Software and hardware integrated intelligent central control system control method and related device

Abstract

The invention provides a software and hardware integrated intelligent central control system control method and a related device, and belongs to the technical field of control, and the method comprises the steps: firstly, loading a training process template which comprises equipment linkage rules corresponding to different training scenes; in response to the training scene selected by the user, determining a corresponding equipment linkage rule, and generating a hardware reconstruction instruction and an equipment control instruction at the same time; a logic gate circuit in a driver is configured based on the hardware reconstruction instruction, and a target hardware control interface uniquely matched with the training scene is instantiated on site; and finally, outputting a device control instruction to the corresponding training device through the target hardware control interface, so that the plurality of training devices execute an accurate linkage operation according to a preset time sequence. According to the invention, training scene requirements are directly mapped into a hardware circuit structure, on-demand dynamic generation of a control interface is realized, and time sequence precision and real-time performance of multi-device cooperative control are significantly improved.

Description

Technical Field

[0001] This application relates to the field of control technology, and particularly to a control method and related devices for a software-hardware integrated intelligent central control system. Background Art

[0002] In vocational education, industrial training, and college experimental teaching, intelligent central control systems have become the core platforms for integrating and controlling multiple training devices. Typical training scenarios such as automotive engine fault diagnosis, industrial robot offline programming, and aircraft maintenance simulation training often require the central control system to simultaneously drive various types of devices such as simulators, oscilloscopes, signal generators, and actuators, and complete coordinated actions according to strict teaching logic timings.

[0003] Existing central control systems generally adopt a technical architecture of fixed hardware interfaces and software scheduling. The controller main board is pre-wired with a fixed number of general input / output pins, serial communication interfaces, or analog output channels, and all training devices need to adapt to the electrical specifications and communication protocols of these fixed interfaces. When performing multi-device linkage tasks, the controller runs embedded software and sequentially sends instructions to each device through operating system scheduling or interrupt service programs. At the software level, approximate timing control is achieved through timer chains, state machines, or real-time task orchestration.

[0004] However, multi-device linkage requires strict time alignment relationships between control signals. In some training scenarios such as high-voltage system power-off detection, engine fuel injection timing observation, and radar pulse triggering, the timing deviation between multiple signals is required to be controlled within microseconds or even nanoseconds. Limited by operating system interrupt response jitter, bus competition delays, and instruction interpretation and execution overheads, pure software solutions are difficult to meet such high-precision synchronization requirements. Summary of the Invention

[0005] Embodiments of this application provide a control method and related devices for a software-hardware integrated intelligent central control system to improve the above problems.

[0006] To achieve the above object, this application adopts the following technical solutions: In a first aspect, this application proposes a control method for a software-hardware integrated intelligent central control system. The intelligent central control system includes a controller, a driver, and training devices. The method is executed by the controller and includes: Loading a training process template, where the training process template includes device linkage rules corresponding to different training scenarios; Responding to the selected training scenario by the user to determine the corresponding device linkage rules, and generating hardware reconstruction instructions and device control instructions according to the device linkage rules; Configuring the logic gate circuit in the driver based on the hardware reconstruction instructions to generate a target hardware control interface matching the training scenario; Based on the target hardware control interface, output device control commands to the corresponding training devices so that multiple training devices can perform linked operations according to a preset timing sequence.

[0007] In conjunction with the first aspect, optionally, the logic gate circuit includes multiple virtual slots, wherein each virtual slot corresponds to a reconfigurable partition, and each virtual slot corresponds to a training scenario; the method further includes: Receive concurrent selections from users for the first and second training scenarios; Configure a first target hardware control interface in the first virtual slot and configure a second target hardware control interface in the second virtual slot. The first training equipment group is controlled based on the first target hardware control interface, and the second training equipment group is controlled based on the second target hardware control interface.

[0008] In conjunction with the first aspect, the method may optionally also include: Obtain the user's historical operation sequence and input the historical operation sequence into the neural network model. Based on the output of the neural network model, determine the prediction training scenario. Obtain the prediction confidence level. When the prediction confidence level exceeds the preset threshold, before the user confirms the selection, configure the logic gate circuit in the driver to generate a target hardware control interface that matches the prediction training scenario based on the hardware reconfiguration instructions corresponding to the prediction training scenario. Configure candidate hardware control interfaces in available virtual slots; If the user confirms that the selected training scenario is the same as the predicted training scenario, the candidate hardware control interface will be switched to the active state.

[0009] In conjunction with the first aspect, optionally, the user's historical operation sequence is obtained and input into a neural network model. Based on the output of the neural network model, the prediction training scenario is determined, including: Based on the user's wiring operation sequence on the operation panel of the training equipment, the knob adjustment angle curve, the button pressing duration and the equipment response state switching time, a time series feature map is determined. The time series feature map is input into the spatiotemporal attention network, where the spatiotemporal attention network determines the temporal dependency features of the operation event and the spatial association features of the device port through a multi-head attention mechanism. The time-dependent features and spatial correlation features are cross-referenced, and the matching probability distribution between the current operation sequence and each candidate training scenario is output. The candidate training scenario with the highest probability value in the matching probability distribution is determined as the prediction training scenario.

[0010] In conjunction with the first aspect, optionally, a time series feature graph is determined based on the user's wiring operation sequence on the training equipment's control panel, the knob adjustment angle curve, the button press duration, and the device response state switching time, including: Get the terminal numbers and insertion / removal event timestamps corresponding to the wiring operation sequence, get the continuous voltage sampling sequence corresponding to the knob adjustment angle curve, get the pulse width sequence corresponding to the button press duration, and get the response delay time difference corresponding to the device response state switching time. Map the terminal block number to port space coordinate encoding, and align the insertion / removal event timestamp, continuous voltage sampling sequence, pulse width sequence and response delay time difference to the same clock domain to generate a heterogeneous operation event stream. The heterogeneous operation event stream is divided into slice sequences of equal time windows along the time axis, and the dense feature vectors corresponding to different operation events are determined within each time window. Dense feature vectors are stacked in chronological order to form a three-dimensional tensor to construct a time series feature map. The three dimensions of the three-dimensional tensor are the time window index, the operation event category, and the port space coordinates, respectively.

[0011] In conjunction with the first aspect, optionally, the time-series feature map is input into a spatiotemporal attention network, wherein the spatiotemporal attention network determines the temporal dependency features of the operation event and the spatial association features of the device port through a multi-head attention mechanism, including: The time series feature map is divided into multiple time heads along the dimension of the time window index. Each time head determines the corresponding time self-attention, and the time-dependent features are determined based on the output of multiple time heads. The time series feature map is divided into multiple spatial heads along the dimensions of the port spatial coordinates. Each spatial head determines the corresponding spatial self-attention, and spatial correlation features are determined based on the outputs of multiple spatial heads.

[0012] In conjunction with the first aspect, optionally, feature cross-validation can be performed between time-dependent features and spatially related features, and the matching probability distribution between the current operation sequence and each candidate training scenario can be output, including: Based on the cross-attention mechanism, attention is calculated by using time-dependent features as queries and spatial association features as keys to obtain fused features. The matching probability distribution between the current operation sequence and multiple candidate training scenarios is determined based on the fusion features.

[0013] Secondly, this application proposes an integrated hardware and software intelligent central control system, characterized in that the intelligent central control system includes a controller, a driver, and training equipment, and the system is configured as follows: Load the training process template, which includes device linkage rules for different training scenarios; The corresponding device linkage rules are determined in response to the training scenario selected by the user, and hardware reconfiguration instructions and device control instructions are generated according to the device linkage rules; Configure the logic gates in the driver based on hardware reconfiguration instructions to generate a target hardware control interface that matches the training scenario. Based on the target hardware control interface, output device control commands to the corresponding training devices so that multiple training devices can perform linked operations according to a preset timing sequence.

[0014] Optionally, in conjunction with the second aspect, the logic gate circuit includes multiple virtual slots, wherein each virtual slot corresponds to a reconfigurable partition, and each virtual slot corresponds to a training scenario, and the system is configured as follows: Receive concurrent selections from users for the first and second training scenarios; Configure a first target hardware control interface in the first virtual slot and configure a second target hardware control interface in the second virtual slot. The first training equipment group is controlled based on the first target hardware control interface, and the second training equipment group is controlled based on the second target hardware control interface.

[0015] Optionally, in conjunction with the second aspect, the system is configured as follows: Obtain the user's historical operation sequence and input the historical operation sequence into the neural network model. Based on the output of the neural network model, determine the prediction training scenario. Obtain the prediction confidence level. When the prediction confidence level exceeds the preset threshold, before the user confirms the selection, configure the logic gate circuit in the driver to generate a target hardware control interface that matches the prediction training scenario based on the hardware reconfiguration instructions corresponding to the prediction training scenario. Configure candidate hardware control interfaces in available virtual slots; If the user confirms that the selected training scenario is the same as the predicted training scenario, the candidate hardware control interface will be switched to the active state.

[0016] Optionally, in conjunction with the second aspect, the system is configured as follows: Obtain the user's historical operation sequence and input it into a neural network model. Based on the output of the neural network model, determine the prediction training scenario, including: Based on the user's wiring operation sequence on the operation panel of the training equipment, the knob adjustment angle curve, the button pressing duration and the equipment response state switching time, a time series feature map is determined. The time series feature map is input into the spatiotemporal attention network, where the spatiotemporal attention network determines the temporal dependency features of the operation event and the spatial association features of the device port through a multi-head attention mechanism. The time-dependent features and spatial correlation features are cross-referenced, and the matching probability distribution between the current operation sequence and each candidate training scenario is output. The candidate training scenario with the highest probability value in the matching probability distribution is determined as the prediction training scenario.

[0017] Optionally, in conjunction with the second aspect, the system is configured as follows: Based on the user's wiring operation sequence on the training equipment's control panel, the knob adjustment angle curve, the button press duration, and the device's response state switching time, a time series characteristic graph is determined, including: Get the terminal numbers and insertion / removal event timestamps corresponding to the wiring operation sequence, get the continuous voltage sampling sequence corresponding to the knob adjustment angle curve, get the pulse width sequence corresponding to the button press duration, and get the response delay time difference corresponding to the device response state switching time. Map the terminal block number to port space coordinate encoding, and align the insertion / removal event timestamp, continuous voltage sampling sequence, pulse width sequence and response delay time difference to the same clock domain to generate a heterogeneous operation event stream. The heterogeneous operation event stream is divided into slice sequences of equal time windows along the time axis, and the dense feature vectors corresponding to different operation events are determined within each time window. Dense feature vectors are stacked in chronological order to form a three-dimensional tensor to construct a time series feature map. The three dimensions of the three-dimensional tensor are the time window index, the operation event category, and the port space coordinates, respectively.

[0018] Optionally, in conjunction with the second aspect, the system is configured as follows: The time-series feature map is input into a spatiotemporal attention network, which uses a multi-head attention mechanism to determine the temporal dependency features of the operation event and the spatial association features of the device port, including: The time series feature map is divided into multiple time heads along the dimension of the time window index. Each time head determines the corresponding time self-attention, and the time-dependent features are determined based on the output of multiple time heads. The time series feature map is divided into multiple spatial heads along the dimensions of the port spatial coordinates. Each spatial head determines the corresponding spatial self-attention, and spatial correlation features are determined based on the outputs of multiple spatial heads.

[0019] Optionally, in conjunction with the second aspect, the system is configured as follows: The time-dependent features and spatial correlation features are cross-referenced, and the matching probability distribution between the current operation sequence and each candidate training scenario is output, including: Based on the cross-attention mechanism, attention is calculated by using time-dependent features as queries and spatial association features as keys to obtain fused features. The matching probability distribution between the current operation sequence and multiple candidate training scenarios is determined based on the fusion features.

[0020] A third aspect of this invention provides an electronic device, which includes: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method proposed in the first aspect of the present invention.

[0021] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method as described in the first aspect of the present invention.

[0022] In summary, the above method and apparatus have the following technical effects: This invention discloses a control method and related device for an integrated hardware and software intelligent central control system. First, a training process template is loaded, containing device linkage rules corresponding to different training scenarios. Responding to the user-selected training scenario, the corresponding device linkage rules are determined, and hardware reconfiguration instructions and device control instructions are generated simultaneously. Based on the hardware reconfiguration instructions, the logic gate circuits in the driver are configured, and a target hardware control interface uniquely matching the training scenario is instantiated on-site. Finally, the device control instructions are output to the corresponding training devices through the target hardware control interface, enabling multiple training devices to perform precise linkage operations according to a preset timing sequence. This invention directly maps training scenario requirements to hardware circuit structures, realizing on-demand dynamic generation of control interfaces, significantly improving the timing accuracy and real-time performance of multi-device collaborative control. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating a hardware and software integrated intelligent central control system control method proposed in an embodiment of this application. Detailed Implementation

[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] This application proposes a control method for an integrated hardware and software intelligent central control system. The intelligent central control system includes a controller, a driver, and a training device. The method is executed by the controller. Please refer to [link to relevant documentation]. Figure 1 ,include: S101: Load training process template. The training process template includes device linkage rules for different training scenarios.

[0026] Understandably, the controller reads a pre-programmed training process template. This template is a set of customized equipment collaboration specifications for different teaching scenarios; the core component of the training process template is the equipment linkage rules. Taking automotive engine fault diagnosis training as an example, the template defines that when the user presses the "Start Diagnosis" button, the engine simulator must output a specific speed signal within 300 milliseconds, simultaneously the multimeter on the teaching board displays the corresponding voltage value, and the fault simulator illuminates a specific fault light after a 2-second delay. The timing relationships, signal characteristics, and logical dependencies between these devices are all encapsulated in the equipment linkage rules.

[0027] A training process template can include multiple training scenarios. For example, in an industrial robot operation course, a palletizing programming scenario, a welding trajectory teaching scenario, and an emergency stop and reset scenario can be defined separately. In each scenario, the startup sequence, speed curve, and interlocking conditions of the robot body, control cabinet, end effector, safety light curtain, and other equipment are completely different. These differentiated control requirements are described through corresponding equipment linkage rules.

[0028] S102: In response to the training scenario selected by the user, determine the corresponding device linkage rules, and generate hardware reconfiguration instructions and device control instructions according to the device linkage rules.

[0029] Understandably, the controller extracts physical layer requirements for the hardware interface from the rules. For example, the position sensor of an older model might output a non-standard 5-volt pulse signal with a settling time of only 15 nanoseconds. These requirements are compiled separately into a set of hardware reconfiguration instructions. The sole purpose of these instructions is to change the connection configuration of the internal logic gates of the driver, enabling it to generate a physical interface that perfectly matches the current scenario.

[0030] On the other hand, the controller also acquires the control content regarding the equipment's behavioral logic from the rules. For example, it might first send a start signal to the hydraulic pump station, and after the pressure feedback reaches 21 MPa, wait 100 milliseconds before sending an opening command to the actuator solenoid valve. These state transitions and conditional judgments can all be encapsulated as equipment control commands.

[0031] S103: Configure the logic gates in the driver based on hardware reconfiguration instructions to generate a target hardware control interface that matches the training scenario.

[0032] Understandably, the controller sends a hardware reconfiguration instruction to the driver, and the programmable logic gates inside the driver then begin to reconfigure their hardware structure. The specific reconfiguration method and hardware structure of the driver are disclosed in relevant technical documents and are not limited in this application. The hardware reconfiguration instruction carries a description of the interconnection topology of the logic gates, directly changing the physical connection between the logic units inside the driver chip. That is, previously idle lookup tables are written with new Boolean functions, previously straight-through wiring paths are redirected, and previously high-impedance input / output pins are given specific voltage levels.

[0033] For example, consider the training on the maintenance of high-voltage systems in new energy vehicles. When a user selects the "battery module voltage sampling" scenario, the logic gates in the driver, upon receiving a hardware reconfiguration command, generate a set of control interfaces specifically for the high-voltage battery simulator. This set of interfaces includes six parallel sampling start signal pins with 2.5V logic levels, one trigger output pin with a programmable rising edge slope, and four differential analog input channels capable of withstanding 60V common-mode voltage.

[0034] S104: Output device control commands to the corresponding training devices based on the target hardware control interface, so that multiple training devices can perform linkage operations according to a preset timing sequence.

[0035] Understandably, the controller device delivers control commands to the target hardware control interface. These commands are not forwarded one by one as software messages, but are instead written directly into the interface's internal transmit buffer or status register.

[0036] Taking intelligent manufacturing production line debugging training as an example, when the user selects the "Automated Warehouse Stacker Crane Linkage" scenario, the target hardware control interface is reconfigured into a four-way parallel control unit that simultaneously drives the stacker crane's travel motor, fork extension mechanism, laser rangefinder, and storage location status indicator. After the controller issues a "retrieve" command, the hardware timing state machine inside the interface immediately starts: at 0 microseconds, it outputs an acceleration curve pulse sequence to the travel motor; at 320 milliseconds, when the motor reaches the target position, it automatically cuts off the drive current and triggers the mechanical brake, while simultaneously sending an enable signal to the fork extension mechanism; after the fork positioning sensor confirms the position, it delays for 15 milliseconds to light up the storage location indicator and returns a completion status to the host computer. In other words, this series of operations does not require controller intervention.

[0037] Understandably, the entire linkage process is a physical-level synchronization. For example, in hydraulic system training, the interface simultaneously controls the electric pump start contactor, the hydraulic pipeline valve switching, and the pressure sensor sampling trigger. The timing deviation between pressure jumps and valve opening and closing can be controlled within the range of gate-level propagation delay.

[0038] Furthermore, in some implementations, the logic gates may also include multiple virtual slots, each corresponding to a reconfigurable partition and a training scenario. That is, the driver's logic gates do not operate as a single, unified whole, but are further divided into multiple independent reconfigurable regions. Each such region is called a virtual slot. Each virtual slot physically occupies a specific range of logic cells, wiring resources, and input / output pins within the chip. More importantly, each virtual slot has an independent configuration state, allowing it to run completely different hardware logic simultaneously with other slots without interference.

[0039] Each virtual slot is bound to a training scenario. This means that when a training scenario is activated, its corresponding virtual slot is reconstructed into a target hardware control interface dedicated to that scenario; when the scenario ends, the hardware circuitry in the virtual slot can be erased or left silent, but this hardware resource is always reserved for that scenario and will not be preempted by other scenarios.

[0040] Therefore, this embodiment may also include steps S105-S107: S105: Receive the user's concurrent selection of the first training scenario and the second training scenario.

[0041] This is understandable, meaning that the controller receives explicit operational instructions from one or more users for two different training scenarios within the same time window.

[0042] S106: Configure a first target hardware control interface in the first virtual slot and configure a second target hardware control interface in the second virtual slot.

[0043] Understandably, the first virtual slot and the second virtual slot are two pre-defined, isolated reconfigurable regions within the driver. The controller issues two independent sets of hardware reconfiguration instructions to these two virtual slots respectively. The two sets of instructions do not interfere with each other, and the two reconfiguration processes proceed in parallel.

[0044] S107: Control the first training equipment group based on the first target hardware control interface, and control the second training equipment group based on the second target hardware control interface.

[0045] Understandably, at this point, both the first target hardware control interface in the first virtual slot and the second target hardware control interface in the second virtual slot have been reconstructed and each has complete and independent device driver capabilities.

[0046] The controller sends its respective device control commands to two sets of hardware control interfaces. The first target hardware control interface receives control commands for the first training equipment group, and the second target hardware control interface receives control commands for the second training equipment group. The two sets of interfaces operate in parallel, without dependence on or interference between them.

[0047] Furthermore, to improve the system's response speed, this embodiment may also include steps S201-S204: S201: Obtain the user's historical operation sequence and input the historical operation sequence into the neural network model. Based on the output of the neural network model, determine the prediction training scenario.

[0048] Understandably, the controller no longer passively waits for the user to select a training scenario from the menu. Instead, it attempts to infer the user's next move by observing the actions the user has already performed. Specifically, the controller continuously records every subtle action the user takes on the training device. These actions include which terminal the test lead is inserted into and when it is pulled out; the angle and duration of knob rotation when adjusting it; the duration and force sequence of pressing function buttons; and the delay time for changes in device status after each operation. All this information is strung together chronologically to form the current user's historical operation sequence.

[0049] Specifically, step S201 may include the following steps: S2021: Based on the user's wiring operation sequence on the operation panel of the training equipment, the knob adjustment angle curve, the button pressing duration and the equipment response state switching time, determine the time series feature map.

[0050] For example, the terminal numbers and plug-in / plug-out event timestamps corresponding to the wiring operation sequence can be obtained, the continuous voltage sampling sequence corresponding to the knob adjustment angle curve can be obtained, the pulse width sequence corresponding to the button press duration can be obtained, and the response delay time difference corresponding to the device response state switching time can be obtained.

[0051] Then, the terminal block numbers are mapped to port space coordinate codes, and the insertion / removal event timestamps, continuous voltage sampling sequences, pulse width sequences, and response delay time differences are aligned to the same clock domain to generate a heterogeneous operation event stream. The heterogeneous operation event stream is divided into slice sequences of equal time windows along the time axis, and the dense feature vectors corresponding to different operation events within each time window are determined. The dense feature vectors are then stacked in chronological order into a three-dimensional tensor to form a time series feature map, where the three dimensions of the three-dimensional tensor are the time window index, the operation event category, and the port space coordinates, respectively.

[0052] Specifically, for wiring operations, the precise time of each plugging / unplugging event and the terminal numbers involved are recorded. For knob adjustments, the voltage curve of the knob angle changing over time is continuously recorded at a high sampling rate. For button operations, the start and end times of each press are recorded, and the pulse width is calculated. For device response, the time delay between user operation and device state transition is recorded.

[0053] Taking the diagnostic training of an automotive engine electronic control unit as an example. The user inserts an oscilloscope probe into the throttle position sensor signal terminal, and the controller records the terminal number C30 and the insertion time of 0 seconds; the user rotates the knob clockwise to adjust the simulated throttle opening, and the controller records a continuous sampling sequence in which the voltage linearly rises from 0.8 volts to 4.2 volts within 0 to 1.8 seconds; the user presses the fault code reading button and holds it for 2.3 seconds, and the controller records a pulse sequence with a pulse width of 2300 milliseconds; the device illuminates the fault indicator light 340 milliseconds after the button is released, and the controller records the response delay time difference of 340 milliseconds.

[0054] Since terminal block numbers are simply meaningless numerical codes and cannot directly reflect the physical positional relationship between terminals, the controller maps them to port space coordinate codes. For example, it maps the physical layout of the engine control unit junction box to a two-dimensional grid coordinate system. Adjacent terminals also maintain an adjacent relationship in spatial coordinates.

[0055] Understandably, the original time bases for the four types of data are different: plug-in / plug-out events are discrete time points, knob curves are continuous sampling sequences, button pulses are interval events, and response delays are time differences. The controller aligns all time information to the same system clock domain and interweaves them according to the actual time sequence to generate a complete heterogeneous operation event stream. This event stream is a timeline, and each segment may be marked with wiring plug-in / plug-out, knob angle sampling values, button state changes, or device response flags.

[0056] Furthermore, to input the data into the neural network, the controller can divide the entire timeline into a series of time windows of equal length. Within each time window, all events occurring within the window are compressed into a dense feature vector.

[0057] For example, within a time window of 1 to 1.5 seconds, two wiring operations occurred; the average knob angle was 2.1 volts, with a rate of change of 1.2 volts per second; the total button press duration was 0 milliseconds; and the device responded to one event with a delay of 340 milliseconds. These values ​​were arranged into a fixed-length feature vector.

[0058] This process is repeated, producing a feature vector for each time window. Finally, the feature vectors from all time windows are stacked in chronological order to form a three-dimensional tensor. The first dimension of this tensor is the sequence number of the time window, the second dimension is the category of the operation event, and the third dimension is the spatial coordinates of the terminal block.

[0059] S2022: Input the time series feature map into the spatiotemporal attention network, where the spatiotemporal attention network determines the temporal dependency features of the operation event and the spatial association features of the device port through a multi-head attention mechanism.

[0060] Specifically, the time series feature map can be divided into multiple time heads along the dimension of the time window index. Each time head determines a corresponding temporal self-attention, and temporal dependency features are determined based on the outputs of multiple time heads. Similarly, the time series feature map can be divided into multiple spatial heads along the dimension of the port spatial coordinates. Each spatial head determines a corresponding spatial self-attention, and spatial association features are determined based on the outputs of multiple spatial heads.

[0061] Understandably, spatiotemporal attention networks can separate two types of patterns from feature maps: one is the dependency relationship of operation events over time, and the other is the collaborative association of device ports in physical layout.

[0062] Specifically, the 3D feature map is first split along the dimension of the time window index. Assuming the feature map contains forty consecutive time windows, the network treats these forty windows as forty independent time heads. Each time head independently focuses on the sequence of operational events within a specific segment on the entire timeline, calculating the attention weights between different moments within that segment. That is, each time head captures local temporal dependency patterns from its own perspective. Finally, the network stitches together and merges the local patterns captured by all time heads to form a global temporal dependency feature. This feature can characterize whether the current user's operation rhythm is rapid or steady, which operational habits are related, and whether the device response latency conforms to the expected pattern.

[0063] Similarly, the network performs another set of splits along the dimensions of the port spatial coordinates. The third dimension of the feature map records the spatial location of the terminals where each operation event occurs. The network divides these spatial coordinate points into multiple spatial heads, each responsible for focusing on a specific group of ports within a particular area of ​​the panel. Taking an automotive engine diagnostic panel as an example, spatial head A focuses on the sensor signal terminal cluster in the upper left corner, calculating the self-attention weights between terminals C30, C31, and C32. It finds that whenever the user measures the throttle position sensor, they always immediately measure the intake pressure sensor. Spatial head B focuses on the actuator drive terminal cluster in the lower right corner, noting that the fuel injector drive terminal and the fuel pump control terminal are rarely operated simultaneously. Each spatial head independently learns the port co-occurrence patterns and spatial semantic associations within its respective area. By splicing and fusing the outputs of all spatial heads, a complete spatial association feature is obtained. This feature represents which physical locations the user tends to operate simultaneously, and which ports constitute logical functional groups in the teaching task.

[0064] S2023: Perform feature cross-referencing between time-dependent features and spatially related features, and output the matching probability distribution between the current operation sequence and each candidate training scenario.

[0065] S2024: The candidate training scenario with the highest probability value in the matching probability distribution is determined as the prediction training scenario.

[0066] Specifically, based on the cross-attention mechanism, time-dependent features are used as queries and spatial association features are used as keys for attention calculation to obtain fused features. Based on the fused features, the matching probability distribution between the current operation sequence and multiple candidate training scenarios is determined.

[0067] Understandably, time-dependent features are used as queries, and spatial association features are used as keys. In other words, the user's operation rhythm is used to activate the most relevant port spatial layout.

[0068] Taking automotive engine electronic control unit (ECU) diagnostic training as an example, the user's time-dependent characteristics show that they habitually read fault codes within 300 milliseconds after completing a throttle opening adjustment, and the duration of each read button press is 2.3 seconds. This is a highly stable operating rhythm. Spatial correlation characteristics show that the ports frequently operated by the user are concentrated at three spatial points: the throttle position sensor signal terminal, the K-line of the fault diagnosis interface, and the engine control unit ground wire. These three terminals are arranged in a triangular pattern on the panel's physical layout.

[0069] In the example, whenever the time pattern "short press the read button 300 milliseconds after adjusting the knob" appears, the strongly associated spatial pattern is always the specific triangular area of ​​"throttle position sensor terminal, diagnostic K-line, and ground wire". Meanwhile, port areas that do not match the current time rhythm, such as the oxygen sensor heating control terminal and the fuel pump relay drive terminal, although also marked as frequently used areas in the spatial characteristics, have their attention weight significantly reduced.

[0070] In this matching method, the generated fused features no longer distinguish between time and space, and are fed into a fully connected classification layer. The classification layer compares the current operation sequence with each candidate training scenario in the database. For example, it calculates the cosine similarity between the fused features and the feature templates of the engine throttle position sensor calibration scenario, the oxygen sensor heating circuit diagnostic scenario, and the fuel pump relay control circuit measurement scenario. The specific computational process of the neural network has been disclosed in relevant technical documents and will not be limited here.

[0071] Each understandable similarity calculation outputs a probability value between 0 and 1. All these probability values ​​form a probability distribution across all candidate training scenarios. The scenario with the highest sum of probabilities (1) is the training scenario the network determines is most likely being performed or about to be selected by the current user.

[0072] S202: Obtain the prediction confidence level. When the prediction confidence level exceeds the preset threshold, before the user confirms the selection, configure the logic gate circuit in the driver to generate a target hardware control interface that matches the prediction training scenario based on the hardware reconfiguration instruction corresponding to the prediction training scenario.

[0073] Understandably, the confidence score reflects the model's degree of certainty about its current inference. A higher score indicates a closer match between the user's current operational pattern and the typical characteristics of a training scenario, thus reducing the likelihood of prediction errors. The controller presets a threshold, which can be a balance point calibrated using real teaching data.

[0074] When the prediction confidence exceeds a preset threshold, the controller determines that the prediction is highly reliable. At this point, the user has not yet clicked to confirm any scenario on the interface. The controller immediately retrieves the hardware reconfiguration instruction uniquely bound to this prediction training scenario from the storage medium. This instruction, which should normally be issued only after the user clicks to confirm, is now activated in advance. The instruction is transmitted to the driver, and the logic gates in the driver then begin reconfiguration.

[0075] The entire pre-configuration process is completed quickly. Through this step, the controller reduces the time between user intent and hardware response.

[0076] S203: Configure a candidate hardware control interface in an unused virtual slot.

[0077] Understandably, a virtual slot is a pre-defined reconfigurable partition within a logic gate circuit. Some slots are currently serving the scenario, used for the running hardware control interface; others are unoccupied, in an unconfigured or erased idle state. The controller scans the resource occupancy table of all virtual slots and selects a currently completely idle slot.

[0078] After selecting an available virtual slot, the controller issues a hardware reconfiguration command to the reconfigurable partition corresponding to that slot. The driver immediately initiates the configuration process for the logic gates within that partition.

[0079] S204: If the user confirms that the selected training scenario is the same as the predicted training scenario, then the candidate hardware control interface will be switched to the active state.

[0080] Understandably, if the training scenario is the same as the predicted training scenario, the controller sends an activation command to the virtual slot hosting the candidate hardware control interface. If the scenario confirmed by the user is different from the predicted scenario, the candidate hardware control interface will not be activated.

[0081] This invention discloses a control method for an integrated hardware and software intelligent central control system. First, a training process template is loaded, containing device linkage rules corresponding to different training scenarios. Responding to the user-selected training scenario, the corresponding device linkage rules are determined, and hardware reconfiguration instructions and device control instructions are generated simultaneously. Based on the hardware reconfiguration instructions, the logic gate circuits in the driver are configured, and a target hardware control interface uniquely matching the training scenario is instantiated on-site. Finally, the device control instructions are output to the corresponding training devices through the target hardware control interface, enabling multiple training devices to perform precise linkage operations according to a preset timing sequence. This invention directly maps training scenario requirements to hardware circuit structures, realizing on-demand dynamic generation of control interfaces, significantly improving the timing accuracy and real-time performance of multi-device collaborative control.

[0082] Based on the same inventive concept, this application also proposes an integrated hardware and software intelligent central control system, which includes a controller, a driver, and training equipment. The system is configured as follows: Load the training process template, which includes device linkage rules for different training scenarios; The corresponding device linkage rules are determined in response to the training scenario selected by the user, and hardware reconfiguration instructions and device control instructions are generated according to the device linkage rules; Configure the logic gates in the driver based on hardware reconfiguration instructions to generate a target hardware control interface that matches the training scenario. Based on the target hardware control interface, output device control commands to the corresponding training devices so that multiple training devices can perform linked operations according to a preset timing sequence.

[0083] Optionally, the logic gate circuit includes multiple virtual slots, wherein each virtual slot corresponds to a reconfigurable partition, and each virtual slot corresponds to a training scenario, and the system is configured as follows: Receive concurrent selections from users for the first and second training scenarios; Configure a first target hardware control interface in the first virtual slot and configure a second target hardware control interface in the second virtual slot. The first training equipment group is controlled based on the first target hardware control interface, and the second training equipment group is controlled based on the second target hardware control interface.

[0084] Optionally, the system is configured as follows: Obtain the user's historical operation sequence and input the historical operation sequence into the neural network model. Based on the output of the neural network model, determine the prediction training scenario. Obtain the prediction confidence level. When the prediction confidence level exceeds the preset threshold, before the user confirms the selection, configure the logic gate circuit in the driver to generate a target hardware control interface that matches the prediction training scenario based on the hardware reconfiguration instructions corresponding to the prediction training scenario. Configure candidate hardware control interfaces in available virtual slots; If the user confirms that the selected training scenario is the same as the predicted training scenario, the candidate hardware control interface will be switched to the active state.

[0085] Optionally, the system is configured as follows: Obtain the user's historical operation sequence and input it into a neural network model. Based on the output of the neural network model, determine the prediction training scenario, including: Based on the user's wiring operation sequence on the operation panel of the training equipment, the knob adjustment angle curve, the button pressing duration and the equipment response state switching time, a time series feature map is determined. The time series feature map is input into the spatiotemporal attention network, where the spatiotemporal attention network determines the temporal dependency features of the operation event and the spatial association features of the device port through a multi-head attention mechanism. The time-dependent features and spatial correlation features are cross-referenced, and the matching probability distribution between the current operation sequence and each candidate training scenario is output. The candidate training scenario with the highest probability value in the matching probability distribution is determined as the prediction training scenario.

[0086] Optionally, the system is configured as follows: Based on the user's wiring operation sequence on the training equipment's control panel, the knob adjustment angle curve, the button press duration, and the device's response state switching time, a time series characteristic graph is determined, including: Get the terminal numbers and insertion / removal event timestamps corresponding to the wiring operation sequence, get the continuous voltage sampling sequence corresponding to the knob adjustment angle curve, get the pulse width sequence corresponding to the button press duration, and get the response delay time difference corresponding to the device response state switching time. Map the terminal block number to port space coordinate encoding, and align the insertion / removal event timestamp, continuous voltage sampling sequence, pulse width sequence and response delay time difference to the same clock domain to generate a heterogeneous operation event stream. The heterogeneous operation event stream is divided into slice sequences of equal time windows along the time axis, and the dense feature vectors corresponding to different operation events are determined within each time window. Dense feature vectors are stacked in chronological order to form a three-dimensional tensor to construct a time series feature map. The three dimensions of the three-dimensional tensor are the time window index, the operation event category, and the port space coordinates, respectively.

[0087] Optionally, the system is configured as follows: The time-series feature map is input into a spatiotemporal attention network, which uses a multi-head attention mechanism to determine the temporal dependency features of the operation event and the spatial association features of the device port, including: The time series feature map is divided into multiple time heads along the dimension of the time window index. Each time head determines the corresponding time self-attention, and the time-dependent features are determined based on the output of multiple time heads. The time series feature map is divided into multiple spatial heads along the dimensions of the port spatial coordinates. Each spatial head determines the corresponding spatial self-attention, and spatial correlation features are determined based on the outputs of multiple spatial heads.

[0088] Optionally, the system is configured as follows: The time-dependent features and spatial correlation features are cross-referenced, and the matching probability distribution between the current operation sequence and each candidate training scenario is output, including: Based on the cross-attention mechanism, attention is calculated by using time-dependent features as queries and spatial association features as keys to obtain fused features. The matching probability distribution between the current operation sequence and multiple candidate training scenarios is determined based on the fusion features.

[0089] Based on the same inventive concept, embodiments of this application also propose an electronic device, which includes: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to execute a hardware-software integrated intelligent central control system control method according to an embodiment of this application.

[0090] Furthermore, to achieve the above objectives, embodiments of this application also propose a computer-readable storage medium storing a computer program, which, when executed by a processor, implements a hardware-software integrated intelligent central control system control method according to embodiments of this application.

[0091] The following is a detailed introduction to the various components of the electronic device: In this context, the processor is the control center of the electronic device. It can be a single processor or a collective term for multiple processing elements. For example, a processor can be one or more central processing units (CPUs), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention, such as one or more digital signal processors (DSPs), or one or more field-programmable gate arrays (FPGAs).

[0092] Alternatively, the processor can perform various functions of the electronic device by running or executing software programs stored in memory, and by calling data stored in memory.

[0093] The memory is used to store the software program that executes the solution of the present invention, and the execution is controlled by the processor. The specific implementation method can be referred to the above method embodiment, and will not be repeated here.

[0094] Optionally, the memory can be read-only memory (ROM) or other types of static storage devices capable of storing static information and instructions, random access memory (RAM) or other types of dynamic storage devices capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. The memory can be integrated with the processor or exist independently and coupled to the processor through an interface circuit of an electronic device; the embodiments of the present invention do not specifically limit this.

[0095] A transceiver is used to communicate with network devices or with terminal devices.

[0096] Optionally, the transceiver may include a receiver and a transmitter. The receiver is used to implement the receiving function, and the transmitter is used to implement the sending function.

[0097] Optionally, the transceiver can be integrated with the processor or exist independently and coupled to the processor through the router's interface circuit. This embodiment of the invention does not specifically limit this.

[0098] Furthermore, the technical effects of the electronic device can be referred to the technical effects of the data transmission method in the above method embodiments, and will not be repeated here.

[0099] It should be understood that the processor in the embodiments of the present invention can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0100] It should also be understood that the memory in the embodiments of the present invention can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous DRAM (DDRSDRAM), enhanced synchronous DRAM (ESDRAM), synchronous linked DRAM (SLDRAM), and direct rambus RAM (DRRAM).

[0101] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the flow or function according to the embodiments of the present invention is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. A computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. Semiconductor media can be solid-state drives.

[0102] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0103] In this invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be a single item or multiple items.

[0104] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0105] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

Claims

1. A control method for an integrated hardware and software intelligent central control system, characterized in that, The intelligent central control system includes a controller, a driver, and a training device. The method is executed by the controller and includes: Load the training process template, which includes device linkage rules corresponding to different training scenarios; In response to the training scenario selected by the user, the corresponding device linkage rules are determined, and hardware reconfiguration instructions and device control instructions are generated according to the device linkage rules; Configure the logic gate circuits in the driver based on the hardware reconfiguration instructions to generate a target hardware control interface that matches the training scenario. Based on the target hardware control interface, the device control command is output to the corresponding training device, so that multiple training devices perform linkage operations according to a preset timing sequence.

2. The control method for an integrated hardware and software intelligent central control system according to claim 1, characterized in that, The logic gate circuit includes multiple virtual slots, wherein each virtual slot corresponds to a reconfigurable partition, and each virtual slot corresponds to a training scenario. The method further includes: Receive concurrent selections from users for the first and second training scenarios; Configure a first target hardware control interface in the first virtual slot and configure a second target hardware control interface in the second virtual slot. The first training equipment group is controlled based on the first target hardware control interface, and the second training equipment group is controlled based on the second target hardware control interface.

3. The control method for an integrated hardware and software intelligent central control system according to claim 2, characterized in that, The method further includes: Obtain the user's historical operation sequence and input the historical operation sequence into a neural network model. Based on the output of the neural network model, determine the prediction training scenario. Obtain the prediction confidence level. When the prediction confidence level exceeds a preset threshold, before the user confirms the selection, configure the logic gate circuit in the driver based on the hardware reconfiguration instruction corresponding to the prediction training scenario to generate the target hardware control interface that matches the prediction training scenario. Configure a candidate hardware control interface in the idle virtual slot; If the user confirms that the selected training scenario is the same as the predicted training scenario, then the candidate hardware control interface is switched to the active state.

4. The control method for an integrated hardware and software intelligent central control system according to claim 3, characterized in that, Obtain the user's historical operation sequence and input the historical operation sequence into a neural network model. Based on the output of the neural network model, determine the prediction training scenario, including: Based on the user's wiring operation sequence, knob adjustment angle curve, button pressing duration and device response state switching time on the operation panel of the training equipment, a time series feature graph is determined. The time series feature map is input into a spatiotemporal attention network, wherein the spatiotemporal attention network determines the temporal dependency features of the operation event and the spatial association features of the device port through a multi-head attention mechanism. The time-dependent features and spatial association features are cross-referenced, and the matching probability distribution between the current operation sequence and each candidate training scenario is output. The candidate training scenario with the highest probability value in the matching probability distribution is determined as the predicted training scenario.

5. The control method for an integrated hardware and software intelligent central control system according to claim 4, characterized in that, Based on the user's wiring operation sequence, knob adjustment angle curve, button press duration, and device response state switching time on the operation panel of the training equipment, a time series feature graph is determined, including: Obtain the terminal number and insertion / removal event timestamp corresponding to the wiring operation sequence, obtain the continuous voltage sampling sequence corresponding to the knob adjustment angle curve, obtain the pulse width sequence corresponding to the button pressing duration, and obtain the response delay time difference corresponding to the device response state switching time. The terminal block number is mapped to port space coordinate encoding, and the insertion / removal event timestamp, the continuous voltage sampling sequence, the pulse width sequence, and the response delay time difference are uniformly aligned to the same clock domain to generate a heterogeneous operation event stream; The heterogeneous operation event stream is divided into slice sequences of equal time windows along the time axis, and the dense feature vectors corresponding to different operation events are determined within each time window. The dense feature vectors are stacked in chronological order to form a three-dimensional tensor to construct the time series feature map, wherein the three dimensions of the three-dimensional tensor are the time window index, the operation event category, and the port space coordinates, respectively.

6. The control method for an integrated hardware and software intelligent central control system according to claim 5, characterized in that, The time-series feature map is input to a spatiotemporal attention network, wherein the spatiotemporal attention network determines the temporal dependency features of the operation event and the spatial association features of the device port through a multi-head attention mechanism, including: The time series feature map is divided into multiple time heads along the dimension of the time window index, wherein each time head determines a corresponding time self-attention, and the time dependency feature is determined based on the output of the multiple time heads; The time series feature map is divided into multiple spatial heads along the dimensions of the port spatial coordinates. Each spatial head determines a corresponding spatial self-attention, and the spatial association feature is determined based on the output of the multiple spatial heads.

7. The control method for an integrated hardware and software intelligent central control system according to claim 6, characterized in that, The time-dependent features and spatial association features are cross-referenced, and the matching probability distribution between the current operation sequence and each candidate training scenario is output, including: Based on the cross-attention mechanism, the time-dependent features are used as queries and the spatial association features are used as keys for attention calculation to obtain fused features; Based on the fusion features, the matching probability distribution between the current operation sequence and multiple candidate training scenarios is determined.

8. A hardware and software integrated intelligent central control system, characterized in that, The intelligent central control system includes a controller, a driver, and training equipment, and the system is configured as follows: Load the training process template, which includes device linkage rules corresponding to different training scenarios; In response to the training scenario selected by the user, the corresponding device linkage rules are determined, and hardware reconfiguration instructions and device control instructions are generated according to the device linkage rules; Configure the logic gate circuits in the driver based on the hardware reconfiguration instructions to generate a target hardware control interface that matches the training scenario. Based on the target hardware control interface, the device control command is output to the corresponding training device, so that multiple training devices perform linkage operations according to a preset timing sequence.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; And, a memory communicatively connected to at least one of the processors; The memory stores instructions that can be executed by at least one of the processors, and the instructions are executed by at least one of the processors to enable at least one of the processors to execute the hardware-software integrated intelligent central control system control method as claimed in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, It stores a computer program, which, when executed by a processor, implements a hardware-software integrated intelligent central control system control method as proposed in any one of claims 1-7.

Images