A pan / tilt inertia identification method, pan / tilt and storage medium

Abstract

The present invention discloses a gimbal inertia identification method, a gimbal, and a storage medium, belonging to the field of gimbal control technology. The gimbal inertia identification method includes: receiving a gimbal inertia identification instruction; driving the gimbal to rotate in a preset manner and collecting angular velocity and current data corresponding to the preset rotation; and calculating the gimbal's inertia parameters using a fitting algorithm based on the collected angular velocity and current data corresponding to the preset rotation. The gimbal inertia identification method drives the gimbal to rotate in a preset manner and collects angular velocity and current data corresponding to the preset rotation. After data sampling, the gimbal's moment of inertia is automatically calculated using the sample data through a fitting algorithm. This allows the gimbal to automatically identify its own moment of inertia, improves gimbal design efficiency, and enables the gimbal to adapt to load changes to a certain extent, thereby enhancing the market competitiveness of gimbal products.

Description

Technical Field

[0001] The present invention relates to the technical field of pan / tilt control technology, and in particular to a pan / tilt inertia identification method, a pan / tilt and a storage medium. Background Art

[0002] Existing gimbals lack inertia identification capabilities, and the control parameter tuning process for the gimbal mainly uses trial-and-error methods, which is inefficient. Furthermore, the unknown inertia parameters of the gimbal restrict the use of model-based control algorithms. Other model-based control algorithms, such as auto-disturbance rejection, require known inertia parameters. Furthermore, after the gimbal design is completed, the control parameters are stored in the gimbal storage device and no longer change. This makes it difficult for the same set of control parameters to adapt to application scenarios with variable loads, such as cameras. In scenarios with variable loads, the traditional approach is to adjust the parameters more conservatively to provide a higher stability margin for the gimbal, thereby preventing oscillation or even divergence after changing the load. However, this approach also restricts the improvement of the gimbal's control performance. Summary of the Invention

[0003] In view of this, an embodiment of the present invention aims to provide a gimbal inertia identification method, a gimbal, and a storage medium to solve the technical problems that the current gimbal control parameter tuning method is inefficient and cannot meet the requirements of variable loads.

[0004] The technical solutions adopted by the present invention to solve the above technical problems are as follows:

[0005] According to one aspect of an embodiment of the present invention, a gimbal inertia identification method is provided, the method comprising:

[0006] Receive gimbal inertia identification command;

[0007] Drive the gimbal to rotate according to a preset method and collect the angular velocity and current data corresponding to the preset rotation method;

[0008] The inertia parameters of the gimbal are calculated using a fitting algorithm based on the collected angular velocity and current data corresponding to the preset rotation mode.

[0009] Optionally, receiving the gimbal inertia identification instruction includes receiving the gimbal inertia identification instruction through a UAV ground station, a UAV remote controller, or a handheld gimbal button.

[0010] Optionally, the driving the gimbal to rotate in a preset manner and collecting angular velocity and current data corresponding to the rotation in the preset manner includes:

[0011] The angular velocity target of the first axis of the gimbal is set to 0 and maintained for one step period. Meanwhile, quadrature-axis current data of the first motor corresponding to the first axis during the entire maintenance process and the angular velocity of the first motor at the end of the maintenance process are collected. The step period is greater than or equal to the settling time of the step response of the angular velocity loop corresponding to the first axis.

[0012] Set the positive step speed target of the first axis to n*ω r1 , maintaining a step cycle, and simultaneously collecting the motor quadrature-axis current data of the first motor during the entire maintaining process, and the angular velocity of the first motor at the end of the maintaining process, wherein the initial value of n is 1;

[0013] Setting the angular velocity target of the first axis to 0 and maintaining it for one step cycle, while collecting the quadrature-axis current data of the first motor during the entire maintenance process and the angular velocity of the first motor at the end of the maintenance process;

[0014] Set the negative step speed target of the first axis to -n*ω r1 , while collecting the motor quadrature-axis current data of the first motor during the entire maintenance process, and the angular velocity of the first motor at the end of the maintenance process;

[0015] n is incremented by 1. If n is less than a first preset value, the process returns to setting the angular velocity target of the first axis of the gimbal to 0 and maintaining it for one step cycle, while simultaneously collecting the motor quadrature-axis current data of the first motor corresponding to the first axis during the entire maintenance process, and the angular velocity of the first motor at the end of the maintenance process. If n is greater than or equal to the first preset value, executing the operation of rotating the gimbal according to the preset rotation method and the corresponding angular velocity and current data collected, and calculating the inertia parameters of the gimbal using a fitting algorithm.

[0016] Optionally, before setting the angular velocity target of the first axis of the gimbal to 0 and maintaining it for one step cycle, and simultaneously collecting the motor quadrature-axis current data of the first motor corresponding to the first axis during the entire maintenance process, and the angular velocity of the first motor at the end of the maintenance process, the method further includes:

[0017] Close the attitude angle closed loop corresponding to the first axis of the gimbal.

[0018] Optionally, the fitting algorithm includes a gradient descent method or a least squares method. When the fitting algorithm is the gradient descent method, calculating the inertia parameters of the gimbal using the fitting algorithm based on the collected angular velocity and current data corresponding to the preset rotation method includes:

[0019] The motor quadrature-axis current data collected in each step cycle is integrated to obtain the current integral value, and the angular velocity collected in the corresponding step cycle is subtracted from the angular velocity collected in the previous step cycle to obtain the angular velocity differential value. An angular velocity differential value and a current integral value are regarded as a complete sample data;

[0020] The sample data is fitted using the gradient descent method. If the loss function is less than the preset threshold within the preset number of gradient descent steps, the inertia identification of the first axis of the gimbal is determined to be successful, and the inertia parameters of the first axis obtained by fitting are saved.

[0021] Optionally, when the gimbal includes a roll axis, a pitch axis, and a yaw axis, driving the gimbal to rotate according to a preset manner, and collecting angular velocity and current data corresponding to the rotation in the preset manner includes:

[0022] The roll, pitch, and yaw axes of the gimbal are driven in sequence to rotate according to the preset method. The other two axes that are not currently driven are set to the mechanical angle servo mode with a mechanical angle target of 0. The angular velocity and current data corresponding to the currently driven axis rotating according to the preset method are collected.

[0023] Optionally, the calculating the inertia parameters of the gimbal using a fitting algorithm according to the collected angular velocity and current data corresponding to the preset rotation mode includes:

[0024] After completing the collection of the angular velocity and current data corresponding to the three axes, the angular velocity and current data corresponding to each axis are rotated according to the preset rotation method collected, and the inertia parameters corresponding to each axis are calculated using a fitting algorithm; or, after completing the collection of the angular velocity and current data corresponding to each axis, the inertia parameters corresponding to the axis currently completing the collection are calculated using a fitting algorithm based on the angular velocity and current data corresponding to the preset rotation method collected for the axis currently completing the collection.

[0025] Optionally, driving the gimbal to rotate in a preset manner includes:

[0026] The square wave signal drives the gimbal to rotate in a preset manner.

[0027] According to another aspect of an embodiment of the present invention, a gimbal is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program implements the steps of the gimbal inertia identification method when executed by the processor.

[0028] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided, on which a gimbal inertia identification program is stored. When the gimbal inertia identification program is executed by a processor, the steps of the gimbal inertia identification method described above are implemented.

[0029] The gimbal inertia identification method, gimbal, and storage medium provided in an embodiment of the present invention receive a gimbal inertia identification instruction; drive the gimbal to rotate in a preset manner and collect angular velocity and current data corresponding to the preset rotation; and calculate the gimbal's inertia parameters using a fitting algorithm based on the collected angular velocity and current data corresponding to the preset rotation. The gimbal inertia identification method of an embodiment of the present invention drives the gimbal to rotate in a preset manner and collects angular velocity and current data corresponding to the preset rotation. After data sampling, the gimbal's moment of inertia is automatically calculated using the sample data using a fitting algorithm. This allows the gimbal to automatically identify its own moment of inertia, improves gimbal design efficiency, and enables the gimbal to adapt to load changes to a certain extent, thereby enhancing the market competitiveness of gimbal products. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The present invention will be further described below with reference to the accompanying drawings and embodiments, in which:

[0031] Figure 1 This is a flow chart of a gimbal inertia identification method provided by an embodiment of the present invention;

[0032] Figure 2 is a flow chart of another gimbal inertia identification method provided by an embodiment of the present invention;

[0033] Figure 3 This is a schematic diagram of a pan / tilt system according to the present invention;

[0034] Figure 4 This is a flow chart of another gimbal inertia identification method provided by an embodiment of the present invention;

[0035] Figure 5 This is a flow chart of another gimbal inertia identification method provided by an embodiment of the present invention;

[0036] Figure 6 This is a schematic diagram of a pan / tilt structure provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0037] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0038] Example 1

[0039] Figure 1 This is a flow chart of a gimbal inertia identification method provided by an embodiment of the present invention. The method of this embodiment is automatically executed by the gimbal, wherein each step can be performed in the order shown in the flow chart, or multiple steps can be performed simultaneously according to actual conditions, which is not limited here. The gimbal inertia identification method provided by the present invention includes:

[0040] Step S100, receiving a gimbal inertia identification instruction;

[0041] Step S200: driving the gimbal to rotate according to a preset method, and collecting angular velocity and current data corresponding to the rotation in the preset method;

[0042] Step S300 , calculating the inertia parameters of the gimbal using a fitting algorithm based on the collected angular velocity and current data corresponding to the preset rotation mode.

[0043] Through the above-described implementation, a gimbal inertia identification command is first received; then, the gimbal is driven to rotate in a preset manner, and the angular velocity and current data corresponding to the preset rotation are collected; finally, based on the collected angular velocity and current data corresponding to the preset rotation, a fitting algorithm is used to calculate the gimbal's inertia parameters. The gimbal inertia identification method of this embodiment, by driving the gimbal to rotate in a preset manner and collecting the angular velocity and current data corresponding to the preset rotation, automatically calculates the gimbal's moment of inertia using the sample data through a fitting algorithm after data sampling. This allows the gimbal to automatically identify its own moment of inertia, improves gimbal design efficiency, and enables the gimbal to adapt to load changes to a certain extent, thereby enhancing the market competitiveness of gimbal products.

[0044] In this embodiment, it should first be noted that, considering the technical problems in the prior art, the control parameter adjustment method of the gimbal is inefficient and cannot meet the requirements of variable loads. Therefore, in order to solve the above technical problems, in this embodiment, a gimbal inertia identification instruction is received; the gimbal is driven to rotate in a preset manner, and the angular velocity and current data corresponding to the rotation in the preset manner are collected; and the inertia parameters of the gimbal are calculated using a fitting algorithm based on the collected angular velocity and current data corresponding to the rotation in the preset manner. The gimbal inertia identification method of this embodiment drives the gimbal to rotate in a preset manner and collects the angular velocity and current data corresponding to the rotation in the preset manner. After the data sampling is completed, the gimbal's moment of inertia is automatically calculated using the sample data through a fitting algorithm, so that the gimbal can automatically identify its own moment of inertia, improve the gimbal design efficiency, and enable the gimbal to have a certain ability to adapt to load changes, thereby improving the market competitiveness of the gimbal product.

[0045] The above steps will be described in detail below in conjunction with specific implementation methods.

[0046] In step S100 , a gimbal inertia identification instruction is received.

[0047] Specifically, after receiving the gimbal inertia identification command, the gimbal enters the gimbal inertia identification mode.

[0048] In one embodiment, receiving the gimbal inertia identification instruction includes receiving the gimbal inertia identification instruction through a UAV ground station, a UAV remote controller, or a handheld gimbal button.

[0049] In this embodiment, when the gimbal is mounted on a drone, the gimbal inertia identification command can be issued through the drone ground station or the drone remote control. When the gimbal is a handheld gimbal, the gimbal inertia identification command can be issued through the handheld gimbal button.

[0050] In step S200, the gimbal is driven to rotate in a preset manner, and angular velocity and current data corresponding to the rotation in the preset manner are collected.

[0051] Specifically, as the gimbal rotates in a preset manner, the angular velocity corresponding to the preset rotation is collected, and the angular velocity differential value of the gimbal rotating in the preset manner can be further obtained. Simultaneously, current data corresponding to the preset rotation is collected, and sample data of the angular velocity differential value and current data corresponding to the gimbal rotating in the preset manner can be obtained. Subsequently, this sample data is fitted using a fitting algorithm to obtain the gimbal's inertia, thereby improving the efficiency of gimbal control parameter adjustment. Furthermore, each time the gimbal changes its load, the gimbal inertia identification process of this embodiment is run once to obtain the gimbal inertia after the load change, thereby improving the gimbal's load adaptability and making gimbal products using this gimbal inertia identification method more competitive in the market.

[0052] Optionally, the angular velocity corresponding to the rotation in the preset manner is collected by a speed sensor capable of measuring the speed of the motor rotor. Further optionally, the angular velocity corresponding to the rotation in the preset manner is collected by a gyroscope.

[0053] Optionally, driving the pan / tilt platform to rotate in a preset manner includes: driving the pan / tilt platform to rotate in a preset manner through a square wave signal.

[0054] Specifically, by setting the amplitude and period of the square wave according to the specific pan-tilt structure, the pan-tilt can be driven to rotate in a preset manner through the square wave signal.

[0055] In one embodiment, please refer to Figure 2 , Figure 2 This is a flow chart of another gimbal inertia identification method provided by an embodiment of the present invention, wherein driving the gimbal to rotate according to a preset manner and collecting angular velocity and current data corresponding to the preset rotation manner includes:

[0056] Step S211: Setting the angular velocity target of the first axis of the gimbal to 0 and maintaining the target for one step period, while simultaneously collecting quadrature-axis current data of the first motor corresponding to the first axis throughout the entire maintenance process, and the angular velocity of the first motor at the end of the maintenance process, wherein the step period is greater than or equal to the stabilization time of the angular velocity loop step response corresponding to the first axis;

[0057] Step S212: Set the positive step speed target of the first axis to n*ω r1 , maintaining a step cycle, and simultaneously collecting the motor quadrature-axis current data of the first motor during the entire maintaining process, and the angular velocity of the first motor at the end of the maintaining process, wherein the initial value of n is 1;

[0058] Step S213, setting the angular velocity target of the first axis to 0 and maintaining it for one step cycle, while collecting the motor quadrature-axis current data of the first motor during the entire maintenance process and the angular velocity of the first motor at the end of the maintenance process;

[0059] Step S214: Set the negative step speed target of the first axis to -n*ω r1 , while collecting the motor quadrature-axis current data of the first motor during the entire maintenance process, and the angular velocity of the first motor at the end of the maintenance process;

[0060] In step S215, n is incremented by 1. If n is less than a first preset value, the process returns to setting the angular velocity target of the first axis of the gimbal to 0 and maintaining it for one step period. At the same time, the quadrature-axis current data of the first motor corresponding to the first axis during the entire maintenance process and the angular velocity of the first motor at the end of the maintenance process are collected. If n is greater than or equal to the first preset value, the process executes the operation of calculating the inertia parameters of the gimbal using a fitting algorithm according to the collected angular velocity and current data corresponding to the preset rotation method.

[0061] In this embodiment, it should be noted that, during the process of studying the control parameter tuning method of the gimbal, the inventors found that the gimbal generally operates in a low speed range and the dynamic friction of the gimbal is almost negligible. Therefore, a simplified gimbal model can be established as follows:

[0062]

[0063] Where a is the angular acceleration in rad / s; b is the inertia parameter in rad / s / ma; i is the quadrature axis current i that generates the torque. q , the unit is ma; d is the load parameter, It is an inertial link for simplified gyroscope sampling and mechanical response.

[0064] Since the angular acceleration a cannot be measured directly, and directly sampling the angular acceleration a at a certain moment is easily affected by burr noise, the differential value of the angular velocity ω is selected as Δω=ω t1 -ω t0 , where ω t1 is the gyroscope angular velocity value in the stable stage of step response, ω t0 is the gyroscope angular velocity value at the beginning of the step response. The adjustment model is:

[0065]

[0066] Where ∑i*Δt is the integral value of the entire step response current.

[0067] By selecting the angular velocity value of the stable stage of the sampling step response during the sampling process, the stabilization time of the attitude angular velocity loop step response is generally much longer than that of the inertia link. The time constant T ω , so the inertia link can be ignored The model is finally simplified to:

[0068] Δω=b*∑i*Δt+∑d*Δt (Formula 1-3)

[0069] Based on this, a method is proposed in this embodiment. Figure 3 The pan-tilt system shown includes a pan-tilt 1 and a host computer 2 that is connected to the pan-tilt for sending pan-tilt inertia identification instructions. The pan-tilt includes a motor 11, a motor board 12 connected to the motor for controlling the motor, and a posture board 13 connected to the motor board for controlling the pan-tilt attitude. The posture board includes posture board software 131, a FLASH 132 for storing inertia data, and an IMU (Inertial Measurement Unit) 133 for collecting angular velocities corresponding to a preset rotation mode. The FLASH 132 and the IMU 133 are respectively connected to the posture board software 131. The posture board software 131 includes a speed target generator 1311 for generating a target angular velocity, an angular velocity controller 1312 for converting the target angular velocity into a current target value to control the motor board, and a fitting algorithm 1313 for performing fitting operations. Based on the pan-tilt attitude angular velocity closed loop, it is necessary to first simply debug a basically stable angular velocity closed loop; design a speed target generator for inertia identification to generate positive and negative step targets ±n*ω according to a set step period. r1The output is fed into the attitude angular velocity closed loop, with n gradually increasing until it reaches a first preset value. During each step cycle, the angular velocity during the stable phase of the step response is sampled. The angular velocity during the stable phase of the previous cycle is subtracted to obtain Δω, and the current is integrated to obtain ∑i / Δt. A single Δω and ∑i*Δt form a complete sample. By adjusting the target angular velocity value, sufficiently dispersed sample data can be collected. Once sufficient sample data is collected, the fitting process is initiated to calculate the inertia parameters of each gimbal axis.

[0070] Specifically, in step S212, the angular velocity target of the first axis of the gimbal is set to 0 via a velocity target generator and maintained for one step cycle. Simultaneously, quadrature-axis current data of the first motor corresponding to the first axis during the entire maintenance process and the angular velocity of the first motor at the end of the maintenance process are collected. The step cycle is greater than or equal to the stabilization time of the angular velocity loop step response corresponding to the first axis. This allows for the acquisition of the angular velocity of the first axis when the angular velocity loop step response stabilizes and quadrature-axis current data of the first motor corresponding to the first axis during the entire maintenance process when the angular velocity target is 0. The step cycle being greater than or equal to the stabilization time of the angular velocity loop step response corresponding to the first axis is intended to allow for the motor speed response. For example, when the positive and negative step targets are square wave signals, the signals change very quickly. After receiving the positive and negative step targets, the motor needs time to accelerate to the target speed. Furthermore, the motor must maintain the target speed for at least one step cycle before reaching the target speed. Therefore, it is necessary to sample the angular velocity of the first motor at the end of the maintenance process.

[0071] In step S212, the positive step speed target of the first axis is set to n*ω r1 , maintain a step cycle, and collect the motor quadrature axis current data of the first motor during the entire maintenance process, and the angular velocity of the first motor at the end of the maintenance process, wherein the initial value of n is 1. Thus, the angular velocity target changes from 0 to n*ω r1 The angular velocity of the first axis when the angular velocity loop step response is stable and the motor quadrature-axis current data samples of the first motor corresponding to the first axis during the entire maintenance process.

[0072] In step S213, the angular velocity target of the first axis is set to 0 and maintained for one step cycle, while collecting the motor quadrature axis current data of the first motor during the entire maintenance process and the angular velocity of the first motor at the end of the maintenance process. r1 When becomes 0, the angular velocity of the first axis when the angular velocity loop step response is stable and the motor quadrature-axis current data samples of the first motor corresponding to the first axis during the entire maintenance process.

[0073] In step S214, the negative step speed target of the first axis is set to -n*ω r1At the same time, the quadrature axis current data of the first motor during the entire maintenance process and the angular velocity of the first motor at the end of the maintenance process are collected. Thus, the angular velocity target is changed from 0 to -n*ω r1 The angular velocity of the first axis when the angular velocity loop step response is stable and the motor quadrature-axis current data samples of the first motor corresponding to the first axis during the entire maintenance process.

[0074] In step S215, n is incremented by 1. If n is less than a first preset value, the process returns to the step of setting the angular velocity target of the first axis of the gimbal to 0 and maintaining the target for one step cycle. Meanwhile, the process collects quadrature-axis current data of the first motor corresponding to the first axis throughout the entire maintenance process, as well as the angular velocity of the first motor at the end of the maintenance process. If n is greater than or equal to the first preset value, the process calculates the inertia parameters of the gimbal using a fitting algorithm based on the collected angular velocity and current data corresponding to the preset rotation method. Thus, by accumulating n from 1 to the first preset value, samples of the angular velocity of the first axis corresponding to multiple step velocity targets when the angular velocity loop step response is stable, as well as quadrature-axis current data of the first motor corresponding to the first axis throughout the entire maintenance process, are sequentially collected. These sample data are then fitted using the fitting algorithm to obtain the inertia of the gimbal. Gimbals are categorized as single-axis, dual-axis, and three-axis. This embodiment uses the first axis of a gimbal as an example to illustrate the gimbal inertia identification process. Those skilled in the art will appreciate that, when the gimbal product is a dual-axis or three-axis gimbal, the same inertia identification process can be performed on each axis in sequence to complete the inertia identification of each axis.

[0075] Optionally, before setting the angular velocity target of the first axis of the gimbal to 0 and maintaining a step cycle, and collecting the motor quadrature-axis current data of the first motor corresponding to the first axis during the entire maintenance process, and the angular velocity of the first motor at the end of the maintenance process, the method also includes: closing the attitude angle closed loop corresponding to the first axis of the gimbal.

[0076] Specifically, closing the attitude angle closed loop corresponding to the first axis of the gimbal can avoid the influence of the attitude angle closed loop corresponding to the first axis on the inertia identification process of the first axis.

[0077] In one embodiment, please refer to Figure 4 , Figure 4 This is a flow chart of another gimbal inertia identification method provided by an embodiment of the present invention. When the gimbal includes a roll axis, a pitch axis, and a yaw axis, the driving gimbal to rotate according to a preset method and collecting angular velocity and current data corresponding to the preset rotation method include:

[0078] In step S221, the roll, pitch, and yaw axes of the gimbal are sequentially driven to rotate according to a preset method, the other two axes that are not currently driven are set to a mechanical angle servo mode with a mechanical angle target of 0, and the angular velocity and current data corresponding to the rotation of the currently driven axis according to the preset method are collected.

[0079] In this embodiment, when the gimbal is a three-axis gimbal including a roll axis, a pitch axis, and a yaw axis, the roll axis, pitch axis, and yaw axis of the gimbal are driven in sequence to rotate in a preset manner, the other two axes that are not currently driven are set to a mechanical angle servo mode with a mechanical angle target of 0, and the angular velocity and current data corresponding to the rotation of the currently driven axis in the preset manner are collected, and the angular velocity and current data corresponding to the rotation of the three axes in the preset manner can be obtained in sequence. It can be understood by those skilled in the art that the step period, the first preset value, the step value ω in the positive step velocity target, and the negative step velocity target corresponding to different axes r1 They can be the same or different, and the step period corresponding to each axis must be greater than or equal to the stabilization time of the corresponding angular velocity loop step response. If the step period is too large, the gimbal will easily hit the mechanical limit. r1 The setting of needs to consider the sensor accuracy limit and the limit of the gimbal mechanical structure. The larger the first preset value, the more sample data, which is more conducive to the recognition accuracy. However, a too large first preset value requires a larger rotation range; the step value ω r1 The smaller the value, the better it is for narrowing the rotation range, but too small a step value ω r1 Higher sensor accuracy will be required, so the step period, first preset value and step value ω need to be reasonably set according to the sensor accuracy and mechanical limit of the specific application scenario. r1 .

[0080] In step S300, the inertia parameters of the gimbal are calculated using a fitting algorithm based on the collected angular velocity and current data corresponding to the preset rotation mode.

[0081] Specifically, the gimbal's inertia parameters are obtained by fitting the collected angular velocity and current data corresponding to the preset rotation pattern through a fitting algorithm, thereby improving the efficiency of gimbal control parameter tuning. Furthermore, each time the gimbal's load is changed, the gimbal inertia identification process of this embodiment is run again to obtain the gimbal's inertia after the load change, thereby improving the gimbal's load adaptability and making gimbal products using this gimbal inertia identification method more competitive in the market.

[0082] Optionally, the fitting algorithm includes a gradient descent method or a least squares method. Those skilled in the art will appreciate that the fitting algorithm may also employ other fitting algorithms that meet the gimbal inertia identification error requirements, and this embodiment does not limit the specific fitting algorithm type.

[0083] In one embodiment, please refer to Figure 5 , Figure 5 This is a flow chart of another gimbal inertia identification method provided by an embodiment of the present invention. When the fitting algorithm is a gradient descent method, the calculation of the gimbal inertia parameters using the fitting algorithm based on the collected angular velocity and current data corresponding to the preset rotation method includes:

[0084] Step S311: Integrate the motor quadrature-axis current data collected in each step cycle to obtain a current integral value, and subtract the angular velocity collected in the previous step cycle from the angular velocity collected in the corresponding step cycle to obtain an angular velocity differential value. One angular velocity differential value and one current integral value are considered as a complete sample data.

[0085] In step S312, the sample data is fitted using the gradient descent method. If the loss function is less than a preset threshold within a preset number of gradient descent steps, it is determined that the inertia identification of the first axis of the gimbal is successful, and the inertia parameters of the first axis obtained by fitting are saved.

[0086] In this embodiment, sample data is collected by integrating the motor quadrature-axis current data collected during each step cycle to obtain a current integral value. The angular velocity difference value is then subtracted from the angular velocity collected during the previous step cycle to obtain the angular velocity differential value. The collected sample data is then fitted using a gradient descent method. If the loss function is less than a preset threshold within a preset number of gradient descent steps, the inertia identification of the gimbal's first axis is determined to be successful, and the fitted inertia parameters of the first axis are saved. Compared to sampling schemes that sample current and angular acceleration values ​​at a single moment, this embodiment samples the superposition of angular velocity data and current during the stable phase of the step process. This scheme can better filter out sample noise and is unaffected by lags in the model, thereby improving identification accuracy. Furthermore, this embodiment simplifies the model structure based on the characteristics of the gimbal model, reducing algorithm complexity and making it more suitable for engineering implementation.

[0087] Optionally, if the loss function is not less than a preset threshold within a preset number of gradient descent steps, the inertia identification process of the first axis of the gimbal is exited.

[0088] In one embodiment, please refer to Figure 4 When the gimbal includes a roll axis, a pitch axis, and a yaw axis, the inertia parameters of the gimbal calculated using a fitting algorithm based on the collected angular velocity and current data corresponding to the preset rotation method include:

[0089] Step S321: After completing the collection of the angular velocity and current data corresponding to the three axes, the inertia parameters corresponding to each axis are calculated using a fitting algorithm based on the angular velocity and current data corresponding to the preset rotation method collected for each axis; or, after completing the collection of the angular velocity and current data corresponding to each axis, the inertia parameters corresponding to the axis currently completing the collection are calculated using a fitting algorithm based on the angular velocity and current data corresponding to the preset rotation method collected for the axis currently completing the collection.

[0090] In this embodiment, when the gimbal includes a roll axis, a pitch axis, and a yaw axis, the inertia parameters of each axis of the gimbal can be calculated using a fitting algorithm after the collection of the angular velocity and current data corresponding to the three axes is completed. Alternatively, the inertia parameters of the axis of the gimbal that has currently completed the collection can be calculated using a fitting algorithm each time the collection of the angular velocity and current data corresponding to an axis is completed.

[0091] Optionally, after calculating the inertia parameters of the gimbal using a fitting algorithm based on the collected angular velocity and current data corresponding to the preset rotation mode, the inertia identification process of the gimbal is exited.

[0092] Specifically, after the inertia identification of all axes of the gimbal is completed, the gimbal inertia identification process is exited so that the gimbal can perform subsequent work processes.

[0093] In an embodiment of the present invention, a gimbal inertia identification command is received; the gimbal is driven to rotate in a preset manner, and angular velocity and current data corresponding to the preset rotation are collected; and based on the collected angular velocity and current data corresponding to the preset rotation, a fitting algorithm is used to calculate the gimbal's inertia parameters. The gimbal inertia identification method of this embodiment drives the gimbal to rotate in a preset manner and collects angular velocity and current data corresponding to the preset rotation. After data sampling, a fitting algorithm is used to automatically calculate the gimbal's moment of inertia using the sample data. This allows the gimbal to automatically identify its own moment of inertia, improves gimbal design efficiency, and enables the gimbal to adapt to load changes, thereby enhancing the market competitiveness of gimbal products.

[0094] Example 2

[0095] like Figure 6 As shown, an embodiment of the present invention further provides a gimbal 600, which includes a memory 601, a processor 602, and a computer program (not shown in the figure) stored in the memory 601 and executable on the processor 602. When the computer program is executed by the processor 602, the steps of the gimbal inertia identification method described in the above-mentioned embodiment 1 are implemented.

[0096] The gimbal of the embodiment of the present invention and the gimbal inertia identification method of the above-mentioned embodiment 1 have the same concept. The specific implementation process is detailed in the corresponding method embodiment, and the technical features in the method embodiment are applicable to the present gimbal embodiment and will not be repeated here.

[0097] Example 3

[0098] An embodiment of the present invention further provides a computer-readable storage medium storing a gimbal inertia identification program. When the gimbal inertia identification program is executed by a processor, the steps of the gimbal inertia identification method described in the first embodiment are implemented.

[0099] The computer-readable storage medium of the embodiment of the present invention and the method of the above-mentioned embodiment 1 belong to the same concept. The specific implementation process is detailed in the corresponding method embodiment, and the technical features in the method embodiment are applicable in this computer-readable storage medium embodiment, which will not be repeated here.

[0100] The corresponding technical features in the above-mentioned embodiments can be used interchangeably without causing contradiction or impracticality of the solutions.

[0101] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus comprising the element.

[0102] The serial numbers of the above embodiments of the present invention are for description only and do not represent the advantages or disadvantages of the embodiments.

[0103] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better embodiment. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0104] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are protected by the present invention.

Claims

1. A gimbal inertia identification method, characterized in that: The method comprises: Receive gimbal inertia identification command; Drive the gimbal to rotate according to a preset method and collect the angular velocity and current data corresponding to the preset rotation method; Calculate the inertia parameters of the gimbal using a fitting algorithm based on the collected angular velocity and current data corresponding to the preset rotation mode; The driving of the gimbal to rotate in a preset manner and collecting angular velocity and current data corresponding to the rotation in the preset manner include: The angular velocity target of the first axis of the gimbal is set to 0 and maintained for one step period. Meanwhile, quadrature-axis current data of the first motor corresponding to the first axis during the entire maintenance process and the angular velocity of the first motor at the end of the maintenance process are collected. The step period is greater than or equal to the settling time of the step response of the angular velocity loop corresponding to the first axis. Set the positive step speed target of the first axis to n*ω r1 , maintaining a step cycle, and simultaneously collecting the motor quadrature-axis current data of the first motor during the entire maintaining process, and the angular velocity of the first motor at the end of the maintaining process, wherein the initial value of n is 1; Setting the angular velocity target of the first axis to 0 and maintaining it for one step cycle, while collecting the quadrature-axis current data of the first motor during the entire maintenance process and the angular velocity of the first motor at the end of the maintenance process; Set the negative step speed target of the first axis to -n*ω r1 , while collecting the motor quadrature-axis current data of the first motor during the entire maintenance process, and the angular velocity of the first motor at the end of the maintenance process; n is incremented by 1. If n is less than a first preset value, the process returns to setting the angular velocity target of the first axis of the gimbal to 0 and maintaining it for one step cycle, while simultaneously collecting the motor quadrature-axis current data of the first motor corresponding to the first axis during the entire maintenance process, and the angular velocity of the first motor at the end of the maintenance process. If n is greater than or equal to the first preset value, executing the operation of rotating the gimbal according to the preset rotation method and the corresponding angular velocity and current data collected, and calculating the inertia parameters of the gimbal using a fitting algorithm.

2. The method for identifying the inertia of a pan / tilt platform according to claim 1, wherein: The receiving of the gimbal inertia identification instruction includes receiving the gimbal inertia identification instruction through a UAV ground station, a UAV remote controller or a handheld gimbal button.

3. The method for identifying the inertia of a pan / tilt platform according to claim 1, wherein: Before setting the angular velocity target of the first axis of the gimbal to 0 and maintaining a step cycle, and simultaneously collecting motor quadrature-axis current data of the first motor corresponding to the first axis during the entire maintenance process, and the angular velocity of the first motor at the end of the maintenance process, the method further includes: Close the attitude angle closed loop corresponding to the first axis of the gimbal.

4. The method for identifying the inertia of a pan / tilt platform according to claim 1, wherein: The fitting algorithm includes a gradient descent method or a least squares method. When the fitting algorithm is the gradient descent method, the calculation of the inertia parameters of the gimbal using the fitting algorithm based on the collected angular velocity and current data corresponding to the preset rotation method includes: The motor quadrature-axis current data collected in each step cycle is integrated to obtain the current integral value, and the angular velocity collected in the corresponding step cycle is subtracted from the angular velocity collected in the previous step cycle to obtain the angular velocity differential value. An angular velocity differential value and a current integral value are regarded as a complete sample data; The sample data is fitted using the gradient descent method. If the loss function is less than the preset threshold within the preset number of gradient descent steps, the inertia identification of the first axis of the gimbal is determined to be successful, and the inertia parameters of the first axis obtained by fitting are saved.

5. The method for identifying the inertia of a pan / tilt platform according to claim 1, wherein: When the gimbal includes a roll axis, a pitch axis, and a yaw axis, driving the gimbal to rotate according to a preset method and collecting angular velocity and current data corresponding to the rotation in the preset method include: The roll, pitch, and yaw axes of the gimbal are driven in sequence to rotate according to the preset method. The other two axes that are not currently driven are set to the mechanical angle servo mode with a mechanical angle target of 0. The angular velocity and current data corresponding to the currently driven axis rotating according to the preset method are collected.

6. The method for identifying the inertia of a pan / tilt platform according to claim 5, characterized in that: Calculating the inertia parameters of the gimbal using a fitting algorithm based on the collected angular velocity and current data corresponding to the preset rotation method includes: After completing the collection of the angular velocity and current data corresponding to the three axes, the angular velocity and current data corresponding to each axis are rotated according to the preset rotation method collected, and the inertia parameters corresponding to each axis are calculated using a fitting algorithm; or, after completing the collection of the angular velocity and current data corresponding to each axis, the inertia parameters corresponding to the axis currently completing the collection are calculated using a fitting algorithm based on the angular velocity and current data corresponding to the preset rotation method collected for the axis currently completing the collection.

7. The method for identifying the inertia of a pan / tilt platform according to claim 1, wherein: Driving the pan / tilt platform to rotate in a preset manner includes: The square wave signal drives the gimbal to rotate in a preset manner.

8. A pan / tilt head, characterized in that: The gimbal includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, the steps of the gimbal inertia identification method according to any one of claims 1 to 7 are implemented.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a gimbal inertia identification program, which, when executed by a processor, implements the steps of the gimbal inertia identification method according to any one of claims 1 to 7.

Images