2331 results about "Kinetic model" patented technology

A human-guided vision-force fused impedance iterative learning control method for a robotic arm, comprising: analyzing a robot-environment interaction dynamics equation, solving a visual servo acceleration model, and making use of the equation to establish a human-robotic arm-environment interaction dynamics model in an image feature space; acquiring an image feature position and speed curve of a human-guided robot completing an assembly task, and using dynamic movement primitives for coding and generalization; and designing an impedance iterative learning controller which uses image feature tracking errors as control input, learning impedance characteristics when the human-guided robot performs a contact operation, identifying unknown contact dynamics under the interaction between the robot and the environment, and counteracting identified contact interference in the feature space, so as to implement a flexible assembly operation. The control method solves the problems in existing assembly operations that human-robotic arm-environment coupling nonlinear dynamics, unknown contact dynamics of intensive contact assembly tasks and poor generalization of assembly scenarios require relearning for different scenarios, etc.

The invention relates to the technical field of control or regulation systems, and discloses a cooperative control method for a photovoltaic intelligent manufacturing equipment production line, and the method comprises the steps: a control unit collects the stock and change rate of materials in a physical cache region in real time; establishing and mapping the physical cache region into a virtual viscoelastic dynamic model with non-Newtonian fluid characteristics; calculating a virtual elastic restoring force enabling the stock to return to a balance point and a virtual viscous damping force preventing the stock state from changing based on the model; wherein an asymmetric anisotropic damping generation strategy is executed, and a damping coefficient is dynamically split according to a material flowing trend; and finally, superposing the virtual adjustment correction after vector synthesis to the basic transmission speed to generate a dynamic speed instruction, and by constructing a virtual dynamic field with rheological characteristics and an asymmetric damping mechanism, the problem of nonlinear cascade oscillation in discrete logistics transmission is solved, and differential self-adaptive suppression of accumulation and evacuation risks is realized.

The invention relates to a distributed trajectory planning method and system for an unmanned aerial vehicle cluster in an obstacle environment. The method comprises the steps of initializing a system, constructing a local Euclidean symbol distance site map, and realizing real-time interaction of state information of the unmanned aerial vehicle. An initial collision-free path is generated using jump point search. And solving an optimal control point through an L-BFGS algorithm through multi-constraint trajectory optimization in combination with a load swing dynamics model, four-rotor dynamics limitation, cluster collision avoidance and environment obstacle avoidance requirements. And a dynamic time redistribution strategy is adopted, and the track time interval is adjusted according to speed and acceleration overrun conditions. The method further comprises an adaptive re-planning mechanism, and local target points are updated in real time and adjacent aircraft collaborative optimization is triggered based on local map boundary detection and quadrotor track safety detection. According to the method, efficient, safe and stable trajectory planning of the hanging load quad-rotor unmanned aerial vehicle cluster in a complex environment is realized, the requirement of autonomously and efficiently completing tasks is met, and the task execution efficiency and safety are improved.

The invention discloses a method for establishing a high-voltage circuit breaker fault detection model, and the method comprises the following steps: collecting current, voltage, mechanical response, temperature and other multi-dimensional signals of a circuit breaker under different working conditions, and unifying the signals into standardized time sequence data; a nonlinear dynamic sparse identification method is utilized to establish a dynamic model for describing equipment state evolution, and sparse coefficients reflecting physical change rules are extracted from the dynamic model to serve as health features. And the features are fused with current monitoring data to generate a joint feature input vector, and a health prediction model based on a TabPFN architecture is introduced for training and discrimination. And finally, accurate prediction of the current health state or the potential fault of the circuit breaker is realized, and the model self-adaptive updating capability is realized. According to the method, physical modeling and data analysis are combined, so that the accuracy and interpretability of fault prediction are improved.

The invention discloses an intelligent automobile cooperative cruise safety control method under Dos attacks and physical faults, and relates to automobile intelligent safety and automatic driving. A hierarchical control framework is adopted and comprises an observer layer and a tracking layer. The method comprises the following steps: establishing a vehicle longitudinal dynamics model and a DoS attack model for DoS attack in a V2X communication network and the problems of sensor and actuator faults and parameter isomerism of a vehicle; a completely distributed self-adaptive preset time observer based on event triggering is designed for each following vehicle in the observer layer, and rapid and accurate estimation of the state of the pilot vehicle is achieved; the method comprises the following steps: constructing an augmentation system at a tracking layer, designing a distributed intermediate observer, carrying out online estimation and compensation on faults of a sensor and an actuator, designing a distributed active fault-tolerant controller based on a fault estimation value and a pilot vehicle state estimation value, and calculating a wheel driving torque to realize safe cruise control. Multiple threats of coexistence of network attacks and physical faults are effectively handled, and the stability and safety of the cooperative cruise system are ensured.

The invention discloses a fast training type unmanned aerial vehicle reinforcement learning strategy method suitable for a microcontroller, and belongs to the technical field of unmanned aerial vehicle control, and the method comprises the steps: constructing a standard four-rotor dynamic model containing the first-order delay characteristic of a motor, and simulating a real physical system; an asymmetric actor-commentator architecture is adopted to ensure the robustness and mobility of the strategy; a reward function coefficient is dynamically adjusted based on a course learning mechanism, and network parameters are optimized by adopting an offline strategy algorithm; performing network training and process optimization; and carrying out lightweight processing and deployment on the trained strategy network, landing the strategy in simulation to a real hardware platform with limited resources, jointly realizing rapid training, reliable migration and lightweight deployment, and realizing high-frequency autonomous flight control of the unmanned aerial vehicle on an embedded platform. According to the method, the high-performance reinforcement learning control strategy can be stably and efficiently operated on the microcontroller with limited resources, and the hardware cost and the power consumption of the intelligent unmanned aerial vehicle are greatly reduced.

The invention discloses a directional drilling track accurate control technology based on machine learning, and relates to the technical field of drilling engineering. The directional drilling track accurate control technology comprises the following steps that underground parameters of a drilling tool during underground operation are obtained; establishing a dynamic model based on the drilling tool structure and the motion state; based on the dynamic model, introducing a long-short-term memory neural network, and constructing a hybrid prediction model; drilling parameters are input into the hybrid prediction model, and trend information of multiple tracks is output; based on current geological conditions, drilling tool configuration and operation safety constraints, performing simulation evaluation on the trend information of the plurality of candidate tracks, and screening out an optimal track meeting track precision and underground safety requirements from the candidate tracks; based on the optimal track, control strategy input is constructed, and a state space containing a tool face angle, target azimuth deviation and a drilling tool state is set; and through a deep reinforcement learning method, an advanced adjustment instruction for the guiding tool is generated, and drilling operation is executed according to the optimal track and the advanced adjustment instruction.

This patent discloses a computer system for humanoid robot control system design and implementation, featuring a digital humanoid robot with dynamical models and a set of single-input-single-output (SISO) and multi-input-multi-output (MIMO) controllers. The system comprises a main software program, a generative Al humanoid robot intelligence engine, a robot motion path planner module, and a control system simulation engine. It enables efficient design, testing, validation, and implementation of robot control systems, significantly reducing time to market. The system supports seamless upgrades to accommodate new designs and components, enhancing applications in industrial automation, healthcare, public safety, and more, aligning with the goals of the 4th Industrial Revolution.

The invention belongs to the technical field of hypersonic flight vehicle control, and relates to a rigid-elastic coupling-oriented active and passive integrated control method for a hypersonic flight vehicle. The invention aims to realize stable tracking control of the elastic hypersonic flight vehicle. The method comprises the following steps: constructing a longitudinal dynamic model of the elastic hypersonic aircraft; self-adaptive identification of the elastic vibration frequency is realized through a cascaded self-adaptive filter; an elastic modal filtering estimation method is designed, and high-precision and low-cost elastic modal state quantity is provided for subsequent active feedback controller design; and then rigid-elastic coupling model decomposition is carried out, the control performance is ensured by using active disturbance rejection passive control for a rigid body subsystem, an RBF neural network is introduced for an elastic subsystem, an elastic mode is actively inhibited by using sliding mode control, and stable tracking of a reference instruction is realized. The method is an active and passive integrated control method for the hypersonic flight vehicle oriented to rigid-elastic coupling, and the application prospect is wide.

The invention belongs to the technical field of urban rail vehicle detection and maintenance, and discloses a full-process optimization method and system for polygonal wear of a metro vehicle wheel. The method comprises the following steps: firstly, constructing a digital twin-driven train rigid-flexible coupling dynamic model, carrying out global sensitivity analysis, establishing a sensitivity index model, screening key dynamic performance indexes, and carrying out batch simulation to construct a dynamic response database; feature extraction and classification model training are carried out on the index data, multi-layer wavelet packet decomposition is carried out on the one-dimensional vibration signals, and a multi-channel feature vector is constructed and input into a one-dimensional residual network model; inputting actually acquired data into the trained model, calculating a relative close degree to generate a comprehensive index and a grading result, and generating turning repair suggestions based on grading; meanwhile, multi-source monitoring data are collected, a long-short-term memory network is used for predicting the abrasion evolution trend, finally, turning repair suggestions and trends are integrated, an accounting model and an evaluation system are constructed, and an optimal maintenance decision is generated through a multi-target optimization algorithm.

A multi-robotic-arm collaborative control method. The method comprises the steps of: according to process requirements of an execution task and on the basis of a plurality of robotic arms, constructing a multi-arm coupled dynamics model, the multi-arm coupled dynamics model being used for position decomposition and force distribution for the plurality of robotic arms; on the basis of the multi-arm coupled dynamics model, decomposing the execution task to form a plurality of sub-task segments; on the basis of the sub-task segments, forming a collaborative controller group, the collaborative controller group consisting of controllers corresponding to a plurality of robotic arms under a master terminal; and, by means of the collaborative controller group, controlling each robotic arm to execute a path of a sub-task segment. The present application solves the problems of poor coordination and adaptability among various robotic arms when confronted with unknown working environments and uncertainties of task objectives during cooperative operation of the multiple robotic arms. Further provided are a multi-robotic-arm collaborative robot system and a storage medium.

The invention relates to a neural network-based adaptive control method for a vehicle formation system. Compared with the prior art, the method solves the defects that a complete target trajectory cannot be known in advance due to the influence of an actual environment and the control direction is unknown due to the unstable state of a vehicle engine. The method comprises the following steps: acquiring tracking data of a vehicle formation system; establishing a dynamic model of the vehicle formation system; establishing an error equation; designing a nusturb type function; designing a radial basis function neural network architecture; designing a GRNN training set and an evaluation index for trajectory reconstruction; designing an event trigger function of an actual controller of the system; carrying out self-adaptive control on the vehicle formation; and controlling the vehicle formation with an unknown target trajectory and an unknown control direction. According to the method, a neural network (NN)-based adaptive control architecture is adopted for a vehicle formation system (VPSs), a real-time trajectory can be predicted online by using a GRNN based on historical data of a target trajectory, and an unknown nonlinear term in the system is compensated, so that the position of a vehicle tracks the predicted target trajectory.

The invention relates to the field of nonlinear system control, and provides a preset time reinforcement learning method and system of a continuous nonlinear system, and an electronic device, and the method comprises the steps: constructing a zero-sum game framework based on a kinetic model and a performance index of the nonlinear system; determining a value function and a Hamiltonian function based on a zero-sum game framework; applying a preset neural network model to carry out approximation on the value function, and determining an approximation error; based on the Hamiltonian function and the approximation error, an approximate optimal control strategy and a worst interference strategy are determined; constructing a Lyapunov function based on the value function and the weight error of the value function; and based on a Lyapunov function, an approximate optimal control strategy and a worst interference strategy, a reinforcement learning result is verified. The method and the device are used for overcoming the defects that convergence time cannot be dynamically adjusted, parameter complexity is high and robustness is insufficient in the prior art, and the scheme of the invention can meet dual requirements of a continuous nonlinear system on dynamic convergence and anti-interference performance.

The invention discloses a rotor reliability constrained rolling bearing assembly parameter robust design method. The method comprises the following steps: constructing a dynamic model for an actual rotor-bearing system; constructing an uncertainty parameter vector and a design variable vector; a target function based on robustness and a constraint function based on reliability are constructed, so that an uncertainty optimization model is obtained; constructing an augmented input variable, and establishing a candidate orthogonal polynomial basis function set; on the basis, constructing and evaluating polynomial chaos-Kriging models for the target function and the constraint function respectively, and screening out an optimal polynomial chaos-Kriging model; calculating the expectation and the standard deviation of the target function and the failure probability of the constraint function under each design variable vector; and converting the uncertainty optimization model into an unconstrained single-target optimization model, randomly generating population individuals of a heuristic optimization algorithm in a feasible region of design variables, and iteratively searching an optimal solution of the unconstrained single-target optimization model as a rolling bearing assembly scheme.

The invention relates to a DRL-based AUV (Autonomous Underwater Vehicle) path planning and obstacle avoidance method in a partially observable environment, which comprises the following steps of: firstly, executing an AUV path planning and obstacle avoidance task under the condition that marine environment information is partially observable; secondly, constructing an underwater environment model and an AUV kinetic model, and constructing a dynamic and static separated dual-channel adaptive sensing model through local environment information acquired by detection equipment; then, path planning is modeled as a Markov decision process, and a multi-objective optimization reward function is designed for collaborative optimization; finally, an improved maximum entropy SAC algorithm is combined with a priority experience playback mechanism to train a strategy network, a gliding mode is dynamically triggered through a physical constraint layer, attitude angle and thrust output is corrected in real time, and the safety boundary condition is monitored in real time. According to the method, the real-time performance of path planning, the dynamic and static obstacle avoidance success rate and the ocean current utilization rate under the complex sea condition can be remarkably improved, and a high-robustness autonomous navigation solution is provided for ocean exploration.

The present disclosure discloses a method for attitude control of a quadrotor UAV, comprising establishing an attitude dynamics model of the quadrotor UAV, establishing a motion equation and a state-space equation of a UAV control system, determining an LADRC-CFO, and establishing a differential tracker for reducing a system overshoot.

The invention belongs to the technical field of semiconductor defect detection, and relates to a deep learning-based deep energy level defect identification method, system and device, and a medium. According to the method, transient curve data of an external capacitor is modulated to obtain a time sequence feature vector containing time, temperature and bias voltage information, then a multi-defect-induced weighted capacitor feature vector is obtained through calculation and modulation of a neural network model, and the time sequence feature vector and the multi-defect-induced weighted capacitor feature vector are trained after being spliced. Until the neural network model converges, the trained neural network model predicts a defect parameter vector, defect weight distribution and transient capacitance response at the (i + 1) th moment at the ith moment; according to the method, DLTS test data is modeled and analyzed in combination with a deep learning algorithm and a semiconductor deep energy level defect kinetic model, precise recognition and quantification of a complex defect system are achieved, the problem that defect recognition precision is insufficient in an existing DLTS test method is solved, and meanwhile the method is low in requirement for test equipment and high in applicability.

The invention belongs to the technical field of active suspension control, and particularly relates to an active suspension control method and system based on Skhook-ADRC-FFORC. According to the method, key parameters are obtained by establishing a 1 / 4 vehicle suspension dynamical model, a state space model is constructed, active disturbance suppression control is introduced on the basis, and the stability of the vehicle suspension is improved. Real-time estimation and compensation of vehicle body displacement and total disturbance of a system are realized, fractional order sliding mode control is combined, a fractional order sliding mode surface is constructed by using a Caputo operator, high robustness and smoothness are considered, the buffeting problem of a traditional sliding mode is suppressed, and finally, the high-frequency vibration isolation advantage of Skyhook control and the anti-interference capability of ADRC-FFORC are subjected to weighted fusion. The problems that in the prior art, model simplification and control distortion are caused, and a single strategy cannot give consideration to multi-target performance are solved, and by fusing ADRC, FOSMC and Skyhook control and realizing adaptive weighted switching, the robustness, smoothness and real-time performance of the active suspension under complex working conditions are effectively improved, so that collaborative optimization of vehicle comfort and safety is realized.

The invention discloses a floating type offshore wind turbine aerodynamic load calculation method based on a frequency domain method. The method comprises the steps that a state derivative is determined based on a blade element momentum theory quasi-steady-state assumption, and a dynamic inflow model is used for correction; establishing a pneumatic-elastic coupling two-degree-of-freedom linear dynamic model; establishing a wind wheel kinetic model; establishing a control model according to a floating type offshore wind turbine pitch angle control strategy, a torque control strategy and a floating body motion feedback control strategy; based on the pneumatic-elastic coupling two-degree-of-freedom linear dynamic model, the wind wheel dynamic model and the control model, pneumatic additional mass and pneumatic additional damping are obtained; pneumatic exciting force acting on the hub when the position of the cabin is fixed is obtained; and based on the pneumatic additional mass, the pneumatic additional damping and the pneumatic exciting force, an overall linear dynamic model of the fan is established and solved, and overall response is obtained. According to the method, the calculation result precision can be improved while the calculation efficiency is ensured, and a technical support is provided for accelerating the simulation efficiency of the offshore wind turbine.

The invention provides a self-adaptive fractional order nonsingular terminal sliding mode control method based on time delay estimation, and relates to the technical field of manipulator control. Designing a non-singular terminal sliding mode surface containing fractional differential and power transformation, and further constructing a reaching law with variable gain; then, the unknown dynamic state and disturbance of the system are compensated in real time through a time delay estimation technology by utilizing historical control input and state information; finally, a control law is generated by integrating the sliding mode surface, the reaching law and the time delay estimation compensation item, high-precision and strong-robustness control over an uncertain nonlinear system is achieved, and the tracking precision and the anti-interference capacity of the manipulator in the uncertain environment are effectively improved.

The invention discloses a constraint following control method of a two-arm collaborative robot based on differential homeomorphic mapping, and relates to the field of robot control. According to the method, a nominal dynamical model of a robot is established, and a geometric constraint equation is defined according to a cooperative task; establishing global differential homeomorphic mapping by constructing constraint tracking variables and complementary coordinates, and decoupling a system dynamics model into a constraint tracking subsystem and a complementary subsystem; designing a robust adaptive controller fusing sliding mode control and an adaptive law, and estimating and compensating the uncertainty of the system on line; according to the method, the constraint processing complexity is effectively simplified through differential homeomorphic mapping, the constraint tracking precision and robustness of the system under model uncertainty and external interference are remarkably improved, and high-precision and high-stability cooperative task execution of the two-arm robot in the dynamic environment is achieved.

The invention discloses a flood control toughness evolution simulation method and system based on natural-social element interaction, and the method comprises the steps: obtaining a natural element space-time sequence and a social element space-time sequence of a target region, and aligning the natural element space-time sequence and the social element space-time sequence to a unified space-time grid node, thereby obtaining a heterogeneous node sequence; calculating the time-varying interaction strength between the nodes along with the time and the interaction uncertainty of the time-varying interaction strength; compiling and generating a time-varying rule parameter between each pair of interaction nodes by using a preset rule compiler; and inputting the time-varying rule parameters into a pre-constructed toughness dynamic model, driving and updating the toughness state value of each node in continuous time steps, and obtaining a flood control toughness evolution track. According to the method, the problems of data driving and mechanism model splitting in the prior art are solved through a rule compiling mechanism, dynamic conversion of interaction characteristics from soft weights to hard rules is realized, and the accuracy and interpretability of flood control toughness evaluation in a complex time-varying scene are effectively improved.

The invention discloses a dynamic modeling tracking control method for a self-adaptive fiber placement compaction mechanism, and belongs to the technical field of specific model calculation systems, and the method comprises the following steps: S1, building a kinematic model of the self-adaptive fiber placement compaction mechanism; s2, establishing a workpiece surface model, and solving tangent point coordinates and position coordinates of the driving wheel, the driven wheel and the ground; s3, according to the force balance equation and the moment balance equation of each rod, the relation between each bearing reaction and the external force is solved; s4, establishing a dynamic model of the driving device; and S5, error analysis and optimization are carried out, and it is ensured that the main pressure is within the fluctuation range. According to the device, it can be ensured that the rollers output constant compaction force when the abrupt change molded surface structure is laid, and the forming defect caused by uneven compaction force is avoided.

The invention discloses a quadrotor unmanned aerial vehicle finite time sliding mode attitude control method based on reinforcement learning, and the method comprises the steps: building a quadrotor unmanned aerial vehicle attitude dynamic model considering external disturbance and model uncertainty, and rewriting the quadrotor unmanned aerial vehicle attitude dynamic model into a standard state equation form; a controller nominal item based on a nonsingular integral terminal sliding mode method is designed, an arrival control law item containing a time-varying matrix is designed, and the sum of the arrival control law item and the nominal item is an attitude controller; based on the reinforcement learning principle, an extended Markov decision process is constructed, a corresponding state set, an action set, action implementation mapping, a reward function and the like are designed, and a time-varying matrix is obtained through reinforcement learning training and is set online. According to the method, the dynamic performance and the steady-state performance of the system are effectively improved, and meanwhile high robustness is achieved under turbulence disturbance.

The invention relates to the technical field of nano particle heterogeneous reaction kinetic research, in particular to a nano single particle reaction rate measuring system and method based on optical surface wave dark field imaging. The method comprises the following steps: preparing a nano single particle sample; the humidity of the environment where the nano single particles are located is adjusted; starting a heterogeneous reaction of the nano single particulate matter, performing in-situ dark field imaging on the nano single particulate matter in the reaction process, and collecting scattering hot spot imaging data under different reaction times and different humidity; and based on the acquired imaging data, the reaction rate of the nano single particulate matter is inversed in combination with a kinetic model. According to the invention, the defects in the existing nano-particle reaction rate measurement technology can be overcome, and the reaction rate of a single particle as low as 50 nanometers in the heterogeneous reaction process can be accurately measured and inverted under the high-humidity standard atmospheric pressure condition.

Provided are a planning and control method for a legged robot, an apparatus, a robot, and a storage medium. The method includes: determining, based on the movement state and the swing parameter of each leg at the next moment, a foot-end reference state of each leg, and planning, based on a foot-end dynamic model and a discrete collision model, an impact-aware swing leg trajectory for a leg that starts to swing at the next moment; and performing, based on the movement state, the reference state sequence, and the foot-end reference state, whole-body control on the legged robot, and performing, based on a whole-body dynamic model and the discrete collision model, impact-aware whole-body control for a leg that is a supporting leg in planning but does not touch the ground.

The invention discloses an early warning method and device for climbing instability of a crawler unmanned vehicle and a medium, and relates to the technical field of unmanned vehicle climbing control. The early warning method comprises the steps that a digital twin model of the tracked unmanned vehicle is constructed, the digital twin model fuses a motor dynamic model and a whole vehicle dynamic model, and virtual scene data is generated; real-time data of a multi-source sensor of the tracked unmanned vehicle are obtained, wherein the real-time data comprise vehicle posture data, environment data and driving system data; and constructing a time convolutional network prediction model based on transfer learning, performing pre-training by using virtual scene data as source domain data, and performing fine tuning by using real-time data as target domain data to form a target domain prediction model. And inputting the current data into the target domain prediction model, and outputting a predicted vehicle state parameter prediction sequence. According to the vehicle state parameter prediction sequence and a preset safety threshold value, the instability risk is evaluated, an early warning signal is sent out, and the generation logic of the early warning signal is evaluated and optimized through an early warning performance loss function.

The invention relates to the technical field of hoisting apparatus simulation, and discloses a hoisting apparatus simulation and test verification method based on digital twinning. The method comprises the following steps: constructing a virtual twinborn body of the hoisting apparatus, wherein the virtual twinborn body comprises a geometric model, a kinematic model and a dynamic model corresponding to a physical hoisting apparatus; collecting real-time operation monitoring data of a physical hoisting instrument, and synchronizing the real-time operation monitoring data to the virtual twinborn body; in the virtual twin body, performing standardization processing on the received real-time operation monitoring data to generate a standardized data set; inputting each multi-source measurement data set into a pre-configured feature extraction engine, and outputting a corresponding feature group; performing feature integration operation on each feature group to generate integrated feature representation; and mapping the integrated feature representation to the virtual twin, and executing a simulation test in the virtual twin based on the integrated feature representation and a preset evaluation model library to generate an evaluation result set.

The invention discloses a reinforcement learning unmanned ship path control method based on double-track regulation and control random network distillation. The method comprises the operation steps that an unmanned ship builds a path tracking simulation environment and a kinetic model; the unmanned ship builds a core algorithm flexible action evaluation algorithm framework; the unmanned ship deploys a priority experience playback pool based on quality and success guidance; an uncertainty perception and risk perception mechanism is introduced into the unmanned ship; the unmanned ship builds a success rate-based reward attenuation and cold start module, and the unmanned ship calculates a total reward and designs a reward softening mechanism to smooth the total reward; the unmanned ship imports hyper-parameters of all the modules, starts training circulation in a simulation environment, and dynamically adjusts exploration intensity and the like; according to the method, uncertainty and risk indexes are introduced, the exploration intensity of the intelligent agent is controlled, the intelligent agent is prevented from making dangerous actions, and the robustness is improved; a priority experience playback pool based on quality and success guidance is introduced, high-quality samples are better played back, and strategy convergence is accelerated.

The invention discloses a PINN-based magnetic levitation vehicle nonlinear predictive control method and device and a medium, and relates to the field of rail transit, and the predictive control method comprises the following steps: S1, constructing a dynamic model; s2, learning and approaching unknown dynamic and time-varying parameters in the dynamic model by adopting a physical information neural network; s3, designing an optimization cost function meeting system input and output constraints based on the unknown dynamic state of the dynamic model predicted by the physical information neural network; and S4, constructing a step-by-step variable terminal constraint set to process input time lag. In a controller design and theoretical analysis process, a non-linear dynamic form of a system is completely reserved, and a stepping variable terminal constraint set suitable for input time delay is constructed, so that the controller can still maintain stability and dynamic performance under the condition that a system state obviously deviates from a balance point. According to the method, the operation safety and the control precision of the maglev vehicle under complex working conditions such as parameter drift and time lag disturbance are effectively improved.