module is used to filter multiple state sequences according to preset flight boundaries to obtain a target state sequence; The control module is used to determine the current control parameters corresponding to the target state sequence, and control the surfboard according to the current control parameters.

[0015] To address the aforementioned technical problems, this application also provides a computer device that employs the following technical solution: A computer device includes a memory and a processor, the memory storing computer-readable instructions, the processor executing the computer-readable instructions to implement the steps of the electric hydrofoil surfboard control method as described above.

[0016] To address the aforementioned technical problems, this application also provides a computer-readable storage medium, employing the technical solution described below: A computer-readable storage medium storing computer-readable instructions that, when executed by a processor, implement the steps of the electric hydrofoil surfboard control method as described above.

[0017] Compared with the prior art, this application has the following main advantages: The electric hydrofoil surfboard control method disclosed in this application determines the current state prediction value through a preset dynamic model, providing a priori basis for surfboard state estimation. Acquiring current sensing data and fusing it with the state prediction value can suppress the interference of single sensor measurement noise on the state estimation, making the fused state value closer to the actual motion state of the surfboard. Using the fused state value as the initial state and predicting multiple future state sequences using the dynamic model allows for the prediction of various possible evolution trends before the actual surfboard motion deviates, avoiding state estimation lagging behind actual changes due to sudden external disturbances. Filtering the state sequences based on flight boundaries can eliminate flight paths with risks of water contact or capsizing, ensuring that the remaining state sequences are all within a safe motion range. Determining the control parameters corresponding to the filtered target state sequences and using them to control the surfboard translates the safety and optimality at the state level into specific instructions at the actuator level, ensuring that the output of the motors and servos matches the current flight requirements. This allows the surfboard to select a safe path based on the most reliable current status information when facing different water conditions and operating instructions, and to readjust subsequent actions at every moment based on the latest sensing data, thereby maintaining the stability of the lift-off height and pitch attitude within the allowable range. Attached Figure Description

[0018] To more clearly illustrate the solutions in this application, the accompanying drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is an exemplary system architecture diagram to which this application can be applied; Figure 2 This is a flowchart of one embodiment of the electric hydrofoil surfboard control method according to this application; Figure 3 This is a schematic diagram of one embodiment of the electric hydrofoil surfboard control device according to this application; Figure 4 This is a schematic diagram of the structure of one embodiment of the computer device according to this application. Detailed Implementation

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein in the specification of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings of this application are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings of this application are used to distinguish different objects, not to describe a particular order.

[0021] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0022] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0023] like Figure 1 As shown, the system architecture 100 may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 serves as a medium for providing communication links between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0024] Users can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 via the network 104 to receive or send messages, etc. Various communication client applications can be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0025] The first terminal device 101, the second terminal device 102, and the third terminal device 103 can be various electronic devices with displays and support web browsing, including but not limited to smartphones, tablets, e-book readers, MP3 (Moving Picture Experts Group Audio Layer Ⅲ) players, MP4 (Moving Picture Experts Group Audio Layer IV) players, laptops, and desktop computers, etc.

[0026] Server 105 can be a server that provides various services, such as a backend server that supports the pages displayed on the first terminal device 101, the second terminal device 102, and the third terminal device 103.

[0027] It should be noted that the electric hydrofoil surfboard control method provided in this application embodiment is generally executed by a terminal device, and correspondingly, the electric hydrofoil surfboard control device is generally installed in the terminal device.

[0028] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0029] Continue to refer to Figure 2 A flowchart of an embodiment of the electric hydrofoil surfboard control method according to this application is shown. The electric hydrofoil surfboard control method includes the following steps: Step S201: Determine the predicted current state of the surfboard based on the preset dynamic model.

[0030] In this embodiment, the electric hydrofoil surfboard control method operates on electronic devices (e.g., Figure 1The terminal device shown can send or receive data via wired or wireless connection. It should be noted that the aforementioned wireless connection methods may include, but are not limited to, 3G / 4G / 5G connections, Wi-Fi connections, Bluetooth connections, WiMAX connections, Zigbee connections, UWB (ultra-wide band) connections, and other currently known or future wireless connection methods.

[0031] In this embodiment, the preset dynamic model is a discrete-time nonlinear state prediction model used to describe the dynamic mapping relationship between the surfboard's state vector and the control input vector. Its basic mathematical form is: That is: the state fusion value of the previous moment and the control parameters of the previous moment As input, through the nonlinear function f( Perform recursion and output the predicted state value at the current time. ,in The system uses a pre-defined set of model parameters (including water density, effective hydrofoil area, hydrofoil mounting angle, mass parameters, moment of inertia, lift coefficient curve, etc.). Based on the surfboard's current state (e.g., altitude, pitch angle, velocity) and the currently applied control (e.g., thrust, rudder angle), the dynamic model recursively calculates the predicted state of the surfboard at the next moment using hydrodynamic and rigid body dynamics formulas. Specifically, the fused historical state values ​​and the historical control parameters are input into the dynamic model, which then performs state recursion based on these values ​​and outputs the predicted state value for the current moment. Specifically, the dynamic model first calculates the angle of attack of the surfboard hydrofoil based on the historical pitch angle, the historical forward velocity, and the historical control surface angle, and then calculates the hydrofoil lift based on the angle of attack and the historical forward velocity. Next, based on the hydrofoil lift, preset mass parameters, and preset pitch motion parameters, it calculates the vertical acceleration and pitch acceleration of the surfboard, respectively. Finally, based on the historical lift-off height, vertical acceleration, historical pitch angle, pitch acceleration, historical forward velocity, historical forward acceleration, and preset control cycle duration, it recursively obtains the predicted lift-off height, predicted pitch angle, predicted forward velocity, and predicted forward acceleration for the current moment, and combines them as the current state prediction value for output.

[0032] Step S202: Obtain the current perception data of the surfboard, and fuse the current perception data and the current state prediction value to obtain a state fusion value.

[0033] In this embodiment, multiple sensors deployed on the surfboard collect raw measurement signals, including the distance signal from the bottom of the surfboard to the water surface collected by an ultrasonic ranging sensor or a lidar ranging sensor, recorded as the raw height measurement value; the three-axis acceleration and three-axis angular velocity signals of the surfboard collected by an inertial measurement unit (IMU), from which the raw pitch angle measurement value and the raw forward acceleration measurement value are extracted; and the velocity and position signals of the surfboard collected by a Global Navigation Satellite System (GNSS) module, from which the raw forward velocity measurement value is extracted. The collected raw measurement signals are converted into a perception vector that matches the dimension of the state vector. The synchronous measurement values ​​of each sensor are then standardized in units and transformed in coordinates to ensure that their physical units are consistent with the physical units of the parameters in the state vector, thus obtaining the height measurement value. Pitch angle measurement value Forward velocity measurement value and forward acceleration measurement value The measured values ​​processed as described above are combined in a preset order to form the current sensing data. The vector representation is .

[0034] Based on the first covariance matrix of the current state prediction value and the second covariance matrix of the current sensed data, a Kalman gain matrix is ​​calculated. This Kalman gain matrix is ​​used to determine the weight allocation of the current state prediction value and the current sensed data during fusion. Specifically, when the first covariance matrix is ​​relatively large (i.e., the greater the uncertainty of the prediction value), the Kalman gain matrix is ​​correspondingly larger, and the weight of the current sensed data in the fusion is greater; when the second covariance matrix is ​​relatively large (i.e., the greater the measurement noise), the Kalman gain matrix is ​​correspondingly smaller, and the weight of the current state prediction value in the fusion is greater. The first covariance matrix is ​​recursively obtained by passing the covariance matrix of the previous state fusion value through the Jacobian matrix of the dynamic model and superimposing a preset process noise covariance matrix. The second covariance matrix is ​​the measurement noise covariance matrix of the current sensed data, which is preset based on sensor factory calibration data or static test results. Finally, the residual between the current sensed data and the current state prediction value is calculated. The residual is weighted using the Kalman gain matrix as the fusion weight, and the weighted residual is superimposed on the current state prediction value to obtain the state fusion value.

[0035] In addition, high-pass and low-pass filters can be designed to fuse signals with higher built-in confidence in different frequency domains. Specifically, the current sensing data (such as high-frequency signals measured by accelerometers) is processed by a high-pass filter to extract high-frequency components, and the current state prediction value (such as low-frequency signals recursively derived from a dynamic model) is processed by a low-pass filter to extract low-frequency components. Then, the high-frequency and low-frequency components are superimposed to obtain the state fusion value.

[0036] Step S203: Use the state fusion value as the initial state of the surfboard, use the dynamic model as the prediction benchmark for the state of the surfboard, and predict multiple state sequences of the surfboard within a preset time period.

[0037] In this embodiment, the physical limits of the surfboard's actuators are obtained, and the feasible space of the control quantity is determined based on these limits. Multiple candidate control sequences are generated within this feasible space. Each candidate control sequence contains the control parameters corresponding to each moment of the surfboard within a preset time period. The candidate control sequences are generated by superimposing random perturbations on the optimal control sequence of the surfboard at the previous moment, or by random sampling or grid sampling within the feasible space. For each candidate control sequence, the state fusion value is used as the initial state. Following the chronological order, the current state and the control parameters at the current moment are successively substituted into the dynamic model to recursively obtain the predicted state for the next moment, until the end of the preset time period is reached. The initial state and the recursively obtained predicted states at each moment are then concatenated chronologically to obtain the predicted state sequence corresponding to that candidate control sequence. The above operation is repeated for each candidate control sequence to obtain multiple predicted state sequences.

[0038] In addition, starting with the state fusion value, multiple possible state increments are directly sampled within the state transition probability distribution of the surfboard, thus generating multiple branch state sequences. For example, the mean of the state transition is determined based on the dynamic model, and random perturbations conforming to a preset distribution are superimposed near the mean to generate multiple possible state evolution paths; or, based on the Monte Carlo method, multiple random state branches are generated at each time step, forming a set of state sequences with a multi-branch tree structure. Ultimately, multiple state sequences covering various possible states are obtained.

[0039] Step S204: Based on the preset flight boundary, filter the multiple state sequences to obtain the target state sequence.

[0040] In this embodiment, it is detected whether the water-free height and pitch angle at each moment in all state sequences are within a preset safe range. State sequences that are all within the safe range are selected as candidate state sequences. If there is only one candidate state sequence, it is directly determined as the target state sequence. If there are multiple candidate state sequences, the target state sequence is determined according to a preset optimization objective function: a desired state sequence is generated based on the current driving mode and the user's throttle command; the weighted deviation between each candidate state sequence and the desired state sequence is calculated; and the candidate state sequence with the smallest deviation cost is selected as the target state sequence. The weighted deviation includes at least the weighted sum of the water-free height deviation, pitch angle deviation, forward velocity deviation, and forward acceleration deviation.

[0041] In addition, multiple state sequences can be filtered based on preset flight boundaries. Specifically, for each state sequence, the minimum distance (i.e., safety margin) between its state parameters at all times and the flight boundaries is calculated, including the distance between the altitude above water and the upper and lower limits of the safe altitude range, and the distance between the pitch angle and the upper and lower limits of the safe pitch angle range. Then, the state sequences are sorted from largest to smallest according to their safety margin, and the top K sequences with the largest safety margins are selected as candidate state sequences, from which the best sequence is chosen. The sequence with the highest safety margin is retained first.

[0042] Step S205: Determine the current control parameters corresponding to the target state sequence, and control the surfboard according to the current control parameters.

[0043] In this embodiment, the target state sequence includes the predicted states of the surfboard at the current moment and at multiple future moments. During the generation of the target state sequence, each predicted state corresponds to a control parameter, which includes at least motor thrust and / or rudder angle. Therefore, the target state sequence is also associated with a corresponding target control sequence, which includes the control parameters corresponding to the current moment and at multiple future moments. The first control parameter is extracted from the target control sequence as the current control parameter. The current control parameter is the motor thrust command and / or rudder angle command of the surfboard at the current moment. The actuators of the surfboard include at least a motor and a servo. If the current control parameter includes a motor thrust command, the motor thrust command is sent to the motor, and the motor adjusts its output thrust according to the motor thrust command to drive the surfboard forward; if the current control parameter includes a rudder angle command, the rudder angle command is sent to the servo, and the servo adjusts the deflection angle of the hydrofoil rudder according to the rudder angle command to adjust the flight attitude and altitude of the surfboard.

[0044] This application determines the current state prediction value through a pre-defined dynamic model, providing a priori basis for surfboard state estimation. Fusing current sensing data with the state prediction value suppresses the interference of single-sensor measurement noise on state estimation, making the fused state value closer to the surfboard's actual motion state. Using the fused state value as the initial state and predicting multiple future state sequences using the dynamic model allows for the prediction of various possible evolutionary trends before the surfboard's actual motion deviates, avoiding state estimation lagging behind actual changes due to sudden external disturbances. Filtering state sequences based on flight boundaries eliminates flight paths with risks of water contact or capsizing, ensuring that the remaining state sequences are within a safe motion range. Determining the control parameters corresponding to the filtered target state sequences and using them to control the surfboard translates state-level safety and optimality into specific actuator-level commands, ensuring that the output of motors and servos matches current flight requirements. This enables the surfboard to select a safe motion path based on the most reliable current state information when facing different water conditions and operational commands, and to readjust subsequent actions at each moment based on the latest sensing data, thereby maintaining stability of liftoff height and pitch attitude within permissible ranges.

[0045] In some optional implementations of this embodiment, the step of determining the predicted current state of the surfboard based on a preset dynamic model includes: Obtain the historical state fusion value and historical control parameters of the surfboard; The historical state fusion value and the historical control parameters are input into the dynamic model. The dynamic model then performs state recursion based on the historical state fusion value and the historical control parameters to output the current state prediction value.

[0046] In this embodiment, the historical state fusion value is the surfboard's state estimate at the previous moment, including at least the historical liftoff height, historical pitch angle, historical forward velocity, and historical forward acceleration. The historical control parameters are the control commands applied to the surfboard at the previous moment, including at least the historical rudder angle and historical motor thrust. Both the historical state fusion value and the historical control parameters are stored in the surfboard's controller memory for direct reading and retrieval in the current control cycle.

[0047] The dynamic model first calculates the angle of attack of the surfboard hydrofoil based on historical pitch angle, historical forward velocity, and historical control surface angle, and then calculates the hydrofoil lift based on the angle of attack and historical forward velocity. Next, based on the hydrofoil lift, preset mass parameters, and preset pitch motion parameters, the vertical acceleration and pitch acceleration of the surfboard are calculated respectively. Finally, based on historical lift-off height, vertical acceleration, historical pitch angle, pitch acceleration, historical forward velocity, historical forward acceleration, and preset control cycle duration, the predicted lift-off height, predicted pitch angle, predicted forward velocity, and predicted forward acceleration at the current moment are recursively obtained and combined as the current state prediction value output.

[0048] In the above recursive process, the angle of attack is the angle between the hydrofoil airfoil chord and the direction of the incoming flow, which is determined by the sum of the historical pitch angle, the historical control surface angle, and the preset hydrofoil installation angle; the hydrofoil lift is calculated according to the hydrodynamic formula; the vertical acceleration is obtained by dividing the resultant force of the hydrofoil lift and the weight of the surfboard by the preset mass parameter; the pitch acceleration is obtained by dividing the total pitch moment acting on the surfboard by the preset moment of inertia; the predicted lift-off height and the predicted pitch angle are obtained by recursively using the second-order integral method, the predicted forward velocity is obtained by recursively using the first-order integral method, and the predicted forward acceleration is calculated by the historical motor thrust and hydrofoil lift.

[0049] This application uses historical states and control parameters to recursively derive the current state prediction value through a dynamic model, making the prediction result conform to the motion law of the surfboard, providing a priori estimation basis for subsequent fusion, and improving the state estimation's resistance to disturbances.

[0050] In some optional implementations of this embodiment, the historical state fusion values ​​include historical height above water, historical pitch angle, historical forward velocity, and historical forward acceleration, and the historical control parameters include historical control surface angle and historical motor thrust. The steps described above, which involve using the dynamic model to perform state recursion based on the historical state fusion value and the historical control parameters, and outputting the current state prediction value, include: The process of the dynamic model performing state recursion based on the historical state fusion value and the historical control parameters to output the current state prediction value includes: The angle of attack of the surfboard hydrofoil is calculated based on the historical pitch angle, the historical forward velocity, and the historical control surface angle. The lift of the surfboard hydrofoil is calculated based on the angle of attack and the historical forward velocity. The vertical acceleration of the surfboard is calculated based on the hydrofoil lift and preset mass parameters. The pitch acceleration of the surfboard is calculated based on the hydrofoil lift, the historical motor thrust, and preset pitch motion parameters. The predicted current state value is calculated based on the historical lift-off height, vertical acceleration, historical pitch angle, pitch angle acceleration, historical forward velocity, historical forward acceleration, historical motor thrust, hydrofoil lift, and preset control cycle duration. The predicted current state value includes predicted lift-off height, predicted pitch angle, predicted forward velocity, and predicted forward acceleration.

[0051] In this embodiment, the hydrofoil angle of attack refers to the angle between the chord line of the surfboard hydrofoil airfoil and the direction of the incoming flow; the hydrofoil lift refers to the force perpendicular to the direction of the incoming flow generated when the surfboard hydrofoil moves in the water. The hydrofoil angle of attack α is determined by the historical pitch angle θ, the historical control surface angle δ, and the preset hydrofoil mounting angle α. offset The sum is determined as: α = θ + δ + α offset The hydrofoil mounting angle is the fixed mounting angle of the hydrofoil relative to the longitudinal axis of the surfboard, and is a preset model parameter. Therefore, the lift L of the hydrofoil is L = 1 / 2ρ. v 2 S C L (α), where ρ is the preset water density value, v is the historical forward velocity, S is the preset effective hydrofoil area, and C L (α) represents the lift coefficient of the hydrofoil, a function of the angle of attack α, obtained by looking up a preset lift coefficient curve. The lift coefficient curve characterizes the hydrofoil's lift generation capability at different angles of attack and is predetermined by the hydrofoil's geometry and hydrodynamic characteristics. Vertical acceleration is obtained by dividing the resultant force of the hydrofoil's lift and the surfboard's weight by a preset mass parameter, representing the total mass of the surfboard and the rider. Pitch acceleration is obtained by dividing the total pitch moment acting on the surfboard by a preset moment of inertia. The predicted lift-off height is obtained by superimposing the product of the historical lift-off height and the historical vertical velocity, along with the second integral term of the vertical acceleration. Since the historical state fusion value includes the historical lift-off height but not directly the historical vertical velocity, the historical vertical velocity is estimated by dividing the difference between the historical lift-off height at the previous moment and the current moment by the control cycle duration. The formula is: ,in, The historical height above water, The historical vertical velocity is obtained through differential estimation. The vertical acceleration is... The preset control cycle duration (typically 20ms). Predicted pitch angle. ,in, The historical pitch angle is... The historical pitch angular velocity is obtained through differential estimation. Let the pitch angle acceleration be... The duration of the control cycle. Predicted forward velocity. Where v is the historical forward velocity, a is the historical forward acceleration, and Δt is the control cycle duration. Predicted forward acceleration. Where T is the historical motor thrust, θ is the historical pitch angle, and D is the water resistance of the surfboard, calculated using the hydrofoil lift and a preset lift-to-drag ratio parameter. The calculated predicted lift-off height, predicted pitch angle, predicted forward velocity, and predicted forward acceleration are combined in a preset order into a vector form to form the current state prediction value.

[0052] This application calculates lift by hydrofoil angle of attack and further obtains vertical acceleration and pitch acceleration, so that the state prediction process conforms to the physical motion law of surfboard. The obtained prediction values ​​can reflect the influence of hydrodynamic changes on flight state, providing a reasonable prior estimate basis for subsequent fusion steps.

[0053] In some optional implementations of this embodiment, the steps of using the state fusion value as the initial state of the surfboard, using the dynamic model as the prediction benchmark for the surfboard state, and predicting multiple state sequences of the surfboard within a preset time period include: The state fusion value is used as the initial state of the surfboard, and the dynamic model is used as the prediction benchmark for the surfboard state to generate multiple candidate control sequences within the preset time period. In chronological order, multiple candidate control sequences and the initial state are substituted into the dynamic model for recursion to obtain the predicted state sequence corresponding to each candidate control sequence.

[0054] In this embodiment, the physical limits of the surfboard's actuators are obtained. These physical limits include at least the motor thrust limit and the rudder angle limit. The motor thrust limit defines the maximum and minimum thrust that the motor can output per unit time, and the rudder angle limit defines the maximum and minimum deflection angle that the rudder can deflect. A feasible space for the control quantity is determined based on the actuator physical limits. This feasible space is a combination of the continuous intervals of allowed values ​​for the motor thrust and the continuous intervals of allowed values ​​for the rudder angle. Multiple candidate control sequences are generated within this feasible space. Each candidate control sequence contains the control parameters corresponding to each moment of the surfboard within a preset time period. These control parameters include at least the motor thrust and / or the rudder angle. The candidate control sequences can be generated by superimposing a random perturbation conforming to a preset probability distribution onto the control quantities at each moment of the benchmark sequence, using the optimal control sequence of the surfboard at the previous moment as a reference. This generates multiple perturbed candidate control sequences, where the preset probability distribution can be a Gaussian distribution or a uniform distribution.

[0055] For any candidate control sequence Where N is the number of prediction steps within the preset time period, expressed as a state fusion value. As an initial state, the following recursive process is executed in chronological order: Starting from the initial state... and the first control parameter in the candidate control sequence As input to the dynamic model, the dynamic model is invoked to perform a one-step state recursion to obtain the predicted state at time 1. ; Based on the predicted state at time 1 The second control parameter in the candidate control sequence As input to the dynamic model, the dynamic model is invoked to perform a one-step state recursion to obtain the predicted state at time 2. And so on, to obtain the predicted state at time N. The initial state and the predicted states at each time step are concatenated in chronological order to obtain the predicted state sequence corresponding to the candidate control sequence. This process is repeated to iterate step-by-step for each candidate control sequence, resulting in the predicted state sequence for each candidate control sequence, thus forming multiple state sequences within the preset time period. The process of calling the dynamic model to perform a one-step state recursion based on the predicted state and control parameters is consistent with step S201.

[0056] This application generates multiple candidate control sequences within the physical limits of the actuator and obtains the corresponding predicted state sequences through a dynamic model, so that each sequence meets the motion feasibility requirements of the surfboard, providing a sufficient number of physically reasonable state evolution paths for subsequent screening.

[0057] In some optional implementations of this embodiment, the step of filtering multiple state sequences according to a preset flight boundary to obtain a target state sequence includes: Detect whether each state parameter in all the state sequences is within the flight boundary, and use the state sequences that are all within the flight boundary as candidate state sequences; If there are multiple candidate state sequences, the target state sequence is determined from the multiple candidate state sequences according to a preset optimization objective function.

[0058] In this embodiment, the preset flight boundaries are the range of state parameters allowed for the surfboard during normal operation, including at least a safe range for water lift and a safe range for pitch angle. The safe range for water lift limits the allowable water lift range during flight; its upper limit prevents the surfboard from flying too high and causing loss of control, and its lower limit prevents the surfboard from touching the water. The safe range for pitch angle limits the allowable pitch angle range during flight; its upper and lower limits prevent the surfboard from overturning or stalling due to excessive pitch angle. It is checked whether all state parameters in all state sequences are within the flight boundaries. Specifically, for each state sequence in the trajectory set, it is sequentially checked whether the water lift at each moment in the state sequence is within the safe range for water lift, and whether the pitch angle at each moment is within the safe range for pitch angle. If the water lift or pitch angle at any moment does not meet either condition, the state sequence is determined not to meet the flight boundaries and is removed from the trajectory set. A set of candidate state sequences is obtained by taking a state sequence in which all state parameters are within the flight boundary. If there is only one candidate state sequence, then the candidate state sequence is directly determined as the target state sequence.

[0059] If there are multiple candidate state sequences, the target state sequence is determined from among them according to a preset optimization objective function. Specifically, the desired liftoff height is determined based on the surfboard's current driving mode, with different driving modes corresponding to different desired liftoff heights; the desired forward speed is determined based on the throttle command input by the user via remote control; the desired pitch angle is set to a preset small positive value, and the desired forward acceleration is set to zero. Using the current state fusion value as a starting point and the above-mentioned desired values ​​as targets, a preset smoothing algorithm is used to generate desired state sequences for multiple future times. Then, the weighted deviation between each candidate state sequence and the desired state sequence is calculated according to the optimization objective function to obtain the deviation cost value corresponding to each candidate state sequence. The weighted deviation includes at least the weighted sum of liftoff height deviation, pitch angle deviation, forward speed deviation, and forward acceleration deviation. Finally, the candidate state sequence with the smallest deviation cost value is selected as the target state sequence.

[0060] This application eliminates state sequences with risks of water contact or capsizing by eliminating them at the flight boundary, ensuring that the remaining sequences are all within the safe flight range. Then, it selects the sequence with the smallest deviation from the desired state by using an optimization objective function, so that the screening results have both safety and desired flight quality.

[0061] In some optional implementations of this embodiment, the step of determining the target state sequence from the multiple candidate state sequences according to a preset optimization objective function includes: Obtain the desired state sequence of the surfboard within the preset time period; Calculate the weighted deviation between each candidate state sequence and the desired state sequence according to the optimization objective function, and obtain the deviation cost corresponding to each candidate state sequence; The candidate state sequence with the smallest deviation cost is selected as the target state sequence.

[0062] In this embodiment, the desired state sequence , Where N is the number of prediction steps within the preset time period. For the desired state at step i, each desired state includes at least the desired lift-off height, desired pitch angle, desired forward velocity, and desired forward acceleration. The desired lift-off height is determined based on the surfboard's current riding mode. Specifically, the driving modes include at least one of novice mode, intermediate mode, and competitive mode, with different preset values ​​for the desired liftoff height corresponding to different driving modes. For example, in novice mode, the desired liftoff height is set to a lower value (e.g., 0.3m) to reduce the difficulty of operation; in competitive mode, the desired liftoff height is set to a higher value (e.g., 0.6m) to unleash the performance of the surfboard. Simultaneously, the desired forward speed is determined based on the throttle command input by the user via remote control. Specifically, there is a preset mapping relationship between the throttle command opening and the desired forward velocity; when the throttle opening increases, the desired forward velocity increases accordingly. A preset desired pitch angle is obtained. and expected forward acceleration The desired pitch angle is set to a small positive value (e.g., 2° to 5°) to ensure sufficient lift from the hydrofoil; the desired forward acceleration is set to 0 to maintain a constant speed cruise. Then, using the current state fusion value as the starting point, and the desired lift height, desired pitch angle, desired forward velocity, and desired forward acceleration as targets, a preset smoothing algorithm is used to generate a transition trajectory from the current state to the target state, resulting in the desired state sequence for the next N time moments. .

[0063] For any candidate state sequence , where the state vector is: The corresponding deviation cost value The expression is: in, The weighting coefficient for the deviation in height from water. This is the weighting coefficient for pitch angle deviation. The weighting coefficient for the forward velocity deviation, The weighting coefficients for forward acceleration are non-negative. The first term in the above expression evaluates the degree to which the height above water deviates from the desired height in the candidate state sequence; the second term evaluates the degree to which the pitch angle deviates from the desired pitch angle; the third term evaluates the degree to which the forward velocity deviates from the desired velocity; and the fourth term evaluates the degree to which the forward acceleration deviates from zero (i.e., the drastic change in velocity). The objective function value for each sequence in the candidate state sequence set is calculated, and the sequence with the smallest objective function value is selected as the target state sequence.

[0064] This application evaluates the gap between each candidate sequence and the desired state by using weighted bias, so that the selected target state sequence can comprehensively reflect driving needs in terms of height above water, pitch angle, forward velocity and forward acceleration, thereby taking into account the comprehensive performance of high stability, smooth attitude, speed following and smooth acceleration and deceleration.

[0065] In some optional implementations of this embodiment, the step of fusing the current perceived data and the current state prediction value to obtain a state fusion value includes: Calculate the first covariance of the current state prediction value and the second covariance of the current sensed data, and determine the fusion weight of the current state prediction value and the current sensed data based on the first covariance and the second covariance; The current state prediction value and the current perceived data are weighted and fused according to the fusion weight to obtain the state fusion value.

[0066] In this embodiment, the first covariance matrix is ​​the error covariance matrix of the current state prediction value. This first covariance matrix is ​​used to characterize the degree of uncertainty of the current state prediction value. It is derived from the covariance matrix of the state fusion value from the previous time step. The Jacobian matrix of the aforementioned dynamic model The process noise covariance matrix Q is transmitted and then superimposed with a preset process noise covariance matrix Q, and is obtained recursively. The formula is as follows: ,in, For the dynamic model f( The Jacobian matrix obtained by taking the partial derivative with respect to the state vector, Q is a preset process noise covariance matrix used to characterize the uncertainty of the dynamic model itself. It is predetermined by statistical analysis of the prediction deviation of the dynamic model under different operating conditions and / or by trial-and-error calibration. The second covariance matrix is ​​the measurement noise covariance matrix R of the current sensing data, which is used to characterize the degree of uncertainty of the measurement values ​​of each sensor in the current sensing data. This second covariance matrix is ​​preset according to the noise characteristics of each sensor on the surfboard, for example, obtained through statistical results of sensor factory calibration data or static test data.

[0067] Next, the Kalman gain matrix is ​​calculated based on the first and second covariance matrices. : in, R is the first covariance matrix, R is the second covariance matrix, and H is a preset observation matrix, used to map the state vector to the sensor measurement space. The weights of the current state prediction value and the current sensed data are determined based on the Kalman gain matrix. Specifically, Each element takes a value between 0 and 1, and its magnitude reflects the relative confidence level between the predicted value of the current state and the current perceived data: when the first covariance matrix The larger the relative value (i.e., the greater the uncertainty of the current state prediction), the greater the Kalman gain matrix. The larger the corresponding value, the greater the weight of the current sensed data in the fusion, and the more the fusion result tends to trust the sensed data; conversely, when the second covariance matrix R is relatively large (i.e., the greater the measurement noise of the current sensed data), the Kalman gain matrix... The smaller the value, the greater the weight of the current state prediction in the fusion, and the more likely the fusion result is to believe the prediction.

[0068] Finally, according to the formula Determine the state fusion value. Among them, The predicted value for the current state. The current perceived data, This is the state fusion value. Specifically, the current perceived data is calculated. Corresponding observation component in the current state prediction value The residuals between; then, the Kalman gain matrix. The residuals are weighted as fusion weights, and the weighted residuals are then added to the current state prediction value. The state fusion value is obtained from the above.

[0069] This application determines the fusion weight by comparing the uncertainty of the predicted value and the perceived data, so that the fusion result tends to maintain the prediction trend when the prediction is reliable and tends to follow the measured feedback when the measurement is reliable, thereby achieving a balance between sensor noise and model error and improving the adaptability of state estimation.

[0070] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing related hardware through computer-readable instructions. These computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. The aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, optical disk, or read-only memory (ROM), or random access memory (RAM).

[0071] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0072] Further reference Figure 3 As a response to the above Figure 2 The implementation of the method shown in this application provides an embodiment of an electric hydrofoil surfboard control device, which is similar to... Figure 2 Corresponding to the method embodiments shown, this device can be specifically applied to various electronic devices.

[0073] like Figure 3 As shown, the electric hydrofoil surfboard control device 300 described in this embodiment includes: a determination module 301, a fusion module 302, a prediction module 303, a screening module 304, and a control module 305. Wherein: The determination module 301 is used to determine the current state prediction value of the surfboard according to the preset dynamic model; The fusion module 302 is used to acquire the current sensing data of the surfboard, fuse the current sensing data and the current state prediction value to obtain a state fusion value; The prediction module 303 is used to use the state fusion value as the initial state of the surfboard, use the dynamic model as the prediction benchmark for the state of the surfboard, and predict multiple state sequences of the surfboard within a preset time period. The filtering module 304 is used to filter multiple state sequences according to a preset flight boundary to obtain a target state sequence; The control module 305 is used to determine the current control parameters corresponding to the target state sequence, and control the surfboard according to the current control parameters.

[0074] The electric hydrofoil surfboard control device provided in this application determines the current state prediction value through a preset dynamic model, providing a priori basis for surfboard state estimation. Acquiring current sensing data and fusing it with the state prediction value can suppress the interference of single sensor measurement noise on the state estimation, making the fused state value closer to the actual motion state of the surfboard. Using the fused state value as the initial state and predicting multiple future state sequences using the dynamic model allows for the prediction of various possible evolution trends before the actual surfboard motion deviates, avoiding state estimation lagging behind actual changes due to sudden external disturbances. Filtering the state sequences based on flight boundaries can eliminate flight paths with risks of water contact or capsizing, ensuring that the remaining state sequences are all within a safe motion range. Determining the control parameters corresponding to the filtered target state sequences and using them to control the surfboard translates the safety and optimality at the state level into specific instructions at the actuator level, ensuring that the output of the motors and servos matches the current flight requirements. This allows the surfboard to select a safe path based on the most reliable current status information when facing different water conditions and operating instructions, and to readjust subsequent actions at every moment based on the latest sensing data, thereby maintaining the stability of the lift-off height and pitch attitude within the allowable range.

[0075] In some optional implementations of this embodiment, the determining module 301 is further configured to: Obtain the historical state fusion value and historical control parameters of the surfboard; The historical state fusion value and the historical control parameters are input into the dynamic model. The dynamic model then performs state recursion based on the historical state fusion value and the historical control parameters to output the current state prediction value.

[0076] The electric hydrofoil surfboard control device provided in this application obtains the current state prediction value through recursion of historical state and control parameters using a dynamic model, so that the prediction result conforms to the motion law of the surfboard, providing a priori estimation basis for subsequent fusion and improving the state estimation's resistance to disturbances.

[0077] In some optional implementations of this embodiment, the determining module 301 is further configured to: The angle of attack of the surfboard hydrofoil is calculated based on the historical pitch angle, the historical forward velocity, and the historical control surface angle. The lift of the surfboard hydrofoil is calculated based on the angle of attack and the historical forward velocity. The vertical acceleration of the surfboard is calculated based on the hydrofoil lift and preset mass parameters. The pitch acceleration of the surfboard is calculated based on the hydrofoil lift, the historical motor thrust, and preset pitch motion parameters. The predicted current state value is calculated based on the historical lift-off height, vertical acceleration, historical pitch angle, pitch angle acceleration, historical forward velocity, historical forward acceleration, historical motor thrust, hydrofoil lift, and preset control cycle duration. The predicted current state value includes predicted lift-off height, predicted pitch angle, predicted forward velocity, and predicted forward acceleration.

[0078] The electric hydrofoil surfboard control device provided in this application calculates lift by the angle of attack of the hydrofoil and further obtains vertical acceleration and pitch acceleration, so that the state prediction process conforms to the physical motion law of the surfboard. The obtained prediction values ​​can reflect the influence of hydrodynamic changes on the flight state, providing a reasonable prior estimate basis for subsequent fusion steps.

[0079] In some optional implementations of this embodiment, the prediction module 303 is further configured to: The state fusion value is used as the initial state of the surfboard, and the dynamic model is used as the prediction benchmark for the surfboard state to generate multiple candidate control sequences within the preset time period. In chronological order, multiple candidate control sequences and the initial state are substituted into the dynamic model for recursion to obtain the predicted state sequence corresponding to each candidate control sequence.

[0080] The electric hydrofoil surfboard control device provided in this application generates multiple candidate control sequences within the physical limits of the actuator and obtains the corresponding predicted state sequences through recursion using a dynamic model, so that each sequence meets the motion feasibility requirements of the surfboard and provides a sufficient number of physically reasonable state evolution paths for subsequent screening.

[0081] In some optional implementations of this embodiment, the filtering module 304 is further configured to: Detect whether each state parameter in all the state sequences is within the flight boundary, and use the state sequences that are all within the flight boundary as candidate state sequences; If there are multiple candidate state sequences, the target state sequence is determined from the multiple candidate state sequences according to a preset optimization objective function.

[0082] The electric hydrofoil surfboard control device provided in this application eliminates state sequences with risks of water contact or overturning by eliminating them at the flight boundary, ensuring that the remaining sequences are all within the safe flight range. Then, it selects the sequence with the smallest deviation from the desired state by using an optimization objective function, so that the screening results have both safety and desired flight quality.

[0083] In some optional implementations of this embodiment, the filtering module 304 is further configured to: Obtain the desired state sequence of the surfboard within the preset time period; Calculate the weighted deviation between each candidate state sequence and the desired state sequence according to the optimization objective function, and obtain the deviation cost corresponding to each candidate state sequence; The candidate state sequence with the smallest deviation cost is selected as the target state sequence.

[0084] The electric hydrofoil surfboard control device provided in this application evaluates the difference between each candidate sequence and the desired state through weighted deviation evaluation, so that the selected target state sequence can comprehensively reflect the driving needs in terms of lift-off height, pitch angle, forward speed and forward acceleration, thereby taking into account the comprehensive performance of height stability, smooth attitude, speed following and smooth acceleration and deceleration.

[0085] In some optional implementations of this embodiment, the fusion module 302 is further configured to: Calculate the first covariance of the current state prediction value and the second covariance of the current sensed data, and determine the fusion weight of the current state prediction value and the current sensed data based on the first covariance and the second covariance; The current state prediction value and the current perceived data are weighted and fused according to the fusion weight to obtain the state fusion value.

[0086] The electric hydrofoil surfboard control device provided in this application determines the fusion weight by comparing the uncertainty of the predicted value and the perceived data. This makes the fusion result tend to maintain the predicted trend when the prediction is reliable and tend to follow the measured feedback when the measurement is reliable, thereby achieving a balance between sensor noise and model error and improving the adaptability of state estimation.

[0087] To address the aforementioned technical problems, embodiments of this application also provide a computer device. Please refer to [link / reference needed]. Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0088] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected via a system bus. It should be noted that only the computer device 4 with components 41, 42, and 43 is shown in the figure; however, it should be understood that it is not required to implement all the shown components, and more or fewer components can be implemented alternatively. Those skilled in the art will understand that the computer device described here is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0089] The computer device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The computer device can interact with the user via a keyboard, mouse, remote control, touchpad, or voice control.

[0090] The memory 41 includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 41 may be an internal storage unit of the computer device 4, such as the hard disk or memory of the computer device 4. In other embodiments, the memory 41 may also be an external storage device of the computer device 4, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the computer device 4. Of course, the memory 41 may also include both the internal storage unit and its external storage device of the computer device 4. In this embodiment, the memory 41 is typically used to store the operating system and various application software installed on the computer device 4, such as computer-readable instructions for the control method of an electric hydrofoil surfboard. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or will be output.

[0091] In some embodiments, the processor 42 may be a central processing unit (CPU), controller, microcontroller, microprocessor, or other data processing chip. The processor 42 is typically used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to execute computer-readable instructions stored in the memory 41 or to process data, for example, to execute computer-readable instructions for the electric hydrofoil surfboard control method.

[0092] The network interface 43 may include a wireless network interface or a wired network interface, which is typically used to establish communication connections between the computer device 4 and other electronic devices.

[0093] The computer device provided in this application determines the current state prediction value through a preset dynamic model, providing a priori basis for surfboard state estimation. By acquiring current sensing data and fusing it with the state prediction value, the interference of single sensor measurement noise on state estimation can be suppressed, making the fused state value closer to the actual motion state of the surfboard. Using the fused state value as the initial state and predicting multiple future state sequences using the dynamic model, various possible evolutionary trends can be anticipated before the actual motion of the surfboard deviates, avoiding state estimation lagging behind actual changes due to sudden external disturbances. Filtering the state sequences based on flight boundaries can eliminate flight paths with risks of water contact or capsizing, ensuring that the remaining state sequences are all within a safe motion range. Determining the control parameters corresponding to the filtered target state sequences and using them to control the surfboard transforms the safety and optimality at the state level into specific instructions at the actuator level, ensuring that the output of the motors and servos matches the current flight requirements. This allows the surfboard to select a safe motion path based on the most reliable current state information when facing different water conditions and operational instructions, and to readjust subsequent actions at each moment based on the latest sensing data, thereby maintaining stability of the liftoff height and pitch attitude within permissible ranges.

[0094] This application also provides another embodiment, namely, providing a computer-readable storage medium storing computer-readable instructions that can be executed by at least one processor to cause the at least one processor to perform the steps of the electric hydrofoil surfboard control method described above.

[0095] The computer-readable storage medium provided in this application determines the current state prediction value through a preset dynamic model, providing a priori basis for surfboard state estimation. Acquiring current sensing data and fusing it with the state prediction value can suppress the interference of single sensor measurement noise on the state estimation, making the fused state value closer to the actual motion state of the surfboard. Using the fused state value as the initial state and predicting multiple future state sequences using the dynamic model allows for the prediction of various possible evolution trends before the actual motion of the surfboard deviates, avoiding state estimation lagging behind actual changes due to sudden external disturbances. Filtering the state sequences based on flight boundaries can eliminate flight paths with risks of water contact or capsizing, ensuring that the remaining state sequences are all within a safe motion range. Determining the control parameters corresponding to the filtered target state sequences and using them to control the surfboard translates the safety and optimality at the state level into specific instructions at the actuator level, ensuring that the output of the motors and servos matches the current flight requirements. This allows the surfboard to select a safe path based on the most reliable current status information when facing different water conditions and operating instructions, and to readjust subsequent actions at every moment based on the latest sensing data, thereby maintaining the stability of the lift-off height and pitch attitude within the allowable range.

[0096] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0097] Obviously, the embodiments described above are only some embodiments of this application, not all embodiments. The accompanying drawings show preferred embodiments of this application, but do not limit the patent scope of this application. This application can be implemented in many different forms; rather, these embodiments are provided to provide a more thorough and comprehensive understanding of the disclosure of this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments, or make equivalent substitutions for some of the technical features. Any equivalent structures made using the content of this application's specification and drawings, directly or indirectly applied to other related technical fields, are similarly within the scope of patent protection of this application.

Claims

1. A control method for an electric hydrofoil surfboard, characterized in that, Includes the following steps: The current state prediction value of the surfboard is determined based on a preset dynamic model; Obtain the current sensing data of the surfboard, and fuse the current sensing data with the current state prediction value to obtain a state fusion value; The state fusion value is used as the initial state of the surfboard, and the dynamic model is used as the prediction benchmark for the state of the surfboard to predict multiple state sequences of the surfboard within a preset time period. Based on the preset flight boundary, multiple state sequences are filtered to obtain the target state sequence; Determine the current control parameters corresponding to the target state sequence, and control the surfboard according to the current control parameters.

2. The electric hydrofoil surfboard control method according to claim 1, characterized in that, Determining the predicted current state of the surfboard based on a preset dynamic model includes: Obtain the historical state fusion value and historical control parameters of the surfboard; The historical state fusion value and the historical control parameters are input into the dynamic model. The dynamic model then performs state recursion based on the historical state fusion value and the historical control parameters to output the current state prediction value.

3. The electric hydrofoil surfboard control method according to claim 2, characterized in that, The historical state fusion values ​​include historical height above water, historical pitch angle, historical forward velocity, and historical forward acceleration; the historical control parameters include historical control surface angle and historical motor thrust. The step of using the dynamic model to perform state recursion based on the historical state fusion value and the historical control parameters, and outputting the current state prediction value, includes: The angle of attack of the surfboard hydrofoil is calculated based on the historical pitch angle, the historical forward velocity, and the historical control surface angle. The lift of the surfboard hydrofoil is calculated based on the angle of attack and the historical forward velocity. The vertical acceleration of the surfboard is calculated based on the hydrofoil lift and preset mass parameters. The pitch acceleration of the surfboard is calculated based on the hydrofoil lift, the historical motor thrust, and preset pitch motion parameters. The predicted current state value is calculated based on the historical lift-off height, vertical acceleration, historical pitch angle, pitch angle acceleration, historical forward velocity, historical forward acceleration, historical motor thrust, hydrofoil lift, and preset control cycle duration. The predicted current state value includes predicted lift-off height, predicted pitch angle, predicted forward velocity, and predicted forward acceleration.

4. The electric hydrofoil surfboard control method according to claim 1, characterized in that, The step of using the fused state value as the initial state of the surfboard and the dynamic model as the prediction benchmark for the surfboard's state to predict multiple state sequences of the surfboard within a preset time period includes: The state fusion value is used as the initial state of the surfboard, and the dynamic model is used as the prediction benchmark for the surfboard state to generate multiple candidate control sequences within the preset time period. In chronological order, multiple candidate control sequences and the initial state are substituted into the dynamic model for recursion to obtain the predicted state sequence corresponding to each candidate control sequence.

5. The electric hydrofoil surfboard control method according to claim 1, characterized in that, The step of filtering multiple state sequences according to a preset flight boundary to obtain a target state sequence includes: Detect whether each state parameter in all the state sequences is within the flight boundary, and use the state sequences that are all within the flight boundary as candidate state sequences; If there are multiple candidate state sequences, the target state sequence is determined from the multiple candidate state sequences according to a preset optimization objective function.

6. The electric hydrofoil surfboard control method according to claim 5, characterized in that, The step of determining the target state sequence from multiple candidate state sequences according to a preset optimization objective function includes: Obtain the desired state sequence of the surfboard within the preset time period; Calculate the weighted deviation between each candidate state sequence and the desired state sequence according to the optimization objective function, and obtain the deviation cost corresponding to each candidate state sequence; The candidate state sequence with the smallest deviation cost is selected as the target state sequence.

7. The control method for an electric hydrofoil surfboard according to any one of claims 1 to 6, characterized in that, The step of fusing the current perceived data and the current state prediction value to obtain a state fusion value includes: Calculate the first covariance of the current state prediction value and the second covariance of the current sensed data, and determine the fusion weight of the current state prediction value and the current sensed data based on the first covariance and the second covariance; The current state prediction value and the current perceived data are weighted and fused according to the fusion weight to obtain the state fusion value.

8. A control device for an electric hydrofoil surfboard, characterized in that, include: The determination module is used to determine the predicted current state of the surfboard based on a preset dynamic model; The fusion module is used to acquire the current sensing data of the surfboard, fuse the current sensing data and the current state prediction value to obtain a state fusion value; The prediction module is used to use the state fusion value as the initial state of the surfboard, use the dynamic model as the prediction benchmark for the state of the surfboard, and predict multiple state sequences of the surfboard within a preset time period. The filtering module is used to filter multiple state sequences according to preset flight boundaries to obtain a target state sequence; The control module is used to determine the current control parameters corresponding to the target state sequence, and control the surfboard according to the current control parameters.

9. A computer device, characterized in that, The device includes a memory and a processor, wherein the memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement the steps of the electric hydrofoil surfboard control method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the electric hydrofoil surfboard control method as described in any one of claims 1 to 7.

Images