A method and system for dynamic space reconstruction of a six-degree-of-freedom motion platform

By predicting interference risks and dynamically correcting motion perception weights, a dynamic safe workspace is constructed, and a six-degree-of-freedom motion platform is adaptively controlled. This solves the problems of motion continuity and somatosensory fidelity in existing technologies when motion platforms avoid actuator interference, and improves the realism and control smoothness of simulation training.

CN122837256BActive Publication Date: 2026-11-17CHINA SOUTHERN TECHNOLOGY (GUANGDONG HENGQIN) CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202611327159.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2026-08-31
Publication Date
2026-11-17
Estimated Expiration
2046-08-31

AI Technical Summary

Technical Problem

Existing six-degree-of-freedom motion platforms struggle to balance motion continuity and somatosensory fidelity while mitigating actuator interference risks. Existing control methods are prone to discontinuities and disruptions of somatosensory characteristics in highly dynamic and complex motions, affecting the realism and effectiveness of simulation training.

Method used

By acquiring the platform's operating status and motion prompts, potential interference risks are predicted, the risk contribution of degrees of freedom is calculated, motion perception weights are dynamically adjusted, a dynamic safety workspace is constructed, adaptive control reconstruction is performed, and target safety control commands are generated to ensure motion continuity and motion perception fidelity.

Benefits of technology

It achieves the maximum preservation of motion cues while avoiding actuator interference, improves control smoothness and haptic fidelity, reduces the disruption of the original haptic experience caused by safety reconfiguration, and enhances the realism of flight simulation and the consistency of training.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122837256B_ABST
    Figure CN122837256B_ABST
Patent Text Reader

Abstract

This application belongs to the field of flight simulator control technology, and relates to a dynamic space reconstruction method and system for a six-degree-of-freedom motion platform. It aims to solve the problem in existing technologies where it is difficult to balance motion continuity and motion perception fidelity when avoiding interference risks. The method includes: acquiring operational status information and original motion prompts; predicting the potential spatial interference risks of each actuator and generating a predicted interference risk index; applying perturbations to each degree-of-freedom direction, calculating the contribution of each degree-of-freedom direction to the predicted interference risk, and constructing a degree-of-freedom risk feature set; correcting the initial motion perception weights based on the feature set, forming a dynamic motion perception weight matrix and constructing a safe workspace; and reconstructing the original motion prompts based on the predicted interference risk index and the dynamic motion perception weight matrix under the constraints of the safe workspace, generating target safety control commands. This application balances motion control continuity and motion perception fidelity while avoiding actuator interference risks.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application belongs to the field of flight simulator control technology, and relates to data processing technology for flight simulators or complex simulation systems, and particularly to a dynamic space reconstruction method and system for a six-degree-of-freedom motion platform. Background Technology

[0002] Six-degree-of-freedom motion platforms, especially systems based on parallel mechanisms (such as the Stewart platform), are core components of flight simulators, pilot simulation systems, and various high-end equipment simulation training devices. Through the coordinated extension and retraction of multiple actuators, they can accurately reproduce six-degree-of-freedom spatial motion of translation and attitude, providing operators with highly realistic haptic feedback.

[0003] In existing technologies, the control systems of six-degree-of-freedom motion platforms, while pursuing precise tracking of motion commands, also integrate multiple safety protection mechanisms to avoid hardware damage. One is passive protection based on limit thresholds, which sets fixed boundaries for parameters such as actuator length, speed, acceleration, or platform pose. When a command or actual state is detected to approach or touch the boundary, amplitude limiting, clipping, or emergency shutdown is triggered. However, this method is essentially reactive. When the platform performs complex motions with high dynamics and strong coupling of multiple degrees of freedom, the control system struggles to intervene smoothly before risks occur. Sudden and abrupt command clipping can easily lead to discontinuities and unevenness in platform motion, disrupting the haptic continuity of the simulated motion.

[0004] Secondly, some existing technologies introduce geometric models to detect the spatial distance between actuators or use predictive models to estimate the future state of the platform. However, when risks are identified, traditional methods often perform indiscriminate or coarse overall compression of all degrees of freedom. This lacks both a multi-degree-of-freedom coordination and reconstruction mechanism for the overall workspace of the platform and an analysis of the differences in the contribution of different degrees of freedom directions to interference risk. This limits the full utilization of the platform's remaining available motion capabilities, making it difficult to achieve refined and optimized control strategies.

[0005] More importantly, in high-fidelity applications such as flight simulation, existing safety control methods focus excessively on mechanical safety and rarely take into account the perceptual fidelity of motion cues. This can lead to safety interventions potentially damaging the somatosensory characteristics of the original motion cues and reducing the realism and effectiveness of simulation training.

[0006] Therefore, how to proactively avoid the risk of actuator interference while ensuring the continuity of motion control and the fidelity of motion perception has become a pressing technical challenge in this field. Summary of the Invention

[0007] To address the aforementioned problems in the prior art, namely the difficulty in balancing motion continuity and motion fidelity while mitigating interference risks, the first aspect of this application proposes a dynamic space reconstruction method for a six-degree-of-freedom motion platform, the method comprising the following steps: Acquire the operating status information and original motion prompts of the six-degree-of-freedom motion platform, and obtain the initial motion perception weights to reflect the importance of different degrees of freedom to the motion perception. Based on the operational status information and the original motion prompts, the potential interference state of the platform in the prediction time domain is predicted, and a predicted interference risk index is generated. Positive and negative virtual numerical perturbations are applied to each degree of freedom direction of the platform in the candidate control command space to calculate the risk contribution of each degree of freedom direction to the predicted interference risk index, and the risk contribution of each degree of freedom direction is normalized to obtain the degree of freedom risk feature set. Based on the normalized risk contribution of each degree of freedom direction in the set of degree of freedom risk features, the initial motion perception weights are dynamically corrected to obtain a dynamic motion perception weight matrix, and a dynamic safe workspace is constructed based on the dynamic motion perception weight matrix. Under the constraints of the dynamic safety workspace, the original motion prompting command is adaptively reconstructed based on the predicted interference risk index and the dynamic motion perception weight matrix to generate the target safety control command. The target safety control command is converted into an actuator length command and output to drive the six-degree-of-freedom motion platform to perform closed-loop actions.

[0008] In some preferred embodiments, generating predictive intervention risk indicators includes: A sliding time window is used to extract a historical state sequence constructed based on the running state information and a historical instruction sequence constructed based on the original motion prompt instruction, and then synthesizes the input time sequence. The input time series is input into a preset time series prediction model to calculate the platform prediction pose at each prediction time in the prediction time domain; Based on the platform's predicted pose at each prediction time, the predicted length of each actuator at each prediction time is calculated through the platform's inverse kinematics mapping. Based on the platform's structural parameters and the predicted spatial positions of the hinge points at both ends of each actuator, the minimum spatial distance between any two actuators at each prediction time is calculated. Based on a preset safety distance threshold, the minimum spatial distance is nonlinearly mapped to the interference risk value at each prediction time. The maximum value of all interference risk values ​​generated within the prediction time domain is extracted as the peak risk parameter, and the interference risk values ​​at each time point within the prediction time domain are summed in the time domain to obtain the cumulative risk parameter. The peak risk parameter and the cumulative risk parameter are linearly weighted to obtain the comprehensive interference risk of the corresponding actuator pair. The maximum value of the comprehensive interference risk among all actuator pairs is extracted and established as the final predicted interference risk index.

[0009] In some preferred embodiments, positive and negative virtual numerical perturbations are applied to each degree of freedom direction of the platform within the candidate control command space to calculate the risk contribution of each degree of freedom direction to the predicted interference risk index. The risk contribution of each degree of freedom direction is then normalized to obtain a degree-of-freedom risk feature set, specifically including: In the three translational degrees of freedom and the three attitude degrees of freedom of the six-degree-of-freedom motion platform, respectively, a preset micro-scale positive and negative virtual numerical perturbation is superimposed on the original motion prompt command. Calculate the predicted interference risk index corresponding to the superposition of the positive and negative virtual numerical perturbations respectively, and take the ratio of the risk difference between the two to twice the perturbation amplitude as the risk contribution degree in the corresponding degree of freedom direction; The risk contribution in each degree of freedom direction is normalized to obtain the normalized risk contribution in each degree of freedom direction. The normalized risk contributions in each degree of freedom direction are then combined to form a degree of freedom risk feature set.

[0010] In some preferred embodiments, the initial motion sensing weights are dynamically corrected based on the set of freedom risk features to obtain a dynamic motion sensing weight matrix, and a dynamic safe workspace is constructed based on the dynamic motion sensing weight matrix, specifically including: By introducing a weight adjustment coefficient and combining the normalized risk contribution of each degree of freedom direction in the set of degree of freedom risk features, the initial motion perception weight of the corresponding degree of freedom direction is dynamically corrected to generate dynamic motion perception weight of each degree of freedom direction, and the dynamic motion perception weight of each degree of freedom direction is combined to form a dynamic motion perception weight matrix. Within the globally feasible control instruction space determined by the platform actuator extreme values, regions that do not meet the safety reconfiguration requirements are eliminated using a preset interference risk threshold to define a safe and feasible domain. Regions with excessive motion perception distortion are eliminated using a preset motion perception distortion threshold combined with the dynamic motion perception weight matrix to define a motion perception fidelity domain. The intersection of the safe and feasible domain and the motion perception fidelity domain constitutes the dynamic safe workspace.

[0011] In some preferred embodiments, the dynamic safe workspace is represented as a subset of control instructions that simultaneously satisfy the following mathematical set constraints: The control commands to be reconfigured are located within the globally feasible control command space defined by the size of the mechanism and the mechanical limits of the hardware; The predicted interference risk index caused by the control command to be reconstructed is not greater than the preset interference risk threshold, and the motion perception distortion degree of the control command to be reconstructed, which is weighted by the dynamic motion perception weight matrix, is not greater than the preset perception distortion threshold.

[0012] In some preferred embodiments, generating target security control instructions specifically includes: A multi-objective optimization function is constructed, which is formed by weighted summation of evaluation cost sub-items from multiple dimensions. The evaluation cost sub-items include at least: a tracking error term used to constrain the approximation degree of the reconstruction command to the original motion prompt command; a prediction interference risk term used to suppress the hazards of mechanical collisions; a motion perception distortion term used to weight the control reconstruction deviation in different degrees of freedom directions based on the dynamic motion perception weight matrix; and a control smoothing term measured by the change in the amount of change between two adjacent control cycles. Under the conditions of the dynamic safety workspace and the physical constraint boundary of the actuator, the reconstructed target safety control command is obtained by solving the control variables that minimize the multi-objective optimization function in real time.

[0013] In some preferred embodiments, the actuator physical constraint boundary includes at least: The working length of each actuator corresponding to the reconstructed target safety control command calculated in real time is within the range defined by the minimum and maximum lengths of the corresponding actuators; The absolute values ​​of the instantaneous extension speed and extension acceleration of each actuator corresponding to the target safety control command calculated in real time are not greater than the maximum extension speed and maximum extension acceleration allowed by the hardware, respectively.

[0014] In some preferred embodiments, generating target security control instructions includes: Based on the normalized risk contribution of each degree of freedom in the set of risk features and the dynamic motion perception weight matrix, a safe scaling matrix with diagonal features is calculated and generated, wherein each diagonal element of the safe scaling matrix is ​​determined by the normalized risk contribution and the dynamic motion perception weight in the corresponding degree of freedom direction. Based on the normalized risk contribution of each degree of freedom in the set of risk features, a safety bias vector is generated to guide the platform's motion pose to a low-risk motion region within the dynamic safe workspace. Based on the safety scaling matrix of the degrees of freedom and the safety bias vector, a linear affine transformation is performed on the original motion prompt command so that the transformed control command is located in the low-risk motion region within the dynamic safety workspace, thereby generating the target safety control command.

[0015] In some preferred embodiments, before outputting the actuator length instruction: The actuator length command calculated by conversion is imported into a first-order hysteresis filter. Using the historical output actuator length command and smoothing filter coefficients of the previous cycle, the current actuator length command is subjected to weighted smoothing filtering to avoid abrupt operations. The filtered final state value is output to the underlying mechanical actuator to update the closed-loop motion state and start the next predictive control cycle.

[0016] A second aspect of this application proposes a six-degree-of-freedom motion platform dynamic space reconstruction system, comprising: The status acquisition and instruction parsing module is configured to acquire the running status information and original motion prompt instructions of the six-degree-of-freedom motion platform, and to acquire the initial motion perception weights that reflect the importance of different degrees of freedom to the motion sensation. The interference risk prediction and assessment module is configured to predict the potential interference state of the platform in the prediction time domain based on the operating status information and the original motion prompt instructions, and generate a predicted interference risk index. The interference sensitivity analysis module is configured to apply positive and negative virtual numerical perturbations to each degree of freedom direction in the candidate control command space, calculate the risk contribution of each degree of freedom direction to the predicted interference risk index, and normalize the risk contribution to obtain the degree of freedom risk feature set. The workspace construction module is configured to dynamically correct the initial motion perception weights based on the normalized risk contribution of each degree of freedom direction in the set of degree of freedom risk features, to obtain a dynamic motion perception weight matrix, and to construct a dynamic safe workspace based on the dynamic motion perception weight matrix. The adaptive reconstruction module is configured to perform adaptive control reconstruction of the original motion prompting command based on the predicted interference risk index and the dynamic motion perception weight matrix under the constraints of the dynamic safety workspace, and generate a target safety control command. The instruction output module is configured to convert the target safety control instruction into an actuator length instruction and output it to drive the six-degree-of-freedom motion platform to perform actions.

[0017] Compared with the prior art, the technical solutions in the embodiments provided in this application have at least the following beneficial effects: 1) By deeply coupling the forward-looking prediction of actuator interference risk, the risk contribution characteristics of each degree of freedom and the motion perception weight, a dynamic safe workspace and a multi-objective adaptive reconstruction mechanism are constructed. Under the premise of ensuring that the motion platform avoids actuator interference, the realism of motion prompts is preserved to the greatest extent, which helps to achieve the adaptive unity of operational safety, motion continuity and high-fidelity of motion perception.

[0018] 2) By constructing a predictive time domain in the time dimension and extracting peak and cumulative comprehensive risk parameters, proactive and forward-looking avoidance of actuator interference risk is achieved, enabling the control system to accurately capture the potential dangerous motion trend of the platform in the future period of time, thereby providing sufficient smooth correction margin before the platform really approaches the physical interference boundary. Compared with the existing technology, the control smoothness of the system in high dynamic complex motion scenarios is improved.

[0019] 3) By applying perturbations to each degree of freedom to quantify the risk contribution, and combining the initial motion perception weights to establish a dynamic safety workspace and adaptive allocation strategy, it can intelligently identify and prioritize the compression of motion components with high interference risk and low somatosensory weights, while adaptively retaining the high-weight motion features that are most important to the operator's somatosensory stimulation. This reduces the damage to the original flight / pilot simulation somatosensory experience caused by safety reconstruction from a mechanistic perspective, and increases the realism of the simulation and the consistency of training.

[0020] 4) By combining multi-objective optimization and extreme value finding with first-order hysteresis filtering, it can effectively suppress high-frequency noise and instruction jitter caused by complex inputs with multiple frequencies superimposed; and the perception weights are configurable, which can be widely adapted to different configurations of motion platforms and differentiated training subject requirements, and has engineering application value and scalability. Attached Figure Description

[0021] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a flowchart of a dynamic space reconstruction method for a six-degree-of-freedom motion platform provided in one embodiment of this application; Figure 2 This is a schematic diagram of the structure of a computer system used to implement the methods, systems, and electronic devices of this application. Detailed Implementation

[0022] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0023] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0024] Addressing the actuator interference problem that may occur in flight simulators, pilot simulation systems, and other six-degree-of-freedom motion platforms during highly dynamic, multi-degree-of-freedom coupled motion, this application provides a dynamic space reconstruction method for six-degree-of-freedom motion platforms. By comprehensively analyzing the platform's operating state, actuator state, and motion cue commands, it achieves proactive prediction of actuator interference risks, dynamic safe workspace reconstruction, and control command reconstruction for motion perception fidelity. The method includes: Acquire the operating status information and original motion prompts of the six-degree-of-freedom motion platform, and obtain the initial motion perception weights to reflect the importance of different degrees of freedom to the motion perception. Based on the operational status information and the original motion prompts, the potential interference state of the platform in the prediction time domain is predicted, and a predicted interference risk index is generated. Positive and negative virtual numerical perturbations are applied to each degree of freedom direction of the platform in the candidate control command space to calculate the risk contribution of each degree of freedom direction to the predicted interference risk index, and the risk contribution of each degree of freedom direction is normalized to obtain the degree of freedom risk feature set. Based on the normalized risk contribution of each degree of freedom direction in the set of degree of freedom risk features, the initial motion perception weights are dynamically corrected to obtain a dynamic motion perception weight matrix, and a dynamic safe workspace is constructed based on the dynamic motion perception weight matrix. Under the constraints of the dynamic safety workspace, the original motion prompting command is adaptively reconstructed based on the predicted interference risk index and the dynamic motion perception weight matrix to generate the target safety control command. The target safety control command is converted into an actuator length command and output to drive the six-degree-of-freedom motion platform to perform closed-loop actions.

[0025] To more clearly explain the dynamic space reconstruction method of the six-degree-of-freedom motion platform in this application, the following will combine... Figure 1 The steps in the embodiments of this application are described in detail.

[0026] The first embodiment of this application provides a dynamic space reconstruction method for a six-degree-of-freedom motion platform, including steps S1 to S6, as follows: Figure 1 As shown, each step is described in detail below: S1. Obtain the operating status information and original motion prompts of the six-degree-of-freedom motion platform, and obtain the initial motion perception weights to reflect the importance of different degrees of freedom to the motion perception.

[0027] Preferably, in one control cycle At any given moment, the operating state information of the six-degree-of-freedom motion platform is first acquired, including parameters such as platform pose, velocity, acceleration, length of each actuator, and actuator velocity, and then the platform state vector is constructed: ; in, Indicates the current discrete control cycle; The platform state vector represents the current control cycle; This represents the platform's six-degree-of-freedom pose vector; This represents the platform's six-degree-of-freedom velocity vector; This represents the platform's six-degree-of-freedom acceleration vector; This represents the length vector of the six actuators; This represents the vector of the six actuator extension / retraction speeds.

[0028] In one specific embodiment, the running status information includes: The platform's six-degree-of-freedom pose vector is: ; The platform's six-degree-of-freedom velocity vector is: ; The platform's six-degree-of-freedom acceleration vector is: ; The length vectors of the six actuators are: ; The extension and retraction velocity vectors of the six actuators are: ; in, 、 、 These represent the displacements of the platform in the three translational directions, respectively. 、 、 These represent the platform's roll, pitch, and yaw attitude angles, respectively. 、 、 These represent the linear velocities of the platform in the three translational directions, respectively. 、 、 These represent the angular velocities in the platform's roll, pitch, and yaw directions, respectively. 、 、 These represent the linear accelerations of the platform in the three translational directions, respectively. 、 、 These represent the angular accelerations in the platform's roll, pitch, and yaw directions, respectively. Indicates the first The length of the root actuator in the current control cycle, ; Indicates the first The scaling speed of the root actuator in the current control cycle. .

[0029] More preferably, while acquiring the operating status information of the six-degree-of-freedom motion platform, it also receives the raw motion cue instructions output by the upper-level motion cue generation module (such as the washout algorithm). This instruction is a six-DOF pose target that best reflects the ideal tactile effect without safety constraint correction.

[0030] In one specific embodiment, the original motion prompt instruction is: ; in, The original motion prompt command indicating the current control cycle; 、 、 These represent the desired translation commands; 、 、 These represent the desired attitude command.

[0031] More preferably, to describe the importance of different degrees of freedom to the haptic experience of flight simulation, preset initial motion perception weights are obtained: ; in, Represents the motion sensing weight matrix; 、 、 These represent the importance of the three translational degrees of freedom in motion perception fidelity; 、 、 These represent the importance of the roll, pitch, and yaw directions for motion perception fidelity, respectively. Each element in the matrix quantifies the corresponding degree of freedom (e.g., ...). Translate to The importance of haptic stimulation (such as roll) to the operator during simulated tasks. This weight matrix can be configured according to the airworthiness certification requirements of the flight trainer, specific training subjects, or human perception models.

[0032] S2. Based on the operating status information and the original motion prompts, predict the potential interference state of the platform in the prediction time domain and generate a predicted interference risk index.

[0033] Preferably, the generation of predictive intervention risk indicators includes: A sliding time window is used to extract a historical state sequence constructed based on the running state information and a historical instruction sequence constructed based on the original motion prompt instruction, and then synthesizes the input time sequence. The input time series is input into a preset time series prediction model to calculate the platform state at each prediction time in the prediction time domain; Based on the platform's predicted pose at each prediction time, the predicted length of each actuator at each prediction time is calculated through the platform's inverse kinematics mapping. Based on the platform's structural parameters and the predicted spatial positions of the hinge points at both ends of each actuator, the minimum spatial distance between any two actuators at each prediction time is calculated. Based on a preset safety distance threshold, the minimum spatial distance is nonlinearly mapped to the interference risk value at each prediction time. The maximum value of all interference risk values ​​generated within the prediction time domain is extracted as the peak risk parameter, and the interference risk values ​​at each time point within the prediction time domain are summed in the time domain to obtain the cumulative risk parameter. The peak risk parameter and the cumulative risk parameter are linearly weighted to obtain the comprehensive interference risk of the corresponding actuator pair. The maximum value of the comprehensive interference risk among all actuator pairs is extracted and established as the final predicted interference risk index.

[0034] Furthermore, in a specific embodiment, to improve data stability and noise resistance, Kalman filtering or other state estimation algorithms are used to filter the original measurement data: ; in, This indicates the platform state after filtering or state estimation. This indicates the filter gain for the current control cycle; Indicates the sensor measurement value; This represents the observation matrix.

[0035] Furthermore, in a specific embodiment, a sliding time window is introduced to construct a historical state sequence: ; And construct a sequence of historical motion prompt instructions: ; in, Indicates the current control cycle The following is from the most recent A sequence of historical states consisting of the platform states at each historical moment; Indicates the length of the sliding time window; Indicates the current control cycle The following is from the most recent A sequence of historical instructions consisting of motion prompts for each historical moment.

[0036] A prediction time domain is established to predict the platform's motion state, actuator length changes, and spatial relationships between actuators over multiple future control cycles, thereby enabling a forward-looking assessment of interference risks. In a preferred embodiment, the prediction time domain is established as follows: Establish a length of Prediction window: ; in, Indicates the current control cycle The predicted future platform state sequence; Indicates the first The platform status at each predicted moment; This indicates the forecast window length, which is the number of control periods that are predicted into the future. The prediction time domain length is: , Indicates the total length of the prediction time domain; This indicates the sampling period or control period of the control system.

[0037] In one alternative implementation, a Long Short-Term Memory (LSTM) network is used to predict the platform state at any future time within the prediction time domain: ; in, This represents a time-series prediction model based on long short-term memory networks; Indicates the predicted future number. The platform state at each moment. The prediction results of the temporal prediction model based on the Long Short-Term Memory Network further enable the coordinated reconstruction between interference risk, safety space, and motion perception fidelity control.

[0038] More preferably, an indicator for predicting intervention risk is generated, specifically including: 1) For any time in the prediction time domain The predicted length of each actuator is calculated based on the platform's predicted pose and inverse kinematics relationship: ; in, Indicates the predicted time The length vectors of the next six actuators; The inverse kinematic mapping function represents the motion platform with six degrees of freedom; Indicates the state of the prediction platform The six-DOF pose vector of the prediction platform is extracted from it.

[0039] 2) For any actuator With actuator The actuator pair ( , ), calculate the prediction time The minimum spatial distance between the two ;in, Indicates actuator With actuator At the predicted time The minimum spatial distance; ,and .

[0040] 3) Construct a continuous interference risk function based on the minimum spatial distance between actuators: ; in, Indicates actuator With actuator At the predicted time Interference risk value; Indicates actuator With actuator At the predicted time The minimum spatial distance; Indicates the preset safe distance threshold; This represents the slope adjustment coefficient of the risk function, and This is used to control the sensitivity of the interference risk value to changes in the minimum spatial distance between actuators; This represents the natural exponential function.

[0041] The continuous interference risk function is used to nonlinearly map the minimum spatial distance between actuators to a value range of [value range missing]. The interference risk value is determined as follows: as the minimum spatial distance between actuators increases, the interference risk value gradually decreases; as the minimum spatial distance between actuators decreases, the interference risk value gradually increases; when the minimum spatial distance between actuators equals a preset safety distance threshold, the interference risk value is 0.5.

[0042] 4) To simultaneously reflect peak risk and persistent risk, a comprehensive time-domain risk indicator is constructed: ; in, Indicates actuator With actuator Comprehensive interference risk in the prediction time domain; This represents the maximum risk value within the forecast window, used to reflect peak risk; This represents the weighted cumulative risk within the forecast window, used to reflect the degree of risk persistence; and These are the weighting coefficients for peak risk and cumulative risk, respectively. Indicates the first Temporal weights for each prediction time point. , ,and ; ,and: ; By weighted and fused the peak interference risk and cumulative interference risk in the prediction time domain, the comprehensive interference risk can simultaneously characterize the instantaneous high-risk state and the persistent risk state in the prediction time domain.

[0043] 5) Take the maximum value among the combined risks of all actuator pairs as the overall predicted interference risk index for the current platform: ; in, This indicates the platform's overall interference risk within the prediction time domain; This represents the maximum value of the combined risk across all actuator pairs, representing the risk of the most dangerous actuator combination in the current prediction time domain.

[0044] S3. Apply positive and negative virtual numerical perturbations to each degree of freedom direction of the platform in the candidate control command space to calculate the risk contribution of each degree of freedom direction to the predicted interference risk index, and normalize the risk contribution of each degree of freedom direction to obtain the degree of freedom risk feature set.

[0045] Preferably, positive and negative virtual numerical perturbations are applied to each degree of freedom direction of the platform within the candidate control command space to calculate the risk contribution of each degree of freedom direction to the predicted interference risk index. The risk contribution of each degree of freedom direction is then normalized to obtain a degree-of-freedom risk feature set, specifically including: In the three translational degrees of freedom and the three attitude degrees of freedom of the six-degree-of-freedom motion platform, respectively, a preset micro-scale positive and negative virtual numerical perturbation is superimposed on the original motion prompt command. Calculate the predicted interference risk index corresponding to the superposition of the positive and negative virtual numerical perturbations respectively, and take the ratio of the risk difference between the two to twice the perturbation amplitude as the risk contribution degree in the corresponding degree of freedom direction; The risk contribution in each degree of freedom direction is normalized to obtain the normalized risk contribution in each degree of freedom direction. The normalized risk contributions in each degree of freedom direction are then combined to form a degree of freedom risk feature set.

[0046] More preferably, to identify "risk sources," sensitivity analysis is used to determine the contribution of each degree of freedom to the interference risk, specifically including: 1) Define the first The unit perturbation vector in each of the degrees of freedom is .in, , represents any direction of any of the six degrees of freedom; Indicates the first A unit perturbation vector in each degree of freedom direction.

[0047] 2) Apply a small perturbation to the original motion cue commands. And recalculate the corresponding predicted intervention risk, then the first Risk contribution in each degree of freedom direction: ; in, Indicates the number of times within the current control cycle. Risk contribution in each degree of freedom direction; This indicates the initial motion prompt command within the current control cycle; Indicates the first The virtual numerical perturbation amplitude in each degree of freedom direction; Indicates the first The unit direction vector corresponding to each degree of freedom direction; This indicates the predicted interference risk for the corresponding candidate control command. The positive and negative virtual numerical perturbations only affect the candidate control command calculation process in the prediction model and are not directly output to the six-degree-of-freedom motion platform.

[0048] 3) Combine the risk contributions of each degree of freedom into a degree-of-freedom risk contribution vector: ; in, This represents the six-degree-of-freedom risk contribution vector; , , These represent the risk contributions in the three translational degrees of freedom directions, respectively. , , These represent the risk contributions in the roll, pitch, and yaw attitude degrees of freedom, respectively.

[0049] If a certain degree of freedom direction A larger value indicates that motion in that degree of freedom direction is more likely to cause the actuator to approach the interference region, and compression, biasing, or constraint adjustment should be prioritized in subsequent control reconfiguration; if the value in a certain degree of freedom direction is large, it indicates that motion in that direction is more likely to cause the actuator to approach the interference region, and compression, biasing, or constraint adjustment should be prioritized. If the value is small, it indicates that the motion in that direction has a weaker impact on the risk of interference, and its motion cues can be retained preferentially during the control reconstruction process.

[0050] More preferably, to eliminate the differences in numerical range and dimensions of risk contributions in different degrees of freedom directions, the risk contributions in each degree of freedom direction are normalized: ; in, Indicates the number of times within the current control cycle. Normalized risk contribution in each degree of freedom direction; Indicates the number of times within the current control cycle. Original risk contribution in each degree of freedom direction; Indicates the first Normalized adjustment parameters in each degree of freedom direction, and This is used to adjust the sensitivity of the normalization function to changes in risk contribution.

[0051] The absolute value of the normalized risk contribution is used to characterize the degree of influence of the corresponding degree of freedom direction on the predicted interference risk, and the sign of the normalized risk contribution is used to characterize the direction of change of the predicted interference risk when the control command in the corresponding degree of freedom direction increases.

[0052] The normalized risk contributions of each degree of freedom direction are combined to form a normalized degree-of-freedom risk contribution vector: ; in, This represents the risk contribution vector of the normalized degrees of freedom within the current control period.

[0053] S4. Dynamically correct the initial motion perception weights according to the set of risk features of the degrees of freedom to obtain a dynamic motion perception weight matrix, and construct a dynamic safe workspace based on the dynamic motion perception weight matrix.

[0054] Preferably, the initial motion perception weights are dynamically corrected based on the set of risk features of the degrees of freedom to obtain a dynamic motion perception weight matrix, and a dynamic safe workspace is constructed based on the dynamic motion perception weight matrix, specifically including: By introducing a weight adjustment coefficient and combining the normalized risk contribution of each degree of freedom direction in the set of degree of freedom risk features, the initial motion perception weight of the corresponding degree of freedom direction is dynamically corrected to generate dynamic motion perception weight of each degree of freedom direction, and the dynamic motion perception weight of each degree of freedom direction is combined to form a dynamic motion perception weight matrix. Within the globally feasible control instruction space determined by the platform actuator extreme values, regions that do not meet the safety reconfiguration requirements are eliminated using a preset interference risk threshold to define a safe and feasible domain. Regions with excessive motion perception distortion are eliminated using a preset motion perception distortion threshold combined with the dynamic motion perception weight to define a motion-fidelity domain. The intersection of the safe and feasible domain and the motion-fidelity domain constitutes the dynamic safe workspace.

[0055] More preferably, the dynamic safety workspace is represented as a subset of control instructions that simultaneously satisfies the following mathematical set constraints: The control commands to be reconfigured are located within the globally feasible control command space defined by the size of the mechanism and the mechanical limits of the hardware; The predicted interference risk index caused by the control command to be reconstructed is not greater than the preset interference risk threshold, and the motion perception distortion degree of the control command to be reconstructed, which is weighted by the dynamic motion perception weight matrix, is not greater than the preset perception distortion threshold.

[0056] In a specific embodiment, the initial motion perception weights are dynamically corrected based on the freedom risk feature set to obtain a dynamic motion perception weight matrix, and a dynamic safe workspace is constructed based on the dynamic motion perception weight matrix, specifically as follows: 1) Define the platform's globally feasible control instruction space as follows: ; in, This refers to the globally accessible workspace, which is determined by the platform mechanism size, actuator length limitations, speed limitations, acceleration limitations, and mechanical limits. 2) Dynamically adjust the perception weights based on the risk contribution of the degrees of freedom: ; in, This indicates the [number]th [period] within the current control cycle after considering the risk of interference. One degree of freedom dynamic motion sensing weight; Indicates the first The initial motion perception weights corresponding to each degree of freedom; Indicates the first Normalized risk contribution in each degree of freedom direction; This represents the weighting adjustment coefficient, and .

[0057] When the influence of a certain degree of freedom direction on the predicted interference risk increases, the corresponding dynamic motion perception weight decreases accordingly, so that the control command reconstruction process can appropriately increase the adjustment range of the motion component in that degree of freedom direction while ensuring the safety of actuator interference. When the influence of a certain degree of freedom direction on the predicted interference risk is small, the corresponding dynamic motion perception weight is close to the initial motion perception weight, so as to preferentially retain the original motion cue features in that degree of freedom direction.

[0058] Furthermore, the dynamic motion sensing weights in each degree of freedom direction are combined to form the dynamic motion sensing weight matrix for the current control cycle: ; in, This represents the dynamic motion sensing weight matrix within the current control cycle; , and These represent the dynamic motion perception weights in the three translational degrees of freedom directions, respectively. , and These represent the dynamic motion perception weights in the directions of roll, pitch, and yaw, respectively.

[0059] 3) Construct a dynamic safety workspace coupled with risk and motion perception: ; in, This represents the dynamic safety workspace that simultaneously satisfies actuator interference safety constraints and motion perception fidelity constraints within the current control cycle. This represents the global feasible workspace determined by the platform kinematic constraints, actuator stroke constraints, actuator velocity constraints, and actuator acceleration constraints within the current control cycle. This indicates the six-degree-of-freedom control commands to be reconfigured; Indicates the control command to be reconfigured Corresponding predicted interference risk; This represents the preset threshold for interference risk; Indicates the control command to be reconfigured Motion perception distortion relative to the original motion cues; This represents the preset motion perception distortion threshold.

[0060] The dynamic safety workspace is essentially a dynamic feasible region within the six-degree-of-freedom control instruction space, used to limit the set of control instructions that satisfy platform physical constraints, actuator interference safety constraints, and motion perception fidelity constraints within the current control cycle.

[0061] Furthermore, the motion perception distortion within the current control cycle is defined as: ; in, Indicates the control commands to be reconfigured within the current control cycle. Motion perception distortion relative to the original motion cues; This represents the dynamic motion sensing weight matrix obtained after dynamic correction based on the normalized risk contribution in each degree of freedom direction; This indicates the six-degree-of-freedom control commands to be reconfigured; This indicates the initial motion prompt command within the current control cycle; The L2 norm of a vector.

[0062] The motion-sensing distortion is evaluated differently for control reconstruction deviations in different degrees of freedom by using dynamic motion-sensing weights. For degrees of freedom with higher dynamic motion-sensing weights, their control reconstruction deviations will produce larger evaluation results in the motion-sensing distortion, thus limiting over-adjustment in that degree of freedom. For degrees of freedom with lower dynamic motion-sensing weights, larger safety corrections are allowed, provided that actuator interference safety constraints are met.

[0063] S5. Under the constraints of the dynamic safety workspace, the original motion prompting command is adaptively reconstructed based on the predicted interference risk index and the dynamic motion perception weight matrix to generate the target safety control command.

[0064] Preferably, generating target security control instructions specifically includes: A multi-objective optimization function is constructed, which is formed by weighted summation of evaluation cost sub-items from multiple dimensions. The evaluation cost sub-items include at least: a tracking error term used to constrain the approximation degree of the reconstruction command to the original motion prompt command; a prediction interference risk term used to suppress the hazards of mechanical collisions; a motion perception distortion term used to weight the control reconstruction deviation in different degrees of freedom directions based on the dynamic motion perception weight matrix; and a control smoothing term measured by the change in the amount of change between two adjacent control cycles. Under the conditions of the dynamic safety workspace and the physical constraint boundary of the actuator, the reconstructed target safety control command is obtained by solving the control variables that minimize the multi-objective optimization function in real time.

[0065] More preferably, the actuator physical constraint boundary includes at least: The working lengths of each actuator corresponding to the reconstructed target safety control command calculated in real time are within the range defined by the minimum and maximum lengths; The instantaneous extension speed and absolute value of extension acceleration of each actuator corresponding to the reconstructed target safety control command calculated in real time are all limited to the maximum allowable speed and maximum allowable acceleration boundary range allowed by the hardware.

[0066] In one specific embodiment, a multi-objective optimization method is employed in a dynamic safe workspace. Under the constraints, we find the minimum value of a comprehensive cost function J in order to find the optimal target security control command. The optimization problem is formulated as follows: ; in, This represents the objective function for controlling the refactoring and optimization. This indicates a tracking error in the original motion. This indicates the predicted risk of interference. This indicates the motion perception distortion term; Indicates the control smoothing term; , , , These are the weighting coefficients for each of the above items, used to adjust the relative importance among tracking performance, security, motion perception fidelity, and control smoothness; each sub-item specifically includes: Original motion indicates tracking error: , This indicates the tracking error of the reconstructed control command relative to the original motion cue command; predicts interference risk terms: ,, Indicates control commands The corresponding predicted intervention risk item; Motion perception distortion term: , This indicates the degree of motion perception distortion caused by control reconstruction; Control smoothing term: , This indicates the amount of control change between adjacent control cycles; This indicates the six-degree-of-freedom control command of the previous control cycle. This item is used to suppress control abrupt changes and improve the continuity and smoothness of platform motion.

[0067] During the optimization process, the physical constraints of the actuator are satisfied: ; ; ; in, and They represent the first Minimum and maximum length of the root actuator; Indicates the first The maximum scaling speed allowed by the root actuator; Indicates the first The scaling acceleration of the root actuator in the current control cycle; Indicates the first The maximum allowable scaling acceleration for the root actuator.

[0068] The optimal safety control command is obtained by employing real-time optimization algorithms such as Model Predictive Control (MPC) or Quadratic Programming (QP). ; in, This represents the optimal safety control command obtained within the current control cycle; This represents the dynamic safety workspace constructed within the current control cycle based on predicted interference risks and dynamic motion perception weights. This represents the multi-objective optimization function within the current control cycle. This indicates that the objective function The control variable that reaches its minimum value.

[0069] S6. Convert the target safety control command into an actuator length command and output it to drive the six-degree-of-freedom motion platform to perform closed-loop actions.

[0070] Preferably, before outputting the actuator length instruction: The actuator length command calculated by conversion is imported into a first-order hysteresis filter. Using the historical output actuator length command and smoothing filter coefficients of the previous cycle, the current actuator length command is subjected to weighted smoothing filtering to avoid abrupt operations. The filtered final state value is output to the underlying mechanical actuator to update the closed-loop motion state and start the next predictive control cycle.

[0071] Further preferably, the optimal safety control command is obtained. Then, it is converted into an actuator target length instruction vector using the platform's inverse kinematics model: ; in, This represents the actuator length instruction vector generated in the current control cycle; The inverse kinematic mapping function represents the motion platform with six degrees of freedom; This indicates the reconstructed security control instructions; The executor length instruction vector is: ; in, This indicates that the current control cycle output is given to the first... The target length instruction for the root executor.

[0072] More preferably, to avoid abrupt changes in instructions, a first-order hysteresis filter can be applied to the length instructions: ; in, This indicates the smooth length command that will be ultimately output to the actuator control system during the current control cycle. This indicates the actuator length instruction output in the previous control cycle; This represents the filter coefficients, and their values ​​range from [value range missing]. Finally, the smoothed instructions... The output is sent to the actuator servo drive system. After the actuator moves, the system collects the platform feedback status and the actuator feedback status in real time, and enters the next control cycle, forming the following closed-loop process: ; in, This indicates the overall intervention risk predicted during the current control cycle; This represents the risk-motion perception coupled dynamic safety workspace constructed in the current control cycle; This indicates the platform feedback status for the next control cycle after the actuator has executed its function.

[0073] Through this closed-loop mechanism, control commands can be dynamically adjusted continuously based on the actual operating status of the platform, predicted interference risks, and motion perception fidelity requirements, thereby achieving safe, smooth, and high-fidelity motion control of the six-degree-of-freedom motion platform under complex motion conditions.

[0074] The second embodiment of this application provides a dynamic space reconstruction method for a six-degree-of-freedom motion platform. Based on the first embodiment, when computing resources are limited or real-time requirements are extremely high, an analytical calculation method based on affine transformation is used to directly generate safety control commands, replacing complex nonlinear optimization solutions.

[0075] Preferably, generating target security control instructions includes: Based on the normalized degree of freedom risk contribution vector and the dynamic motion sensing weight matrix Calculate and generate a safe scaling matrix with diagonal features for the degrees of freedom. Based on the normalized degree-of-freedom risk contribution vector, a safety bias vector is generated to guide the platform's motion pose to a low-risk motion region within the dynamic safe workspace boundaries. ; Determine the safety bias vector based on the risk contribution vector of the normalized degrees of freedom: ; Among them, the The safety bias components in each degree of freedom direction are: ; in, Indicates the number of times within the current control cycle. Safety bias components in each degree of freedom direction; Indicates the first The maximum allowable safety offset adjustment coefficient for each degree of freedom direction; Indicates the first Normalized risk contribution in each degree of freedom direction.

[0076] Based on the degree-of-freedom safety scaling matrix and safety bias vector, the original motion cue command is subjected to a linear affine transformation to a low-risk motion region, generating the target safety control command: ; in, This represents the six-degree-of-freedom safety control command obtained after safety reconfiguration within the current control cycle; Indicates the reference safe pose command within the current control cycle; This represents the degree-of-freedom safety scaling matrix, which is jointly determined by the normalized degree-of-freedom risk contribution vector and the dynamic motion-sensing weight matrix. This indicates the initial motion prompt command within the current control cycle; This represents the safety bias vector determined based on the risk contribution vector of the normalized degrees of freedom.

[0077] More preferably, a secure scaling matrix It is a diagonal matrix, whose diagonal elements It is calculated in reverse from the risk contribution level, for example: Based on the normalized degree of freedom risk contribution vector and the dynamic motion sensing weight matrix Calculate and generate a safe scaling matrix with diagonal features for the degrees of freedom. : ; in, , and These represent the safety scaling factors for the three translational degrees of freedom within the current control cycle; , and These represent the safety scaling factors for the roll, pitch, and yaw degrees of freedom in the current control cycle, respectively.

[0078] Furthermore, regarding the first The dynamic motion sensing weights in each degree of freedom direction are normalized: ; in, Indicates the first Normalized dynamic motion perception weights in each degree of freedom direction; This represents a preset positive number, used to avoid the denominator being zero.

[0079] The first is determined based on the normalized dynamic motion sensing weights. Dynamic risk compression coefficient in each degree of freedom direction: ; in, Indicates the number of times within the current control cycle. Dynamic risk compression coefficient in each degree of freedom direction; Indicates the first The basic risk compression coefficient for each degree of freedom direction; This represents the motion perception weight adjustment coefficient, and .

[0080] No. The safety scaling factor for each degree of freedom direction is: ; in, Indicates the number of times within the current control cycle. Safety scaling factor for each degree of freedom direction; Indicates the first The minimum safe scaling ratio allowed in each degree of freedom direction; Indicates the first Normalized risk contribution amplitude in each degree of freedom direction.

[0081] When the normalized risk contribution in a certain degree of freedom direction is large and the dynamic motion perception weight is low, the corresponding safety scaling factor is reduced accordingly to prioritize the compression of high-risk motion components in that degree of freedom direction; when the dynamic motion perception weight in a certain degree of freedom direction is high, the corresponding dynamic risk compression factor is reduced, thereby prioritizing the retention of motion indication features in that degree of freedom direction while meeting the actuator interference safety requirements.

[0082] In one specific embodiment ,in This indicates the minimum scaling ratio allowed for that degree of freedom.

[0083] In this embodiment, the complex optimization problem is simplified to a single matrix multiplication and addition operation, which improves computational efficiency. Although it is less refined in terms of fidelity than the optimization method, it can still prioritize the compression of high-risk, low-perception motion components while ensuring safety, thus achieving an adaptive balance between safety and fidelity.

[0084] A third embodiment of this application proposes a dynamic spatial reconstruction system for a six-degree-of-freedom motion platform, comprising: The status acquisition and instruction parsing module is configured to acquire the running status information and original motion prompt instructions of the six-degree-of-freedom motion platform, and to acquire the initial motion perception weights that reflect the importance of different degrees of freedom to the motion sensation. The interference risk prediction and assessment module is configured to predict the potential interference state of the platform in the prediction time domain based on the operating status information and the original motion prompt instructions, and generate a predicted interference risk index. The interference sensitivity analysis module is configured to apply positive and negative virtual numerical perturbations to each degree of freedom direction in the candidate control command space, calculate the risk contribution of each degree of freedom direction to the predicted interference risk index, and normalize the risk contribution to obtain the degree of freedom risk feature set. The workspace construction module is configured to dynamically correct the initial motion perception weights according to the set of risk features of the degrees of freedom, obtain a dynamic motion perception weight matrix, and construct a dynamic safe workspace based on the dynamic motion perception weight matrix. The adaptive reconstruction module is configured to perform adaptive control reconstruction of the original motion prompting command based on the predicted interference risk index and the dynamic motion perception weight matrix under the constraints of the dynamic safety workspace, and generate a target safety control command. The instruction output module is configured to convert the target safety control instruction into an actuator length instruction and output it to drive the six-degree-of-freedom motion platform to perform actions.

[0085] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process and related descriptions of the system described above can be found in the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0086] It should be noted that the six-degree-of-freedom motion platform dynamic space reconstruction system provided in the above embodiments is only an example of the division of the above functional units. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the modules or steps in the embodiments of this application can be further decomposed or combined. For example, the modules in the above embodiments can be merged into one module, or further divided into multiple sub-modules to complete all or part of the functions described above. The names of the modules and steps involved in the embodiments of this application are only for distinguishing the various modules or steps and are not considered as an improper limitation of this application.

[0087] A device according to the fourth embodiment of this application includes: At least one processor; and a memory communicatively connected to at least one of the processors; The memory stores instructions that can be executed by the processor to implement the above-described dynamic space reconstruction method for a six-degree-of-freedom motion platform.

[0088] A computer-readable storage medium according to a fifth embodiment of this application stores computer instructions that are executed by a computer to implement the above-described dynamic space reconstruction method for a six-degree-of-freedom motion platform.

[0089] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process and related descriptions of the storage device and processing device described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0090] The following is for reference. Figure 2 It shows a schematic diagram of the structure of a computer system for implementing the methods, apparatus, and electronic devices of this application. Figure 2 The server shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0091] like Figure 2 As shown, the computer system includes a Central Processing Unit (CPU) 201, which can perform various appropriate actions and processes based on programs stored in Read Only Memory (ROM) 202 or programs loaded from storage into Random Access Memory (RAM) 203. The RAM 203 also stores various programs and data required for system operation. The CPU 201, ROM 202, and RAM 203 are interconnected via a bus 204. An Input / Output (I / O) interface 205 is also connected to the bus 204.

[0092] The following components are connected to I / O interface 205: an input section 206 including a keyboard, mouse, etc.; an output section 207 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 208 including a hard disk, etc.; and a communication section 209 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section 209 performs communication processing via a network such as the Internet. A drive 210 is also connected to I / O interface 205 as needed. Removable media 211, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 210 as needed so that computer programs read from them can be installed into storage section 208 as needed.

[0093] Specifically, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 209, and / or installed from removable medium 211. When the computer program is executed by central processing unit (CPU) 201, it performs the functions defined in the methods of this application. It should be noted that the computer-readable medium described above in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on a computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0094] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0095] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0096] The terms “first”, “second”, etc., are used to distinguish similar objects, not to describe or indicate a specific order or sequence.

[0097] The term "comprising" or any other similar term is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus / device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent in such process, method, article, or apparatus / device.

[0098] The technical solutions of this application have been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of this application is obviously not limited to these specific embodiments. Without departing from the principles of this application, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of this application.

Claims

1. A method for dynamic space reconstruction of a six-degree-of-freedom motion platform, characterized in that, Includes the following steps: Acquire the operating status information and original motion prompts of the six-degree-of-freedom motion platform, and obtain the initial motion perception weights to reflect the importance of different degrees of freedom to the motion perception. Based on the operational status information and the original motion prompts, the potential interference state of the platform in the prediction time domain is predicted, and a predicted interference risk index is generated. Positive and negative virtual numerical perturbations are applied to each degree of freedom direction of the platform in the candidate control command space to calculate the risk contribution of each degree of freedom direction to the predicted interference risk index, and the risk contribution of each degree of freedom direction is normalized to obtain the degree of freedom risk feature set. Based on the normalized risk contribution of each degree of freedom direction in the set of degree of freedom risk features, the initial motion perception weights are dynamically corrected to obtain a dynamic motion perception weight matrix, and a dynamic safe workspace is constructed based on the dynamic motion perception weight matrix. Under the constraints of the dynamic safety workspace, the original motion prompting command is adaptively reconstructed based on the predicted interference risk index and the dynamic motion perception weight matrix to generate the target safety control command. The target safety control command is converted into an actuator length command and output to drive the six-degree-of-freedom motion platform to perform closed-loop actions.

2. The dynamic space reconstruction method for a six-degree-of-freedom motion platform according to claim 1, characterized in that, Generate indicators for predicting intervention risk, including: A sliding time window is used to extract a historical state sequence constructed based on the running state information and a historical instruction sequence constructed based on the original motion prompt instruction, and then synthesizes the input time sequence. The input time series is input into a preset time series prediction model to calculate the platform prediction pose at each prediction time in the prediction time domain; Based on the platform's predicted pose at each prediction time, the predicted length of each actuator at each prediction time is calculated through the platform's inverse kinematics mapping. Based on the platform's structural parameters and the predicted spatial positions of the hinge points at both ends of each actuator, the minimum spatial distance between any two actuators at each prediction time is calculated. Based on a preset safety distance threshold, the minimum spatial distance is nonlinearly mapped to the interference risk value at each prediction time. The maximum value of all interference risk values ​​generated within the prediction time domain is extracted as the peak risk parameter, and the interference risk values ​​at each time point within the prediction time domain are summed in the time domain to obtain the cumulative risk parameter. The peak risk parameter and the cumulative risk parameter are linearly weighted to obtain the comprehensive interference risk of the corresponding actuator pair. The maximum value of the comprehensive interference risk among all actuator pairs is extracted and established as the final predicted interference risk index.

3. The dynamic space reconstruction method for a six-degree-of-freedom motion platform according to claim 1, characterized in that, Virtual numerical perturbations are applied to each degree of freedom of the platform in the candidate control command space to calculate the degree of influence of each degree of freedom direction on the predicted interference risk index, thereby obtaining a degree of freedom risk feature set, specifically including: In the three translational degrees of freedom and the three attitude degrees of freedom of the six-degree-of-freedom motion platform, respectively, a preset micro-scale positive and negative virtual numerical perturbation is superimposed on the original motion prompt command. Calculate the predicted interference risk index corresponding to the superposition of the positive and negative virtual numerical perturbations respectively, and take the ratio of the risk difference between the two to twice the perturbation amplitude as the risk contribution degree in the corresponding degree of freedom direction; The risk contribution in each degree of freedom direction is normalized to obtain the normalized risk contribution in each degree of freedom direction. The normalized risk contributions in each degree of freedom direction are then combined to form the degree of freedom risk feature set.

4. The dynamic space reconstruction method for a six-degree-of-freedom motion platform according to claim 3, characterized in that, The initial motion perception weights are dynamically corrected based on the aforementioned degree-of-freedom risk feature set to obtain a dynamic motion perception weight matrix. A dynamic safe workspace is then constructed based on this dynamic motion perception weight matrix, specifically including: By introducing a weight adjustment coefficient and combining the normalized risk contribution of each degree of freedom direction in the set of degree of freedom risk features, the initial motion perception weight of the corresponding degree of freedom direction is dynamically corrected to generate dynamic motion perception weight of each degree of freedom direction, and the dynamic motion perception weight of each degree of freedom direction is combined to form a dynamic motion perception weight matrix. Within the globally feasible control instruction space determined by the platform actuator extreme values, regions that do not meet the safety reconfiguration requirements are eliminated using a preset interference risk threshold to define a safe and feasible domain. Regions with excessive motion perception distortion are eliminated using a preset motion perception distortion threshold combined with the dynamic motion perception weight matrix to define a motion perception fidelity domain. The intersection of the safe and feasible domain and the motion perception fidelity domain constitutes the dynamic safe workspace.

5. The dynamic space reconstruction method for a six-degree-of-freedom motion platform according to claim 4, characterized in that, The dynamic safe workspace is represented by a subset of control instructions that simultaneously satisfy the following mathematical set constraints: The control commands to be reconfigured are located within the globally feasible control command space, which is defined by the size of the mechanism and the mechanical limits of the hardware. The predicted interference risk index caused by the control command to be reconstructed is not greater than the preset interference risk threshold, and the motion perception distortion degree of the control command to be reconstructed, which is weighted by the dynamic motion perception weight matrix, is not greater than the preset perception distortion threshold.

6. The dynamic space reconstruction method for a six-degree-of-freedom motion platform according to claim 4, characterized in that, Generate target security control instructions, specifically including: A multi-objective optimization function is constructed, which is formed by weighted summation of evaluation cost sub-items from multiple dimensions. The evaluation cost sub-items include at least: a tracking error term used to constrain the approximation degree of the reconstruction command to the original motion prompt command; a prediction interference risk term used to suppress the hazards of mechanical collisions; a motion perception distortion term used to weight the control reconstruction deviation in different degrees of freedom directions based on the dynamic motion perception weight matrix; and a control smoothing term measured by the change in the amount of change between two adjacent control cycles. Under the conditions of the dynamic safety workspace and the physical constraint boundary of the actuator, the reconstructed target safety control command is obtained by solving the control variables that minimize the multi-objective optimization function in real time.

7. The dynamic space reconstruction method for a six-degree-of-freedom motion platform according to claim 6, characterized in that, The physical constraint boundary of the actuator includes at least: The working length of each actuator corresponding to the target safety control command calculated in real time is within the range defined by the minimum allowable length and the maximum allowable length of the corresponding actuator; The absolute values ​​of the instantaneous extension speed and extension acceleration of each actuator corresponding to the target safety control command calculated in real time are not greater than the maximum extension speed and maximum extension acceleration allowed by the hardware, respectively.

8. The dynamic space reconstruction method for a six-degree-of-freedom motion platform according to claim 4, characterized in that, Generate target security control instructions, including: Based on the normalized risk contribution of each degree of freedom in the set of risk features and the dynamic motion perception weight matrix, a safe scaling matrix with diagonal features is calculated and generated, wherein each diagonal element of the safe scaling matrix is ​​determined by the normalized risk contribution and the dynamic motion perception weight in the corresponding degree of freedom direction. Based on the normalized risk contribution of each degree of freedom in the set of risk features, a safety bias vector is generated to guide the platform's motion pose to a low-risk motion region within the dynamic safe workspace. Based on the safety scaling matrix of the degrees of freedom and the safety bias vector, a linear affine transformation is performed on the original motion prompt command so that the transformed control command is located in the low-risk motion region within the dynamic safety workspace, thereby generating the target safety control command.

9. The dynamic space reconstruction method for a six-degree-of-freedom motion platform according to claim 1, characterized in that, Before outputting the executor length instruction: The actuator length command calculated by conversion is imported into a first-order hysteresis filter. Using the historical output actuator length command and smoothing filter coefficients of the previous cycle, the current actuator length command is subjected to weighted smoothing filtering to avoid abrupt operations. The filtered final state value is output to the underlying mechanical actuator to update the closed-loop motion state and start the next predictive control cycle.

10. A dynamic spatial reconfiguration system for a six-degree-of-freedom motion platform, characterized in that, include: The status acquisition and instruction parsing module is configured to acquire the running status information and original motion prompt instructions of the six-degree-of-freedom motion platform, and to acquire the initial motion perception weights that reflect the importance of different degrees of freedom to the motion sensation. The interference risk prediction and assessment module is configured to predict the potential interference state of the platform in the prediction time domain based on the operating status information and the original motion prompt instructions, and generate a predicted interference risk index. The interference sensitivity analysis module is configured to apply positive and negative virtual numerical perturbations to each degree of freedom direction in the candidate control command space, calculate the risk contribution of each degree of freedom direction to the predicted interference risk index, and normalize the risk contribution to obtain the degree of freedom risk feature set. The workspace construction module is configured to dynamically correct the initial motion perception weights based on the normalized risk contribution of each degree of freedom direction in the set of degree of freedom risk features, to obtain a dynamic motion perception weight matrix, and to construct a dynamic safe workspace based on the dynamic motion perception weight matrix. The adaptive reconstruction module is configured to perform adaptive control reconstruction of the original motion prompting command based on the predicted interference risk index and the dynamic motion perception weight matrix under the constraints of the dynamic safety workspace, and generate a target safety control command. The instruction output module is configured to convert the target safety control instruction into an actuator length instruction and output it to drive the six-degree-of-freedom motion platform to perform actions.

Citation Information

Patent Citations

  • Multi-degree-of-freedom parallel motion platform motion trail collision risk early warning method and system for simulator

    CN119623005A

  • Full-dimensional self-interference detection method for parallel robot platform

    CN121893333A