23 results about "Reference tracking" patented technology

The invention belongs to the technical field of multi-agent systems, and particularly relates to a double-layer heterogeneous multi-agent system sliding mode formation control method, which comprises the steps of S1, constructing a double-layer heterogeneous multi-agent system according to a preset form; s2, setting a cross-level hybrid communication topological structure for the double-layer heterogeneous multi-agent system; s3, setting a system control rate of the double-layer heterogeneous multi-agent system; and S4, based on the constructed double-layer heterogeneous multi-agent system, performing formation control and obstacle avoidance cooperative operation of the multi-agent system by using the system control rate in the S3, and ensuring that the system realizes cooperative consistency of reference tracking errors and real-time prevention and control of collision risks under the condition that mismatching uncertainty exists. According to the method, the coordination consistency of tracking errors and real-time prevention and control of collision risks can be ensured while the uncertainties of mismatching are processed.

A machine learning model is trained with a training dataset, where the machine learning model comprises a plurality of layers. During training, values of a plurality of coefficients of one or more layers are monitored. In response to detecting a change of a given coefficient by more than a threshold during a given training run, a given reference to a given input dataset of the given training run is stored. In response to detecting an output error of a trained version of the machine learning model, the given reference to the given input dataset is retrieved if the given coefficient is located on a backward path providing more than a threshold contribution to the output error. Next, the given reference is provided to an application analyzing the trained version of the machine learning model in order to determine a cause of the output error.

The invention discloses a PID (Proportion Integration Differentiation) tracking control method based on error perception nonlinear gain, which is used for high-precision tracking control of an MIMO (Multiple Input Multiple Output) nonlinear system. Comprising the following steps: 1) constructing an adaptive gain by using current error information, historical error information and prediction error information; 2) realizing intrinsic integral saturation resistance through nonlinear integral design; and 3) complete autonomous operation can be realized without accurate system dynamics priori knowledge or manual adjustment. According to the method, the nonlinear compensation capability and the robust adaptive performance are enhanced while the structural simplicity of the PID controller is kept. Through rigorous Lyapunov analysis, the stability of the method to a non-linear system in a general form is verified, and the method shows excellent performance in both adjustment and time-varying reference tracking scenes. The method can be widely applied to the fields of linear motor servo systems, sea surface ship control, unmanned vehicle dynamics control and the like needing high-precision tracking control.

This invention discloses a robust control method for robot visual servoing based on the theory of all-drive systems, belonging to the field of visual servoing for robotic arms. The steps are as follows: Taking the eye-to-hand robot visual servoing system as the research object, based on the efficient second-order minimization (EMS) method and combined with the virtual visual servoing method, the target depth and attitude parameters are estimated online to construct a high-precision image interaction matrix; utilizing the all-drive characteristics of robot joint dynamics, the image feature error is mapped to joint space error through the interaction matrix, establishing an open-loop error dynamic model of the visual servoing all-drive system; based on the theory of all-drive systems, it is transformed into a linear second-order system through nonlinear state feedback, and the state space equation is constructed; a sliding mode robust model reference tracking controller is designed. This method can maintain system stability under conditions of camera calibration error, dynamic parameter disturbance, and unmodeled dynamics coexisting, achieving high-precision, high-dynamic visual servoing tracking control.

The invention discloses an engineering vehicle walking torque distribution and driving anti-skid control method and system, and belongs to the technical field of engineering vehicle control. The method comprises the steps that pedal, gear and wheel speed signals of a vehicle are collected, segmented nonlinear reanalysis and walking mode identification are carried out, preliminary distribution and constraint of four-wheel driving or braking energy recovery torque are carried out, and a four-wheel initial torque request is generated; anti-skid reference tracking wheel speed is generated based on the vehicle state, the actual wheel speed of each wheel is compared with the reference wheel speed, the rotation state of each wheel is independently identified in a distributed mode in combination with the opening degree of an accelerator, and an identifier is output; and when there is wheel slip, calculating a torque takeover value based on a robust integral sliding mode control algorithm by taking the reference wheel speed as a target, and taking a small value of the torque takeover value and an initial torque request of a corresponding wheel as final output. The problem that traditional anti-skid control fails near zero speed is solved, and full-working-condition and high-precision self-adaptive driving anti-skid control is achieved.

The invention relates to an auxiliary positioning test method and device and a storage medium, and the method comprises the steps: determining a test position of to-be-tested electronic equipment in a test environment, and obtaining reference GPS position information corresponding to the test position, the reference GPS position information being predetermined through reference electronic equipment, the to-be-tested electronic equipment and the reference electronic equipment have the same specification. The to-be-tested electronic equipment is controlled to be at the test position, signal scanning is carried out within the reference test power range based on the GPS position information, the tracking sensitivity of the electronic equipment is obtained, and the reference test power range is determined based on the reference tracking sensitivity and the carrier-to-noise ratio of the to-be-tested electronic equipment at the test position; the reference tracking sensitivity is pre-determined by the reference electronic device at the test position and based on the reference GPS position information, the corresponding test signal can be searched more quickly, and a scanning range with a smaller range is set based on the reference tracking sensitivity, so that the test efficiency is effectively improved.

The present application relates to the technical field of pressure control, and discloses a butterfly valve pressure control method, device, equipment and storage medium, which comprises the following steps: performing signal tracking processing on a target pressure set value to obtain a reference tracking signal; performing interference suppression processing on an actual pressure signal to obtain an interference suppression signal; performing signal fusion on the reference tracking signal and the interference suppression signal to obtain a fusion tracking signal; performing differential extraction on the fusion tracking signal to obtain a differential signal; estimating the system state of a butterfly valve system, and performing comparison calculation based on the fusion tracking signal, the differential signal and the system state to generate a butterfly valve control signal; the present application generates a butterfly valve control signal by fusing the signal tracking and the interference suppression dual-channel signals, and combining the system state and the differential extraction result, realizes full compensation and dynamic correction of the time delay influence, and significantly improves the response speed and control precision of the butterfly valve pressure control.

The invention relates to the technical field of pressure control, and discloses a butterfly valve pressure control method, device and equipment and a storage medium, and the method comprises the steps: carrying out the signal tracking processing of a target pressure set value, so as to obtain a reference tracking signal; performing interference suppression processing on the actual pressure signal to obtain an interference suppression signal; performing signal fusion on the reference tracking signal and the interference suppression signal to obtain a fused tracking signal; performing differential extraction on the fused tracking signal to obtain a differential signal; estimating the system state of the butterfly valve system, and performing comparative calculation based on the fused tracking signal, the differential signal and the system state to generate a butterfly valve control signal; according to the method, the butterfly valve control signal is generated by fusing signal tracking and interference suppression dual-channel signals and combining the system state and the differential extraction result, whole-course compensation and dynamic correction of the time lag influence are achieved, and the response speed and the control precision of butterfly valve pressure control are remarkably improved.

Techniques for concurrent lazy reference tracking in an old garbage collection generation are disclosed, including: encountering, by a mutator thread during a garbage collection epoch, a first instruction to write a first value to a field; responsive to encountering the first instruction to write the first value to the field: entering a slow-path write barrier; performing, by the slow-path write barrier, a first one or more reference counting operations with respect to the field; encountering, by the mutator thread during the first garbage collection epoch and subsequent to encountering the first instruction to write the first value to the field, a second instruction to write a second value to the field; responsive to encountering the second instruction to write the second value to the field: entering a fast-path write barrier; wherein the fast-path write barrier does not perform any reference counting operations with respect to the field.

This invention discloses a target tracking and control method for unmanned surface vessels (USVs) based on sea-air collaborative visual perception. The method includes: establishing a nonlinear mathematical model of a USV-US collaborative system; acquiring left and right images in real time using the USV's binocular cameras to construct a dynamic positioning framework that fuses stereo vision with platform pose; performing target detection and coordinate processing on the left and right images to obtain the fused target position coordinates; obtaining a reference tracking path for the collaborative system based on these target position coordinates through multiple optimization and path generation mechanisms; employing a hierarchical collaborative control strategy based on the reference tracking path and the nonlinear mathematical model to obtain the control laws for the USV and the USV, as well as the attitude reference signals; and finally controlling the USV-US collaborative system to achieve the target tracking task. This invention realizes collaborative target tracking of heterogeneous sea-air platforms under visual perception, offering advantages such as high positioning accuracy, reasonable path planning, and strong control robustness.

A machine learning model is trained with a training dataset, where the machine learning model comprises a plurality of layers. During training, values of a plurality of coefficients of one or more layers are monitored. In response to detecting a change of a given coefficient by more than a threshold during a given training run, a given reference to a given input dataset of the given training run is stored. In response to detecting an output error of a trained version of the machine learning model, the given reference to the given input dataset is retrieved if the given coefficient is located on a backward path providing more than a threshold contribution to the output error. Next, the given reference is provided to an application analyzing the trained version of the machine learning model in order to determine a cause of the output error.

The application relates to the technical field of robot tracking, and discloses a dynamic tag tracking method and device for an RFID robot based on perception limitation. The robot is provided with an RFID antenna for detecting a target with an RFID tag; the dynamic tag tracking method comprises the following steps: acquiring phase data for representing the relative position of the RFID tag and the RFID antenna; performing phase unfolding on the phase data, and performing phase correction on the absolute value of the unfolded phase; estimating a virtual candidate position of the RFID tag according to the position information of the RFID tag at the previous moment; determining the position information of the RFID tag at the current moment based on the matching condition of the unfolded phase of the RFID antenna and the phase of the virtual candidate position, so as to obtain the relative position information of the RFID tag and the robot; constructing a potential energy function based on the relative position information of the RFID tag and the robot and the detection constraint of the RFID antenna, and introducing a cost function to calculate the reference tracking speed of the robot. In this way, the RFID robot with the perception limitation can track the target more accurately.

The invention relates to a method for testing finite-time secrecy attacks of a tracking control system. The method comprises the following steps of: constructing a system model based on a state variable and a system matrix of a target system; determining a controller gain matrix based on the system model, a reference tracking model and a preset positive definite symmetric matrix; constructing a state observer and an attack detector of a target system through a holosymmetric polytope method and the controller gain matrix; based on the state observer, the attack detector and the controller gain matrix, a secret attack condition is determined; according to a preset expectation set and the secrecy attack condition, constructing a plurality of objective functions of multi-stage secrecy attacks within a preset time; and based on the optimal solution of each target function, determining the optimal attack mode of the security test of the target system. According to the method, the secret attack condition is determined through the holosymmetric polytope method and the control model, so that the effectiveness of the secret attack test is improved.

This invention relates to a two-degree-of-freedom active disturbance rejection (ADNR) control method for a dual three-phase motor driven by a nine-switch current source, belonging to the field of permanent magnet motor control technology. The invention addresses the coupling problem between tracking and disturbance rejection performance in traditional ADNR control for dual three-phase permanent magnet synchronous motors based on current source inverters. The invention includes: establishing a mathematical model of the dq-axis current loop of a nine-switch current source inverter in a synchronous rotating coordinate system, incorporating both periodic and aperiodic disturbances; designing an improved extended state observer, removing the error correction term for the current tracking signal and its differential signal update law in the traditional extended state observer, and designing the disturbance update law as a proportional-integral-repetitive control structure; constructing a two-degree-of-freedom ADNR controller based on the control characteristics of the current source inverter, and independently adjusting the reference tracking performance and disturbance rejection performance based on the output of the improved extended state observer, achieving complete decoupling between the two.

The invention discloses a test case generation cooperation system for an automobile BCM, and the system comprises an interface layer which provides files, tasks, reviews and questions and answers, and returns an increment result in an SSE streaming manner; the multi-agent collaboration layer is connected and interacted with the interface layer, comprises a test case generation agent and a test case review agent, and is controlled by polling dialogues in combination with termination conditions; the model management layer is connected with and receives the instruction of the multi-agent collaboration layer, and is used for uniformly adapting to multiple models and performing task-level reference tracking and cleaning; the data persistence layer is connected with the interface layer, receives and stores three entities of files, tasks and review history, and records a state machine and a write-back trace; and the tool is connected with the log layer, sends data to the interface layer and is used for robust JSON extraction, unified log and rotation and structured downloading and exporting. A'test case generation Agent-test case review Agent-manual review 'closed-loop collaborative mechanism can be realized, and a test case set which is strictly structured and can fall into a library is automatically generated for a BCM scene.

The invention relates to the technical field of robot tracking, and discloses a dynamic tag tracking method and device based on a perception-limited RFID robot. The robot is provided with an RFID antenna and is used for detecting a target pasted with an RFID tag; the dynamic tag tracking method comprises the following steps: acquiring phase data used for representing relative positions of an RFID tag and an RFID antenna; performing phase unwrapping on the phase data, and performing phase correction on an absolute value of an unwrapped phase; estimating a virtual candidate position of the RFID tag according to the position information of the RFID tag at the previous moment; based on the matching condition of the unwrapped phase of the RFID antenna and the phase of the virtual candidate position, determining the position information of the RFID tag at the current moment so as to obtain the relative position information of the RFID tag and the robot; and constructing a potential energy function based on the relative position information of the RFID tag and the robot and the detection constraint of the RFID antenna, and introducing a cost function to calculate the reference tracking speed of the robot. In this way, the RFID robot with limited perception can track the target more accurately.

Systems and methods of using a common control scheme to autonomously control a vehicle during semi-autonomous and fully autonomous driving modes are provided. In particular, embodiments of the presently disclosed technology incorporate reference tracking for driving input and vehicle state into this common control scheme. In some embodiments, this common control scheme may be implemented using Model Predictive Control (“MPC”).

The application relates to the technical field of industrial vision detection, and discloses a high-temperature-resistant nozzle coaxiality visual detection method, which comprises the following steps: through double-entry judgment of structural heterogeneity and working condition environment, self-adaptive matching of a detection mode and an actual state of a nozzle is realized, detection resource mismatch is avoided, through an in-situ dynamic reference tracking method of the same material, effective stripping of thermal expansion of a support structure and systematic errors of an environment system is realized, and it is ensured that a preliminary coaxiality measurement result truly reflects thermal deformation of the nozzle itself, through a multi-material gradient reference reference body array and a microstructure coding full-field decoding technology, point-by-point fine compensation of non-uniform thermal deformation of a heterogeneous multilayer structure is realized, and an overall differential correction precision bottleneck is broken through.

This invention relates to a testing method for finite-time stealth attacks on tracking control systems. The method includes: constructing a system model based on the state variables and system matrix of the target system; determining a controller gain matrix based on the system model, a reference tracking model, and a preset positive definite symmetric matrix; constructing a state observer and an attack detector for the target system using a fully symmetric polytope method and the controller gain matrix; determining stealth attack conditions based on the state observer, attack detector, and controller gain matrix; constructing multiple objective functions for multi-stage stealth attacks within a preset time period according to a preset expectation set and stealth attack conditions; and determining the optimal attack method for security testing of the target system based on the optimal solution of each objective function. This invention improves the effectiveness of stealth attack testing by determining stealth attack conditions through a fully symmetric polytope method and control model.

The invention discloses a dual three-phase motor nine-switch current source driven two-degree-of-freedom active-disturbance-rejection control method, belongs to the technical field of permanent magnet motor control, and aims to solve the problem of tracking and disturbance rejection performance coupling of traditional active-disturbance-rejection control in current control of a dual three-phase permanent magnet synchronous motor based on a current source type inverter. The method comprises the following steps: establishing a dq-axis current loop mathematical model of a nine-switch current source type inverter containing periodic disturbance and non-periodic disturbance under a synchronous rotating coordinate system; designing an improved extended state observer, deleting a current tracking signal in a traditional extended state observer and an error correction term of a differential signal updating law of the current tracking signal, and designing a disturbance updating law as a proportional-integral-repetitive control structure; and constructing a two-degree-of-freedom active disturbance rejection controller based on the control characteristics of the current source type inverter, and independently adjusting the reference tracking performance and the disturbance rejection performance based on the output of the improved expansion state observer to realize complete decoupling of the two.

The invention discloses a solid propellant stretching optical flow measurement method and device. The method comprises the following steps: acquiring a surface deformation video sequence of a solid propellant sample under the action of a uniaxial stretching load in real time; on the basis of the surface deformation video sequence, a feature point tracking area is established in a scale distance section of the surface of the sample, initial feature points are arranged by adopting a regular grid point method, and the initial feature points serve as reference tracking points of optical flow tracking and displacement transmission; and based on the displacement data of the reference tracking point, constructing full-field continuous displacement distribution by adopting a bilinear interpolation algorithm, and finally obtaining a two-dimensional full-field displacement field of the sample. The device comprises a data acquisition unit, a feature point tracking unit and a two-dimensional full-field displacement field acquisition unit. According to the method, on the basis of a Lucas-Kanade sparse optical flow algorithm, feature points are selected for calculation in combination with propellant material deformation characteristics; a frame superposition algorithm is designed, and the displacement field measurement precision is improved.

The application discloses a kind of multi-rotor unmanned aerial vehicle master control in loop simulation method, system and storage medium, it is related to unmanned aerial vehicle in loop simulation technical field.Under the master control in loop simulation architecture, the virtual total thrust and virtual total torque are generated after the propulsion control instruction of master control output is modeled after propulsion chain pure lag and first-order inertia response, and virtual flight state is solved after wind disturbance is introduced into flight dynamics model;For the multi-source delay and dynamic distortion of propulsion side, dynamics calculation side, platform execution side and platform mechanism side, respectively, compensation is carried out, and the compensated predicted virtual flight state is mapped as Stewart platform reference instruction, and the inertial measurement unit is excited by platform motion, and the master control is fed back to form a closed loop.The application can restore the time sequence relationship between control instruction, propulsion response and body motion, improve the overall consistency and platform reference tracking accuracy of master control in loop simulation under wind disturbance.

The invention relates to the field of photovoltaic power generation, and discloses a photovoltaic panel dust retention light transmittance fluctuation-based tracking angle adjustment method, which comprises the following steps of S1, acquiring a reference tracking angle of a photovoltaic panel; s2, acquiring a periodic angle disturbance signal with specific injection frequency and amplitude; s3, acquiring a compensation angle of the photovoltaic panel; s4, synthesizing the reference tracking angle, the compensation angle and the periodic angle disturbance signal into an actual tracking angle instruction; and S5, a photovoltaic tracker is driven by a servo motor, so that the photovoltaic tracker adjusts the angle along with the actual tracking angle instruction. The power generation efficiency can be improved by dynamically responding to the dust accumulation influence in real time, and the power change caused by dust accumulation light transmittance fluctuation can be captured in real time and an accurate angle mismatch error signal can be generated by actively injecting a periodic angle disturbance signal and combining synchronous demodulation processing; and the compensation angle is dynamically updated through the integral controller, so that the photovoltaic panel always approaches the actual optimal angle.