Intelligent control system for optimizing sorting process of medicine storage

Abstract

The invention discloses a medicine storage sorting process optimization intelligent control system, and relates to the technical field of intelligent control, and the system comprises a chaos construction module which collects medicine order data and storage parameters in real time, constructs a chaos sorting demand field, and generates a space-time probability cloud through a Lorentz equation; the path planning module inputs the space-time probability cloud into a quantum decision tree, generates a sorting path through a quantum annealing algorithm, and obtains a quantum hash value of the molecular vibration spectrum based on the sorting path; and the execution verification module is used for decoding the sorting path into a photon lattice instruction to drive the sorting robot to execute a task, synchronously triggering quantum tunneling verification to compare the actual molecular characteristics with the quantum hash value, and generating a verification confidence coefficient. According to the invention, high-dimensional modeling and prospective path prediction of multivariable dynamic factors in a drug storage environment are realized, and adaptability and planning precision in a complex environment are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent control, in particular to a medicine warehouse sorting process optimization intelligent control system. BACKGROUND

[0002] In medicine warehouse management, the early sorting process mainly relies on manual operation and automated equipment based on simple rules. For example, using barcodes or RFID tags for item identification and tracking, moving goods through pre-set paths, and using database management systems to record inventory information. In the order processing process, genetic algorithms or simulated annealing algorithms are often used to optimize strategies to plan the optimal path in order to improve efficiency and reduce costs. This method relies on historical data for demand forecasting and path planning, trying to adapt to future demand changes through fixed parameter settings.

[0003] However, there are significant limitations: 1. When faced with highly dynamic order demand, it is difficult to make adaptive adjustments due to fixed model structure and pre-set parameters, resulting in low resource utilization and delayed response. 2. Existing verification mechanisms usually rely on scanning barcodes or RFID tags one by one, which not only takes a long time, but also is prone to reading errors in high-density storage environments, seriously affecting sorting accuracy. SUMMARY

[0004] In view of the above existing problems, the present application is proposed.

[0005] Therefore, the present application provides a medicine warehouse sorting process optimization intelligent control system to solve the problems of low resource utilization, delayed response, and poor sorting accuracy of existing medicine warehouse sorting systems in response to dynamic order demand and high-density storage environments.

[0006] To solve the above technical problems, the present application provides the following technical solutions:

[0007] The present application provides a medicine warehouse sorting process optimization intelligent control system, which includes a chaotic construction module that collects real-time medicine order data and warehouse parameters and constructs a chaotic sorting demand field, generates a space-time probability cloud through the Lorenz equation; a path planning module that inputs the space-time probability cloud into a quantum decision tree, generates a sorting path through a quantum annealing algorithm, and obtains a quantum hash value of the molecular vibration spectrum based on the sorting path; an execution verification module that decodes the sorting path into photonic crystal lattice instructions to drive a sorting robot to perform tasks, synchronously triggers quantum tunneling verification to compare actual molecular characteristics with the quantum hash value, and generates a verification confidence, wherein the actual molecular characteristics are the molecular vibration spectrum of the medicine obtained from the medicine order data; a feedback correction module that dynamically corrects the chaotic field parameters according to the verification confidence and feeds back to the quantum decision tree to generate a corrected sorting path.

[0008] As a preferred scheme of the medicine warehouse sorting process optimization intelligent control system, the order data includes a medicine unique identification code, a target sorting port coordinate, a medicine volume, a medicine mass, a priority label, and an invalid time.

[0009] The warehouse parameters include a sorting robot real-time position, a sorting line speed, a sorting port remaining capacity, a sorting port maximum capacity, and an obstacle distribution.

[0010] As a preferred scheme of the medicine warehouse sorting process optimization intelligent control system, the steps of constructing the chaotic sorting demand field are as follows.

[0011] The Euclidean distance of the sorting robot to the target sorting port is calculated, and the Euclidean distance is mapped to a field strength attenuation factor.

[0012] The priority label is mapped to a continuous exponential weight through an exponential decay function, a time attenuation factor is defined according to an inverse square root function of the invalid time, and a comprehensive weight is generated.

[0013] Based on the obstacle distribution, the obstacle density around the robot is calculated through three-dimensional Gaussian convolution, and the obstacle influence factor is calculated in combination with the medicine mass.

[0014] Based on the sorting port remaining capacity and the sorting port maximum capacity, the capacity attenuation factor is calculated using a quadratic function.

[0015] The field strength attenuation factor, the comprehensive weight, the obstacle influence factor, and the capacity attenuation factor are combined into an original field strength matrix, and the chaotic sorting demand field is generated after normalization.

[0016] As a preferred scheme of the medicine warehouse sorting process optimization intelligent control system, the steps of generating the space-time probability cloud through the Lorenz equation are as follows.

[0017] Based on the sorting robot real-time position, the velocity component is calculated through difference and the sorting robot motion direction angle is obtained.

[0018] The sorting robot motion direction angle is injected into the Lorenz equation with the chaotic sorting demand field to generate the space-time probability cloud.

[0019] As a preferred scheme of the medicine warehouse sorting process optimization intelligent control system, the steps of inputting the space-time probability cloud into the quantum decision tree and generating the sorting path through the quantum annealing algorithm are as follows.

[0020] The probability density distribution of the space-time probability cloud is taken as the weight of the quantum decision tree leaf node, and a quantum decision tree node weight matrix is generated.

[0021] Based on the quantum decision tree node weight matrix, a target function is constructed, a constraint condition is set, the target function and the constraint condition are coded into a QUBO matrix, and the QUBO matrix is loaded into a quantum annealing machine to generate a sorting path.

[0022] As a preferred scheme of the medicine warehouse sorting process optimization intelligent control system, wherein: the quantum hash value based on the molecular vibration spectrum obtained by the sorting path, the specific steps are,

[0023] Obtain the molecular vibration spectrum, and extract the geometric features of the sorting path;

[0024] The geometric features of the sorting path and the molecular vibration spectrum are combined into a time domain signal, and converted into a frequency domain complex amplitude through quantum Fourier transform;

[0025] The frequency domain complex amplitude is encoded into a quantum state through a quantum register, and a binary hash sequence is generated through multiple quantum state evolution and measurement;

[0026] The binary hash sequence is anti-collision coded to generate a quantum hash value of the molecular vibration spectrum.

[0027] As a preferred scheme of the medicine warehouse sorting process optimization intelligent control system, wherein: the sorting path is decoded into photonic crystal lattice instructions to drive the sorting robot to perform tasks, and the specific steps are,

[0028] Based on the sorting line speed, the sorting path is decomposed into straight line segments and arc line segments, and the motion parameters of the straight line segments and the arc line segments are calculated respectively;

[0029] Based on the motion parameters, the photonic crystal lattice parameter mapping is performed through displacement coding, speed coding and steering coding, and the photonic crystal lattice instructions are generated and executed.

[0030] As a preferred scheme of the medicine warehouse sorting process optimization intelligent control system, wherein: the quantum tunneling verification is triggered synchronously to compare the actual molecular features with the quantum hash value, and the specific steps are,

[0031] Perform quantum Fourier transform on the molecular vibration spectrum to generate real-time frequency domain amplitude, and generate a two-qubit entangled state by encoding the quantum hash value and the real-time frequency domain amplitude into a quantum state;

[0032] Based on the quantum annealing machine, the quantum tunneling coupling strength parameter is calibrated, and the quantum tunneling Hamiltonian loaded in the quantum processor is obtained;

[0033] The two-qubit entangled state and the quantum tunneling Hamiltonian are evolved through the Schrodinger equation to generate an evolved quantum state;

[0034] The Z basis projection measurement is performed on the evolved quantum state, the joint probability distribution of the quantum hash value and the frequency domain amplitude is counted, the ground state frequency is recorded, and the joint probability distribution matrix is generated.

[0035] As a preferred scheme of the medicine warehouse sorting process optimization intelligent control system, the generation verification confidence refers to the sum of the joint probability of the joint probability distribution matrix and the indicator function, and the verification confidence is calculated.

[0036] As a preferred scheme of the medicine warehouse sorting process optimization intelligent control system, the dynamic correction of the chaotic field parameters according to the verification confidence, and the feedback to the quantum decision tree to generate the corrected sorting path refers to converting the verification confidence into a correction weight coefficient, correcting the chaotic field parameters according to the correction weight coefficient, and generating the corrected sorting path.

[0037] The beneficial effects of the present application are: by constructing a sorting demand field based on chaotic dynamics and generating a space-time probability cloud, high-dimensional modeling and forward-looking path prediction of multivariate dynamic factors in the medicine warehouse environment are realized, and the adaptability and planning accuracy in complex environment are improved; meanwhile, a quantum tunneling verification mechanism is introduced, and quantum entanglement and tunneling effect are used to realize high-speed and high-precision synchronous verification of medicine molecular characteristics. BRIEF DESCRIPTION OF DRAWINGS

[0038] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0039] Fig. 1 It is a schematic diagram of the medicine warehouse sorting process optimization intelligent control system.

[0040] Fig. 2 It is a flowchart for constructing a chaotic sorting demand field.

[0041] Fig. 3 It is a flowchart for path planning.

[0042] Fig. 4 It is a flowchart for generating a joint probability distribution matrix. DETAILED DESCRIPTION

[0043] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings of the specification.

[0044] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0045] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0046] Reference Figs. 1-4 As one embodiment of the present invention, this embodiment provides an intelligent control system for optimizing the pharmaceutical warehousing and sorting process, including the following steps:

[0047] The chaos construction module collects drug order data and warehousing parameters in real time, constructs a chaotic sorting demand field, and generates a spatiotemporal probability cloud through the Lorentz equation.

[0048] The system periodically acquires drug order data and warehousing parameters through a distributed sensor network. The drug order data includes the unique identifier of the drug, the coordinates of the target sorting port, the volume of the drug, the quality of the drug, the priority label, and the expiration time. The warehousing parameters include the real-time position of the sorting robot, the sorting line speed, the remaining capacity of the sorting port, the maximum capacity of the sorting port, and the distribution of obstacles.

[0049] It should be noted that the remaining capacity of the sorting port, the maximum capacity of the sorting port, and the coordinates of the obstacle distribution are captured in real time by LiDAR and vision sensors; the real-time position of the sorting robot is obtained by the global positioning sensor and inertial measurement unit built into the sorting robot; and the speed of the sorting line is measured by the sorting line encoder.

[0050] Calculate the Euclidean distance from the sorting robot to the target sorting point, and map the Euclidean distance to a field strength attenuation factor using a 1.5 power function. The expression is as follows:

[0051] in, For sorting robots To the target sorting port The field strength attenuation factor, For sorting robots To the target sorting port The Euclidean distance;

[0052] Priority labels are mapped to continuous exponential weights using an exponential decay function, and a time decay factor is defined based on the inverse square root function of failure time.

[0053] It should be noted that priority tags generate continuous exponential weights by exponentially calculating the number of priority tags using a natural constant. The smaller the number of priority tags, the higher the priority. The exponential weight increases as the value decreases. The formula is: continuous exponential weight equals five times the natural constant minus the power of the priority tag value. Failure time generates a time decay factor by adding the failure time to the offset and taking the reciprocal of the square root. The smaller the failure time, the more urgent the timeliness. The time decay factor increases as the failure time decreases. The formula is: time decay factor equals the failure time plus the reciprocal of the square root of the offset. The continuous exponential weight is multiplied by the time decay factor to generate a comprehensive weight. The comprehensive weight integrates the impact of priority and timeliness on the sorting path.

[0054] Multiplying the exponential weight by the time decay factor generates the comprehensive weight, expressed as:

[0055] in, For sorting robots For the target sorting port Overall weighting, Priority label Continuous exponential weights, Expiration time Time decay factor;

[0056] Based on obstacle distribution, the obstacle density around the robot is calculated using 3D Gaussian convolution, and the impact of obstacles on the path is adjusted according to drug quality to generate an obstacle influence factor, expressed as:

[0057] in, For sorting robots Density of surrounding obstacles For the first The coordinates of the obstacle. The standard deviation of the Gaussian kernel. The total number of obstacles. The obstacle influence factor, For sorting robots Real-time location, For the sake of drug quality, Indexing obstacles;

[0058] Based on the remaining capacity and maximum capacity of the sorting port, the capacity decay factor is calculated using a quadratic function, expressed as follows: ;

[0059] in, It is the capacity decay factor. for the sorting port remaining capacity, for the sorting port maximum capacity, for the sorting port index;

[0060] multiply the field strength attenuation factor, the comprehensive weight, the obstacle influence factor and the capacity attenuation factor to generate the original field strength matrix, the expression is: ;

[0061] wherein, the original field strength matrix of the target sorting port for the sorting robot ;

[0062] normalize the original field strength matrix to generate the chaotic sorting demand field.

[0063] Based on the real-time position of the sorting robot, the velocity component is calculated by difference and the motion direction angle of the sorting robot is obtained;

[0064] Further, the real-time position coordinates of the sorting robot are stored in time sequence, the horizontal coordinate difference and the vertical coordinate difference of adjacent time stamps are extracted, the ratio of the horizontal coordinate difference to the time stamp difference is calculated to obtain the horizontal velocity component, and the ratio of the vertical coordinate difference to the time stamp difference is calculated to obtain the vertical velocity component; the ratio of the horizontal coordinate difference to the vertical coordinate difference is converted into the motion direction angle of the sorting robot by the inverse tangent function, the direction angle is expressed in radians and corrected to the range of zero to twice the circumference, and the horizontal velocity component, the vertical velocity component and the motion direction angle are output as the motion state parameters.

[0065] The motion direction angle of the sorting robot and the chaotic sorting demand field are injected into the Lorenz equation to generate the coupled Lorenz differential equation set, the expression is: ;

[0066] wherein, the horizontal motion of the sorting robot, the vertical motion of the sorting robot, the vertical direction motion of the sorting robot, the Prandtl number corresponding to the standard Lorenz equation, the Rayleigh number corresponding to the standard Lorenz equation, the geometric parameter corresponding to the standard Lorenz equation, the normalized chaotic sorting demand field component of the chaotic sorting demand field in the horizontal motion of the sorting robot ; the normalized chaotic sorting demand field component of the chaotic sorting demand field in the vertical motion of the sorting robot ; a normalized chaotic sorting demand field component in the vertical direction of the sorting robot, a sorting robot motion direction angle a converted angular velocity parameter, a time;

[0067] The coupled Lorenz differential equation set is numerically solved by the fourth-order Runge-Kutta method to generate three-dimensional chaotic trajectory points of the sorting robot;

[0068] It should be noted that the sorting robot motion direction angle is input as the angular velocity parameter of the Lorenz equation, and the chaotic sorting demand field is injected as an additional control term into the variable evolution term of the Lorenz equation to construct a coupled Lorenz differential equation set containing the dynamics of the direction angle and the field strength constraint. The coupled differential equation set is iteratively calculated by the fourth-order Runge-Kutta method, and the integration step is matched to the control period of the sorting robot. Three-dimensional chaotic trajectory point coordinates are output at each iteration step. The three-dimensional chaotic trajectory points are projected onto the two-dimensional coordinate system of the warehouse plane through coordinate transformation to generate chaotic motion trajectory points required for spatiotemporal probability cloud computing.

[0069] The three-dimensional chaotic trajectory points are projected onto a two-dimensional plane, and the probability density distribution is calculated using the Epanechnikov kernel function to generate a spatiotemporal probability cloud, expressed as: ;

[0070] wherein, is a spatiotemporal probability cloud on a two-dimensional plane coordinate is a bandwidth parameter, is the number of three-dimensional chaotic trajectory points, is a two-dimensional plane coordinate abscissa, is a two-dimensional plane coordinate ordinate, is a two-dimensional plane coordinate abscissa of the i-th three-dimensional chaotic trajectory point, is a two-dimensional plane coordinate ordinate of the i-th three-dimensional chaotic trajectory point, is an indicator function. The path planning module inputs the spatiotemporal probability cloud into the quantum decision tree to generate a sorting path through the quantum annealing algorithm, and obtains a quantum hash value of the molecular vibration spectrum based on the sorting path; The warehouse plane is divided into discrete coordinate grids including an abscissa sequence and an ordinate sequence, and each discrete coordinate grid corresponds to a leaf node of the quantum decision tree,

[0071]

[0072]

[0073] ​​​​The probability density distribution of the spatiotemporal probability cloud at the discrete coordinate grid is taken as the weight of the leaf node, and a quantum decision tree node weight matrix is generated;

[0074] The number of required quantum bits is determined according to the depth of the quantum decision tree, for example, if the discrete coordinate grid is divided into four by four, the depth is two, and two quantum bits are required;

[0075] Each leaf node index is converted into binary encoding, and the number of binary bits is equal to the number of quantum bits;

[0076] Each bit of the binary encoding corresponds to the state of a quantum bit, and the path from the root node to the leaf node is represented by a quantum bit sequence state, and a quantum decision tree path encoding table is obtained, which records the correspondence between the leaf node index and the quantum bit sequence, for example, path eleven indicates that the first quantum bit state is one and the second quantum bit sequence state is one;

[0077] The quantum decision tree node weight matrix is normalized, and the weight of each leaf node is squared to obtain the amplitude of the corresponding quantum state, the quantum register is initialized, the quantum bit sequence state is encoded into a quantum superposition state, and the amplitude of each basis state in the quantum superposition state is the square root of the weight of the corresponding leaf node;

[0078] The amplitude of each basis state in the quantum superposition state is extracted, the basis state amplitude is mapped back to the leaf node weight according to the quantum decision tree path encoding table, and a diagonal matrix form of the quantum decision tree node weight matrix is generated.

[0079] The total length of the minimum sorting path and the weighted sum of the weights of the diagonal matrix form of the quantum decision tree node weight matrix are taken as the objective function, and the constraint condition is set to avoid obstacles and sorting port capacity restrictions, the objective function and the constraint condition are encoded into a QUBO matrix, the QUBO matrix is loaded into a quantum annealer, the ground state is searched through quantum tunneling effect, and the binary decision variable sequence of the sorting path is output. The binary decision variable sequence is mapped back to the warehouse plane coordinate point to generate the sorting path;

[0080] It should be pointed out that the total length of the sorting path is defined as the sum of the Euclidean distance between adjacent coordinate points, the diagonal elements of the quantum decision tree node weight matrix are taken as the path point weight items, and the objective function is constructed as the weighted sum of the total length of the path and the path point weight items; the obstacle distribution is converted into the distance penalty item of the path point and the obstacle, and the remaining capacity restriction of the sorting port is converted into the path point selection constraint item; the objective function and the constraint item are combined into a quadratic unconstrained binary optimization model, and the objective function item and the constraint penalty item coefficient are filled into the upper triangular elements of the QUBO matrix; the QUBO matrix is loaded into the quantum annealer through the quantum annealer hardware interface, the quantum annealer initializes the quantum bits to the superposition state, performs quantum tunneling and thermal annealing mixed evolution, measures the final quantum bit state to generate a binary decision variable sequence; the variable index with a value of one in the binary decision variable sequence is mapped to the warehouse plane discrete coordinate grid point, and the sorting path coordinate sequence is generated by sorting according to the spatial adjacent relationship of the grid points.

[0081] According to the unique identification code of the medicine, the molecular vibration spectrum is obtained, the geometric features of the sorting path including the total length of the sorting path and the path curvature change rate are extracted, the geometric features of the sorting path are combined with the molecular vibration spectrum into a time domain signal, and the time domain signal is converted into a frequency domain complex amplitude through quantum Fourier transform;

[0082] It should be pointed out that when retrieving the molecular vibration spectrum time domain signal according to the unique identification code of the medicine, the total length of the sorting path is calculated by accumulating the displacement of the path point, and the path curvature change rate is calculated by the difference between the direction angles of adjacent path points divided by the interval; the total length of the sorting path and the path curvature change rate are aligned according to the time stamp to convert into a geometric feature time domain signal with equal interval sampling; the molecular vibration spectrum time domain signal and the geometric feature time domain signal are spliced into a mixed time domain sequence; the mixed time domain sequence is converted into a frequency domain complex amplitude through quantum Fourier transform, the modulus of the complex amplitude represents the frequency component intensity, and the argument represents the phase relationship, and a frequency domain complex amplitude array is generated as a quantum state encoding input.

[0083] The frequency domain complex amplitude is encoded into a quantum state through a quantum register, a quantum random walk protocol is applied, a binary hash sequence is generated through multiple quantum state evolution and measurement, and an anti-collision code is applied to the binary hash sequence to generate a quantum hash value of the molecular vibration spectrum.

[0084] It should be pointed out that the frequency domain complex amplitude is mapped to a multi-qubit superposition state through the amplitude encoding operation of the quantum register, and each complex amplitude corresponds to the amplitude component of the quantum state; the quantum random walk protocol performs evolution operation on the quantum state by alternately applying the conditional displacement operator and the coin operator, and performs projection measurement on the quantum state after multiple iterations; the measurement result is converted into a binary hash sequence according to the order of quantum bits; the binary hash sequence is compressed into a fixed-length quantum hash value of the molecular vibration spectrum through a preset anti-collision hash function, and the quantum hash value is output as a sorting task verification identifier.

[0085] The verification module is executed to decode the sorting path into photonic lattice instruction to drive the sorting robot to perform the task, and to synchronously trigger quantum tunneling verification to compare the actual molecular characteristics with the quantum hash value and to generate a verification confidence level;

[0086] Based on the sorting line speed, the maximum allowed line speed of the sorting robot and the longitudinal distance between adjacent path points are obtained, and the longitudinal speed is calculated; the longitudinal speed is normalized, and the maximum light intensity of the laser is combined to generate the photonic lattice intensity modulation parameter; the sorting path is decomposed into straight line segments and arc line segments, and the longitudinal speed of the straight line segments and the arc line segments is calculated respectively; through displacement coding, speed coding and turning coding, photonic lattice parameter mapping is performed to generate photonic lattice instructions;

[0087] Further, the sorting path is decomposed into straight line segments and arc line segments according to the distance between adjacent points and the change in the direction angle; the transverse displacement and the longitudinal displacement of the straight line segments are extracted, and the curvature radius and the change in the direction angle of the arc line segments are extracted; the longitudinal displacement of the straight line segments is divided by the control period time to obtain the longitudinal speed, and the longitudinal speed is divided by the maximum sorting line speed to generate the speed coding parameter; the transverse displacement of the straight line segments is divided by the laser wavelength and multiplied by twice the circumference to generate the displacement coding parameter; the change in the direction angle of the arc line segments is divided by the curvature radius to obtain the turning angle speed, and the turning angle speed is divided by the maximum turning angle speed of the sorting robot to obtain the inverse tangent function to generate the turning coding parameter; the displacement coding parameter is mapped into the photonic lattice phase modulation parameter, the speed coding parameter is mapped into the photonic lattice intensity modulation parameter, and the turning coding parameter is mapped into the photonic lattice polarization angle parameter; the phase modulation parameter, the intensity modulation parameter and the polarization angle parameter are combined into a three-tuple instruction according to the sorting path coordinate sequence to generate the photonic lattice instruction.

[0088] The photonic lattice instruction is encoded into an optical pulse signal, which is transmitted to the sorting robot through an optical fiber; the sorting robot analyzes the optical signal into motor control instructions to drive the sorting robot to perform the task.

[0089] The quantum Fourier transform is performed on the molecular vibration spectrum to generate a real-time frequency domain amplitude, and the quantum hash value and the real-time frequency domain amplitude are encoded into a two-qubit entangled state through quantum state encoding;

[0090] For example, the time-domain signal of the molecular vibration spectrum is converted into a frequency-domain complex amplitude through quantum Fourier transform, and the real part and the imaginary part of the frequency-domain complex amplitude are respectively encoded into the ground state amplitude component of the first qubit; each binary value of the pre-stored quantum hash value is encoded into the ground state amplitude component of the second qubit; the first qubit and the second qubit generate a two-qubit entangled state through a controlled non-gate operation, and the entangled state is in the form of a superposition state of the ground state amplitude of the first qubit and the ground state amplitude of the second qubit; the two-qubit entangled state is used as the input state of the quantum tunneling verification for subsequent joint probability distribution measurement.

[0091] The quantum annealer hardware calibrates the quantum tunneling coupling strength parameter based on the hardware, defines the quantum tunneling Hamiltonian, maps the Hamiltonian to the physical qubit coupler of the quantum processor, calibrates the coupling strength and evolution time, and obtains the quantum tunneling Hamiltonian loaded in the quantum processor.

[0092] For example, the quantum annealer hardware determines the tunneling coupling strength parameter between physical qubits through the calibration process, defines the quantum tunneling Hamiltonian as a linear combination of tunneling interaction terms between qubits, maps the coupling term coefficients of the quantum tunneling Hamiltonian to the adjustable coupler interface of the quantum processor, converts the coupling strength parameter to the driving signal amplitude, sets the time resolution of the quantum annealer annealing protocol as the evolution time parameter, calibrates the matching relationship between the coupler driving signal amplitude and the time parameter, and the quantum annealer hardware feedback interface reads the calibrated driving signal configuration to verify the consistency of the physical implementation and the theoretical definition of the quantum tunneling Hamiltonian, and outputs the quantum tunneling Hamiltonian parameter set loaded in the quantum processor.

[0093] The double-qubit entangled state and the quantum tunneling Hamiltonian are evolved through the Schrödinger equation to generate an evolved quantum state.

[0094] It should be noted that the double-qubit entangled state is input into the quantum processor as an initial quantum state, and the quantum tunneling Hamiltonian acts on the initial quantum state through a time evolution operator; the quantum processor executes time evolution according to the calibrated quantum tunneling Hamiltonian parameters and the preset evolution time parameters, the time evolution operator and the quantum tunneling Hamiltonian satisfy the corresponding relationship of the Schrödinger equation; the quantum processor outputs the evolved quantum state, which contains the interference superposition information of the quantum hash value and the real-time frequency domain amplitude, and is used as the final state of the quantum tunneling verification for subsequent joint probability distribution measurement.

[0095] The evolved quantum state is subjected to Z-basis projection measurement, the joint probability distribution of the quantum hash value and the frequency domain amplitude is calculated, the ground state occurrence frequency is recorded, and the joint probability distribution matrix is generated.

[0096] It should be noted that each qubit of the double-qubit entangled state is subjected to independent Z-basis projection measurement, and the measurement results are recorded as corresponding ground state combinations of the quantum hash value and the frequency domain amplitude; the measurement process is repeated multiple times, the frequency of the occurrence of each bit of the quantum hash value and each bit of the frequency domain amplitude is calculated, and the ratio of the frequency to the total number of measurements generates the joint probability distribution matrix; the row index of the joint probability distribution matrix represents the ground state combination of the quantum hash value, the column index represents the ground state combination of the frequency domain amplitude, and the joint probability distribution matrix element value is the joint probability of the corresponding ground state combination. The joint probability distribution matrix is output as an input parameter for confidence calculation.

[0097] The verification confidence is calculated by summing the joint probabilities of quantum hash value and frequency domain amplitude matching, as expressed by: ;

[0098] in, To verify the confidence level, The first quantum hash value Bit, The first frequency domain amplitude Bit, This represents the joint probability distribution of quantum hash value and frequency domain amplitude. This is an indicator function.

[0099] It should be noted that, when traversing each ground state of the quantum hash value and each ground state of the frequency domain amplitude, if the ground state of the quantum hash value is equal to the corresponding ground state of the frequency domain amplitude, the joint probability value of the ground state combination in the joint probability distribution matrix is ​​extracted; the joint probability values ​​of all matching ground state combinations are summed, and the sum is divided by the total number of bits of the quantum hash value to generate a verification confidence scalar value; the verification confidence scalar value ranges from zero to one, and the larger the value, the higher the degree of matching between the quantum hash value and the real-time molecular vibration spectrum. The verification confidence scalar value is used as the input parameter and output parameter for dynamically correcting the chaotic field parameters.

[0100] The feedback correction module dynamically corrects the chaotic field parameters based on the verification confidence level and feeds them back to the quantum decision tree to generate the corrected sorting path;

[0101] Convert the test confidence level into a corrected weighting coefficient. Used to quantify the magnitude of parameter correction. This indicates that no correction is needed. Indicates the maximum correction;

[0102] Based on the corrected weighting coefficients, the corrected chaotic field parameters include the field strength attenuation factor, the comprehensive weight, the obstacle influence factor, and the capacity attenuation. The corrected sorting path is then generated using the corrected chaotic field parameters.

[0103] Furthermore, the expression for the modified field strength attenuation factor is: ;

[0104] in, This is the corrected field strength attenuation factor;

[0105] The expression for the revised overall weight is as follows: ;

[0106] in, For the corrected priority label Continuous exponential weights, a time decay factor of the revised expiration time a revised sorting robot to the target sorting port a comprehensive weight;

[0107] The expression of the revised obstacle influence factor is: ;

[0108] wherein, a revised obstacle influence factor;

[0109] The expression of the revised capacity decay is: ;

[0110] wherein, a revised obstacle influence factor.

[0111] To sum up, the application realizes high-dimensional modeling and forward-looking path prediction of multi-variable dynamic factors in the medicine storage environment by constructing a sorting demand field based on chaotic dynamics and generating a space-time probability cloud, and improves adaptability and planning accuracy in complex environments. At the same time, a quantum tunneling verification mechanism is introduced to realize high-speed and high-precision synchronous verification of the characteristics of medicine molecules by using quantum entanglement and tunneling effect.

[0112] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and not to limit it. Although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the present application, and they should be covered in the scope of the claims of the present application.​

Claims

1. An intelligent control system for optimizing the pharmaceutical warehousing and sorting process, characterized by: include, The chaos construction module collects drug order data and warehousing parameters in real time, constructs a chaotic sorting demand field, and generates a spatiotemporal probability cloud through the Lorentz equation. The path planning module inputs the spatiotemporal probability cloud into the quantum decision tree, generates sorting paths through the quantum annealing algorithm, and obtains the quantum hash value of the molecular vibration spectrum based on the sorting paths; The execution verification module decodes the sorting path into photonic lattice instructions to drive the sorting robot to perform tasks, and simultaneously triggers quantum tunneling verification to compare the actual molecular features with the quantum hash value, and generates a verification confidence level; the actual molecular features are the molecular vibrational spectra of the drugs obtained from the drug order data; The feedback correction module dynamically corrects the chaotic field parameters based on the verification confidence level and feeds them back to the quantum decision tree to generate the corrected sorting path.

2. The intelligent control system for optimizing the drug warehousing and sorting process as described in claim 1, characterized in that: The order data includes the drug's unique identifier, target sorting port coordinates, drug volume, drug quality, priority label, and expiration time. The warehousing parameters include the real-time position of the sorting robot, the sorting line speed, the remaining capacity of the sorting port, the maximum capacity of the sorting port, and the distribution of obstacles.

3. The intelligent control system for optimizing the drug warehousing and sorting process as described in claim 2, characterized in that: The specific steps for constructing the chaotic sorting demand field are as follows: Calculate the Euclidean distance from the sorting robot to the target sorting port, and map the Euclidean distance to the field strength attenuation factor; Priority labels are mapped to continuous exponential weights using an exponential decay function, a time decay factor is defined based on the inverse square root function of failure time, and a comprehensive weight is generated. Based on obstacle distribution, the obstacle density around the robot is calculated using three-dimensional Gaussian convolution, and the obstacle influence factor is calculated in conjunction with drug quality. Based on the remaining capacity and maximum capacity of the sorting port, a quadratic function is used to calculate the capacity decay factor; The field strength attenuation factor, comprehensive weight, obstacle influence factor and capacity attenuation factor are combined into the original field strength matrix, and then normalized to generate the chaotic sorting demand field.

4. The intelligent control system for optimizing the drug warehousing and sorting process as described in claim 3, characterized in that: The specific steps for generating the spatiotemporal probability cloud using the Lorentz equation are as follows: Based on the real-time position of the sorting robot, the velocity component is calculated by differential calculation and the motion direction angle of the sorting robot is obtained. The sorting robot's motion direction angle and the chaotic sorting demand field are injected into the Lorentz equation to generate a spatiotemporal probability cloud.

5. The intelligent control system for optimizing the drug warehousing and sorting process as described in claim 4, characterized in that: The specific steps for inputting the spatiotemporal probability cloud into the quantum decision tree and generating sorting paths using the quantum annealing algorithm are as follows: The probability density distribution of the spatiotemporal probability cloud is used as the weight of the leaf nodes of the quantum decision tree, and a quantum decision tree node weight matrix is ​​generated. Based on the node weight matrix of the quantum decision tree, an objective function is constructed, constraints are set, the objective function and constraints are encoded into a QUBO matrix, the QUBO matrix is ​​loaded into the quantum annealing machine, and sorting paths are generated.

6. The intelligent control system for optimizing the drug warehousing and sorting process as described in claim 5, characterized in that: The specific steps for obtaining the quantum hash value of the molecular vibrational spectrum based on the sorting path are as follows: Obtain molecular vibrational spectra and extract the geometric features of the sorting path; The geometric features of the sorting path and the molecular vibration spectrum are combined into a time-domain signal, and then converted into a frequency-domain complex amplitude through quantum Fourier transform. The frequency domain complex amplitude is encoded into a quantum state through a quantum register, and a binary hash sequence is generated through multiple quantum state evolutions and measurements. Collision-resistant encoding is performed on the binary hash sequence to generate the quantum hash value of the molecular vibration spectrum.

7. The intelligent control system for optimizing the drug warehousing and sorting process as described in claim 6, characterized in that: The specific steps for decoding the sorting path into photonic lattice instructions to drive the sorting robot to perform tasks are as follows: Based on the sorting line speed, the sorting path is decomposed into straight line segments and arc segments, and the motion parameters of the straight line segments and arc segments are calculated respectively. Based on motion parameters, photonic lattice parameters are mapped through displacement encoding, velocity encoding, and steering encoding to generate and execute photonic lattice instructions.

8. The intelligent control system for optimizing the drug warehousing and sorting process as described in claim 7, characterized in that: The synchronously triggered quantum tunneling verification, which compares actual molecular characteristics with quantum hash values, involves the following steps: A quantum Fourier transform is performed on the molecular vibrational spectrum to generate a real-time frequency domain amplitude. The quantum hash value and the real-time frequency domain amplitude are then encoded into a two-qubit entangled state through quantum state encoding. Based on the quantum tunneling coupling strength parameters calibrated by the quantum annealing machine, the quantum tunneling Hamiltonian loaded in the quantum processor is obtained; The entangled state of two qubits and the quantum tunneling Hamiltonian are evolved through the Schrödinger equation to generate an evolving quantum state; Perform Z-basis projection measurements on the evolving quantum state, statistically analyze the joint probability distribution of quantum hash value and frequency domain amplitude, record the frequency of occurrence of the ground state, and generate a joint probability distribution matrix.

9. The intelligent control system for optimizing the pharmaceutical warehousing and sorting process as described in claim 8, characterized in that: The generation of verification confidence refers to calculating the verification confidence by combining the sum of the joint probabilities of the joint probability distribution matrix with the indicator function.

10. The intelligent control system for optimizing the drug warehousing and sorting process as described in claim 9, characterized in that: The process of generating the corrected sorting path involves converting the verification confidence level into a corrected weight coefficient, correcting the chaotic field parameters based on the corrected weight coefficient, and generating the corrected sorting path.

Images