Space station simulation interactive display system applied to space class

Abstract

The invention discloses a space station simulation interaction display system applied to a space classroom, and the system comprises a scenarized entity module which is used for constructing a space station function partition model which is scaled according to a proportion, and forming a display carrier with a function interaction attribute; the dynamic projection module is used for generating a teaching content picture adaptive to the model structure and generating dynamic projection content in functional linkage with the entity module; the behavior sensing module is used for capturing audience interaction behaviors and generating targeted interaction response instructions; and the multi-dimensional linkage module is used for coordinating all the parts to act cooperatively, controlling the lifting platform to drive the simulation astronaut model to move to the corresponding function partition based on the interaction response instruction, synchronously driving the mechanical arm to simulate operation demonstration, and generating a scenarized space classroom interaction display effect. According to the embodiment of the invention, time limitation can be broken through, multi-component cooperation can be realized, and space station simulation interaction display accurately responding to audience intentions can be realized.

Description

Technical Field

[0001] This invention belongs to the field of simulation and interactive technology, and in particular, it is a space station simulation and interactive display system applied to space classrooms. Background Technology

[0002] With the promotion of space science education brands such as "Tiangong Classroom," the demand for popularizing knowledge about the space station is increasing. Existing space classrooms mostly rely on astronauts to teach in orbit, which is significantly limited by mission time and resources, making it difficult to achieve regular demonstrations. Moreover, existing simulation systems are mostly static models or single video playbacks, lacking scenario-based physical interaction, and cannot allow the audience to intuitively experience the functional divisions and operational logic of the space station.

[0003] Furthermore, the existing system suffers from sluggish interactive response: it can only passively play preset content and cannot dynamically push in-depth knowledge based on the audience's focus; the coordination between various display components (such as models, projections, and mechanical structures) is poor, the projected images are easily misaligned with the physical models, and the mechanical movements are disconnected from the teaching content, resulting in fragmented science popularization effects. At the same time, the system's audience is limited to teenagers, lacking suitable knowledge depth and interactive forms for adult groups, and its social acceptance needs to be improved. Summary of the Invention

[0004] The purpose of this invention is to provide a space station simulation interactive display system for use in space classrooms, in order to overcome the shortcomings of existing technologies and achieve space station simulation interactive display that breaks through time limitations, enables multi-component collaboration, and accurately responds to the audience's intentions.

[0005] One embodiment of this application provides a space station simulation interactive display system for use in a space classroom. The system includes: a scene-based entity module, a dynamic projection module, a behavior perception module, and a multi-dimensional linkage module. The scenario-based entity module is used to construct a scaled-down model of the space station's functional zones. The model's interior is equipped with simulation operation components corresponding to the teaching content. The node module is equipped with a simulated astronaut model carrying a lifting platform, and the top is fitted with a robotic arm that can simulate equipment handling, forming a display carrier with functional interactive attributes. The dynamic projection module is used to generate teaching content images that adapt to the model structure. It uses dual projectors to project the image crosswise and employs a scene anchor point matching algorithm to align the projected image with the model's functional areas in real time. When switching teaching topics, it automatically adjusts the image scaling ratio and content focus to generate dynamic projection content that is linked to the functions of the physical module. The behavior perception module is used to capture audience interaction behavior. It identifies the audience's standing area and gaze direction through a microwave radar array, and judges the interaction intention based on the duration. When the audience focuses on a certain functional area for more than a preset time, it triggers the display of in-depth teaching content corresponding to that area and generates targeted interactive response instructions. The multi-dimensional linkage module is used to coordinate the collaborative actions of various components. Based on interactive response commands, it controls the lifting platform to move the simulated astronaut model to the corresponding functional area, and synchronously drives the robotic arm to simulate operation demonstration. This allows the state changes and mechanical actions of the simulated operation components to match the dynamic projection content in real time, generating a scenario-based interactive display effect for the space classroom.

[0006] Optionally, the scenario-based entity module is specifically used for: Based on the actual structural parameters of the space station, the core functional areas are divided according to a preset ratio, the size, relative position and connection method of each area are determined, and the structural blueprint of the area is output. Create a solid model framework based on the partition structure blueprint, reserve installation interfaces for simulation operation components at corresponding positions inside the model, and output a model framework with interfaces. Install simulation operation components: assemble a triggerable display plant cultivation chamber at the experimental chamber interface, install a force feedback fixed handle at the living chamber interface, and output the model of the basic assembly components. A lifting platform carrying a simulated astronaut model is installed in the node module. A multi-jointed robotic arm is assembled on top of the model. All components are integrated and the interactive functions are tested to form a display carrier with functional interactive attributes.

[0007] Optionally, the dynamic projection module is specifically used for: Based on the teaching focus of each functional area of ​​the space station, a multi-theme teaching content library was created, with each theme content associated with a corresponding area tag, and a structured teaching content library was output. Install the two projectors at the preset positions on the outside of the model, adjust the projection angle so that the image overlaps and covers the surface of the model, calibrate the brightness and color consistency of the two projectors using a grayscale test card, and output the calibrated dual projector system. Physical anchor points are set at the edges of each functional partition of the model. The scene anchor point matching algorithm is used to identify the anchor point position, calculate the geometric deviation between the projected image and the partition, and adjust the output parameters of the projector in real time to make the image and the partition outline accurately aligned, and output the aligned projected image. When a theme switching command is received, the system automatically calls up the teaching content of the corresponding theme, adjusts the screen scaling ratio according to the target partition size, and updates the anchor point matching parameters in sync to maintain alignment, generating dynamic projection content that is linked with the functions of the physical module.

[0008] Optionally, the behavior-aware module is specifically used for: A microwave radar array is distributed in a fan shape around the model, dividing it into multiple monitoring areas. The radar sampling frequency is configured to capture the positional movement and micro-movement signals of the audience, and output the completed radar perception system. The radar sensing system collects signals in real time, identifies the standing area of ​​the audience through signal strength analysis, determines the gaze direction by combining the signal Doppler effect, and outputs raw data of audience behavior. Perform time-series analysis on raw audience behavior data to calculate the dwell time and gaze duration of an audience in a certain functional area. When the duration exceeds a preset value, it is determined that there is a deep interaction intention in that area, and the intention judgment result is output. Based on the intent judgment result, the corresponding in-depth teaching content of the functional area is triggered, and an interactive response instruction containing area identifier and content type is generated.

[0009] Optionally, the multi-dimensional linkage module is specifically used for: Receive interactive response commands, parse the target functional area location in the commands, generate the lifting platform movement path and the robotic arm operation task, and output motion control parameters; Based on motion control parameters, the lifting platform is driven to move the simulated astronaut model along the path. The current position is fed back in real time through position sensors. When the target area is reached, a stop signal is triggered and an astronaut arrival signal is output. After receiving the astronaut's arrival signal, the robotic arm is synchronously driven to move along a preset trajectory. The joint encoder ensures the accuracy of the movement, simulates the equipment handling or operation process, and outputs a signal that the robotic arm has completed its movement. By integrating the astronaut arrival signal and the robotic arm action completion signal, the system triggers the state changes of the corresponding simulated operation components and sends synchronous commands to the dynamic projection module, enabling the projected content to match the physical actions in real time and generating a scenario-based interactive display effect for the space classroom.

[0010] Another embodiment of this application provides a method for interactive display of a space station simulation in a space classroom, the method comprising: Construct a scaled-down model of the space station's functional zones, with simulation operation components corresponding to the teaching content set up inside the model. The node module is equipped with a simulated astronaut model carrying a lifting platform, and the top is fitted with a robotic arm that can simulate equipment handling, forming a display carrier with functional interactive attributes. The system generates teaching content images that are adapted to the model structure. It uses dual projectors to project the images crosswise and employs a scene anchor point matching algorithm to align the projected images with the model's functional areas in real time. When switching teaching topics, it automatically adjusts the image scaling and content focus to generate dynamic projection content that is linked to the functions of the physical modules. Capture audience interaction behavior, identify the audience's standing area and gaze direction through a microwave radar array, and judge the interaction intention based on the duration. When the audience focuses on a certain functional area for more than a preset time, trigger the display of the corresponding in-depth teaching content in that area and generate targeted interactive response instructions. By coordinating the coordinated actions of various components and controlling the lifting platform to move the simulated astronaut model to the corresponding functional area based on interactive response commands, the robot arm is simultaneously driven to simulate operation demonstrations. This allows the state changes and mechanical movements of the simulated operation components to match the dynamic projection content in real time, generating a scenario-based interactive display effect for the space classroom.

[0011] Another embodiment of this application provides a storage medium storing a computer program, wherein the computer program is configured to execute the method described in any of the preceding claims when running.

[0012] Another embodiment of this application provides an electronic device including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the method described in any of the preceding claims.

[0013] Compared with existing technologies, this invention provides a space station simulation interactive display system for use in space classrooms. The system includes: a scenario-based entity module for constructing a scaled-down model of the functional areas of a space station, forming a display carrier with functional interactive attributes; a dynamic projection module for generating teaching content images adapted to the model structure, generating dynamic projection content that is linked to the functions of the entity module; a behavior perception module for capturing audience interaction behavior, identifying the audience's standing area and gaze direction through a microwave radar array, and generating targeted interactive response commands; and a multi-dimensional linkage module for coordinating the collaborative actions of various components, controlling the lifting platform to move the simulated astronaut model to the corresponding functional area based on the interactive response commands, and synchronously driving the robotic arm to simulate operation demonstrations, generating scenario-based space classroom interactive display effects. This enables a space station simulation interactive display that breaks through time limitations, achieves multi-component collaboration, and accurately responds to audience intentions. Attached Figure Description

[0014] Figure 1 A structural block diagram of a space station simulation interactive display system for use in a space classroom, provided by an embodiment of the present invention; Figure 2 A flowchart illustrating a space station simulation interactive display method for use in a space classroom, provided as an embodiment of the present invention; Figure 3 This is a hardware structure block diagram of a computer terminal for a space station simulation interactive display method applied to a space classroom, as provided in an embodiment of the present invention. Detailed Implementation

[0015] The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0016] See Figure 1The present invention provides a space station simulation interactive display system for use in a space classroom. The system includes: a scene-based entity module 101, a dynamic projection module 102, a behavior perception module 102, and a multi-dimensional linkage module 103. The scenario-based entity module is used to construct a scaled-down model of the space station's functional zones. The model's interior features simulation operation components corresponding to the teaching content. The node module houses a simulated astronaut model with a lifting platform, and the top is equipped with a robotic arm capable of simulating equipment handling, forming a display platform with interactive functional attributes. Specifically, the scenario-based entity module is used for: Based on the actual structural parameters of the space station, the core functional areas are divided according to a preset ratio, the size, relative position and connection method of each area are determined, and the structural blueprint of the area is output. The core of the scenario-based physical module is to restore the functional logic of the space station. It needs to be based on the actual space station structure and ensure the consistency between the model and the actual functional areas through scaling, so as to lay the foundation for subsequent component assembly and interaction design.

[0017] Obtaining actual structural parameters of the space station: Referencing the publicly available technical parameters of the Chinese space station's "Tianhe" core module, "Wentian" experimental module, and "Mengtian" experimental module, the core parameters include: Core module: 16.6 meters in total length and 4.2 meters in maximum diameter, including node module (4.2 meters in diameter and 3.0 meters in length), living module (5.0 meters in length) and resource module (8.6 meters in length). Experimental Module: The "Wentian" module is 17.9 meters long and 4.2 meters in diameter, including a scientific experiment area (8.0 meters long) and a plant cultivation area (2.0 meters long). Key interfaces: The "cone-rod" docking mechanism is used between each section, with an interface diameter of 0.8 meters and a docking gap of ≤2mm.

[0018] Preset scale and partitioning: Considering the space limitations of the exhibition setting (such as a science museum or school exhibition hall), the preset scale is 1:10 (the model size is moderate, ensuring both detailed reproduction and ease of operation for visitors). Based on this, the dimensions of each partition of the model are calculated as follows: Node module model: 42cm in diameter (4.2m ÷ 10), 30cm in length (3.0m ÷ 10), serving as the central hub of the model and connecting other sections; Experimental chamber model (including plant cultivation function): 179cm long (17.9m ÷ 10), 42cm in diameter, with the first 20cm designated as the plant cultivation chamber section and the last 159cm as the general experimental section; Living cabin model: 50cm long (5.0 m ÷ 10), 42cm in diameter, including a fixed handle installation area (15cm long) and a living scene display area (35cm long). Storage compartment model (auxiliary partition): 20cm long and 30cm in diameter, used to demonstrate the space station's material storage function.

[0019] Relative position and connection method determined: Relative positions: Centered on the node cabin model (coordinate origin (0,0,0)), the experimental cabin model is arranged along the positive X-axis (the axis coincides with the node cabin axis, with a spacing of 5cm), the living cabin model is arranged along the negative X-axis (with a spacing of 5cm), and the storage cabin model is arranged along the positive Y-axis (connected to the side of the node cabin, with a spacing of 3cm). A 3D layout diagram is drawn using CAD software (such as AutoCAD 2024) to ensure that the spacing between each section is uniform and that there is no obstruction of the audience's view.

[0020] Connection method: The structure adopts "detachable buckle + positioning pin". Each partition interface is designed with 3 buckles (material ABS plastic, load capacity ≥2kg) and 2 positioning pins (diameter 5mm, length 10mm). The positioning pins are inserted into the pin holes of the corresponding partition (tolerance H7 / g6) to ensure that the axis deviation of each partition is ≤1mm after connection, and facilitates disassembly and maintenance in the future.

[0021] Zoned structure blueprint output: The blueprint is in AutoCAD format and includes three views (front view, top view, and side view). Annotations include: The length, width, and height dimensions of each section (accurate to 1mm), such as the node compartment being "420mm in diameter and 300mm in length"; Interface location coordinates (e.g., the center coordinates of the experimental module interface (300,0,0)); Pre-reserved installation locations for key components (e.g., the lifting platform installation area is at the bottom of the node compartment, with dimensions of 100mm × 80mm). Material specifications (e.g., frame made of 6061 aluminum alloy, 2mm thick). The blueprints need to be verified for structural strength (e.g., through ANSYS software simulation to ensure that the deformation is ≤0.5mm when the frame bears a load of 5kg). Once confirmed to be correct, the blueprints will be output as the basis for subsequent model making.

[0022] Create a solid model framework based on the partition structure blueprint, reserve installation interfaces for simulation operation components at corresponding positions inside the model, and output a model framework with interfaces. The physical model framework serves as the carrier for all functional components. It must balance structural strength and interface precision to ensure that the simulation operation components can be accurately assembled and operate stably.

[0023] Frame material selection and processing: Material selection: The main frame is made of 6061 aluminum alloy (density 2.7g / cm³). 3 With a tensile strength of 276MPa, it balances lightweight and strength. The panel is made of transparent acrylic sheet (3mm thick, 92% light transmittance, making it easy to observe the internal structure). Processing technology: According to the blueprint, aluminum alloy profiles (20mm×20mm cross section) are cut by CNC milling machine (accuracy ±0.1mm) to form a frame skeleton, and then welded by argon arc welding (weld height 3mm, no false welds). After welding, the weld is polished with sandpaper (roughness Ra≤1.6μm) to ensure the surface is flat. Panel installation: Acrylic panels are CNC laser cut (accuracy ±0.2mm) into the corresponding shape and fixed to the inside of the frame with M3 countersunk screws (150mm spacing). The screw heads are lower than the panel surface (to avoid scratching the audience). After installation, sealant (silicone glue, temperature resistant -40℃~80℃) is used to seal the gaps to prevent dust from entering.

[0024] Pre-installed mounting interfaces for simulation operation components: Plant cultivation chamber interface: A rectangular slot (150mm×100mm×50mm, 30mm depth) is reserved in the inner 20cm area in front of the experimental chamber model. A 3mm thick EVA buffer pad is pasted on the inner wall of the slot (to prevent damage from collisions during installation). Four M4 threaded holes (50mm×80mm spacing) are reserved at the bottom of the slot for fixing the circuit board of the plant cultivation chamber; two Φ5mm wire holes (20mm spacing) are also reserved for power cables and signal cables to pass through.

[0025] Fixed handle interface: In the 15cm long fixed handle installation area of ​​the living cabin model, 3 sets of M5 threaded holes are reserved (2 holes in each set, 80mm apart, 15mm deep), with a thread precision of 6H, to ensure that the handle will not be loose after installation; the "Fixed handle installation position" (font height 5mm) is laser-engraved around the hole position for easy identification during later assembly.

[0026] Other interfaces: Two Φ8mm hook mounting holes (100mm apart) are reserved on the inside of the storage compartment model for hanging simulated material bags; a lifting platform mounting position (100mm×80mm, with 4 M6 positioning holes) is reserved at the bottom of the node compartment model, with a positioning hole tolerance of H7 to ensure the installation accuracy of the lifting platform.

[0027] Interface accuracy testing and framework output: Accuracy testing: A coordinate measuring machine (accuracy ±0.02mm) is used to test the positional deviation of each interface. For example, the deviation of the center coordinate of the plant cultivation chamber slot from the blueprint is ≤0.5mm, and the verticality deviation of the threaded hole of the fixed handle is ≤0.1mm / m. A feeler gauge (accuracy 0.02mm) is used to test the flatness of the inner wall of the slot, and the gap is ≤0.1mm. Frame output: After passing the inspection, affix partition label stickers (such as "Experimental Chamber - Plant Cultivation Area", font height 8mm, color red) to the outside of the frame, and affix installation instruction labels (such as "Plant Cultivation Chamber: Push in along the slot and tighten 4 M4 screws") to the interface. The final output is a stable and precise interface model frame.

[0028] Install simulation operation components: assemble a triggerable display plant cultivation chamber at the experimental chamber interface, install a force feedback fixed handle at the living chamber interface, and output the model of the basic assembly components. The simulated operating components need to replicate the actual functions of the space station and also have interactive triggering capabilities, allowing viewers to intuitively experience space classroom knowledge through operation. The core is to ensure that the functions of the components match the teaching content.

[0029] The plant cultivation chamber assembly can be triggered for display: Plant cultivation container structural design: The container body is made of transparent acrylic material (dimensions 148mm×98mm×48mm, compatible with pre-reserved slots), and contains 3 parts internally: Simulated plant model: A Arabidopsis thaliana model (50mm high, 6 leaves, silicone material, emerald green color) is made at a 1:10 scale and fixed in a simulated culture dish (30mm in diameter, resin material) at the bottom of the chamber; Trigger display module: Built-in 2.4-inch OLED display (resolution 320×240, brightness 300cd / m²) 2 Suitable for the lighting environment of the exhibition hall), the screen is embedded in the front of the cabin and displays the content of "Arabidopsis thaliana growth cycle under microgravity" (including pictures and text descriptions of the three stages of germination, leaf unfolding and flowering). Touch triggering component: A capacitive touch sensor (model TTP223, sensitivity 10pF, response time ≤100ms) is used. The sensor probe is attached to the side of the chamber (labeled "Touch to view growth data"). When the audience touches the probe, it triggers the display screen to switch the displayed content.

[0030] Assembly process: Connect the touch sensor and the control cable and power cable (5V DC, current ≤100mA) of the OLED screen to the control board (model Arduino Nano) inside the model through the reserved wire hole. Push the plant cultivation chamber into the rectangular slot of the experimental chamber, ensuring that the chamber fits snugly against the inner wall of the slot. Connect the chamber to the pre-drilled hole at the bottom of the slot using four M4×10mm countersunk screws (made of stainless steel) through the threaded holes at the bottom of the chamber. Tighten the screws to a torque of 0.5 N·m (to avoid crushing the chamber). Power-on test: When the touch sensor is triggered, the display screen switches the content of the growth stage normally without any lag or black screen, and the trigger success rate is ≥99%, confirming that the function is normal.

[0031] Installation of fixed handle with force feedback: Fixed handle structure design: The handle body is made of ABS plastic (120mm×30mm×20mm, surface texture simulates the anti-slip texture of a real space station handle), and integrates two parts internally: Force feedback actuator: It adopts a micro DC motor (model 28BYJ-48, torque 0.5N・m, reduction ratio 1:64). The motor output shaft is connected to the cam structure inside the handle. When the audience grips the handle and applies force, the motor drives the cam to rotate, generating reverse resistance (simulating the experience of "requiring force but feeling no weight" in the microgravity environment of space). Force sensor: A strain gauge force sensor (model LZ-F10, range 0-50N, accuracy ±0.5N) is used. It is attached to the connection between the handle and the mounting base to detect the force applied by the audience and control the motor resistance output (the greater the force applied, the greater the resistance, proportional coefficient 0.8).

[0032] Installation process: Connect the control lines (485 communication bus) of the force sensor and the force feedback motor to the control board through the wiring channel near the threaded hole reserved in the living quarters; Secure the handle mounting base (aluminum alloy, 80mm×30mm×10mm) to the pre-drilled threaded hole in the living compartment using two M5×15mm screws. Tighten the screws to a torque of 1.0 N·m to ensure the mounting base is stable. Functional test: When the audience grips the handle and applies a force of 5N, the motor generates a reverse resistance of 4N (0.8×5). The force sensor detection error is ≤0.2N, and the resistance feedback is timely (delay ≤200ms), confirming that the force feedback function is normal.

[0033] Model output of the basic assembly components: After installation, perform linkage tests on all simulation operation components: when touching the plant cultivation chamber, the display screen should show normally; when applying force by gripping the fixed handle, the force feedback should be normal; when not in operation, the components should be in standby mode (the display screen should show a static plant image, and the handle should offer no resistance). Check the connection stability between the components and the model frame (e.g., when shaking the model, the plant cultivation chamber should not loosen, and the handle should not shift). Finally, output a model with the basic components assembled and functions normally.

[0034] A lifting platform carrying a simulated astronaut model is installed in the node module. A multi-jointed robotic arm is assembled on top of the model. All components are integrated and the interactive functions are tested to form a display carrier with functional interactive attributes.

[0035] The lifting platform and robotic arm are the core components of dynamic interaction. They need to be linked with the simulated astronaut model and the functional areas of the space station to simulate the movement and operation of astronauts in the space station and enhance the scene-based effect of the display.

[0036] Installation of the lift platform carrying a simulated astronaut model: Lifting platform selection and structure: A ball screw type lifting platform (model ZOLIX TGA100, stroke 500mm, positioning accuracy ±0.1mm, maximum speed 50mm / s, load capacity 5kg) is adopted to meet the lifting requirements of the node module model (height 300mm); a load plate (size 120mm×100mm, material aluminum alloy) is installed on the top of the lifting platform to fix the simulated astronaut model.

[0037] Astronaut Simulation Model Production: Made at a 1:10 scale, 200mm tall, made of silicone (skin-like texture), wearing a white spacesuit (printed with the Chinese space station logo), holding a mini national flag (50mm×30mm, made of nylon), the bottom of the model is fixed to the platform with 4 M3 screws, and the verticality deviation of the model after fixing is ≤1mm (calibrated with a level).

[0038] Installation process: Secure the lifting platform to the reserved mounting position (100mm×80mm) at the bottom of the node cabin model using four M6×20mm screws. Tighten the screws to a torque of 2.0 N·m to ensure that the lifting platform coincides with the axis of the node cabin (deviation ≤0.5mm). Connect the control line (RS485 communication) and power line (24V DC, current ≤2A) of the lifting platform to the control board. Install a position sensor (model E3Z-LS63, detection distance 5-30mm, accuracy ±0.1mm) on the side of the lifting platform to provide feedback on the current height of the lifting platform. Lifting test: Control the lifting platform to rise from the bottom (0mm) to the top (300mm) without any jamming. The position sensor feedback height error is ≤0.2mm. It automatically stops when it reaches the top (triggers the limit switch), confirming that the lifting function is normal.

[0039] Multi-joint robotic arm assembly: Robotic arm selection and parameters: Select a 6-joint industrial robotic arm (model DOBOT Magician Lite, arm span 400mm, repeatability ±0.2mm, rotation range of each joint 0-360°, end-effector load 100g), adapted to the top space of the model (height from the top of the node cabin model to the ceiling of the exhibition hall ≥500mm), and install a miniature gripper at the end (gripping force 1N, adapted to the simulation experiment box (size 30mm×20mm×15mm)).

[0040] Installation process: Fix the robotic arm base (made of cast iron, weighing 2kg, anti-tipping) to the center of the top of the model (coordinates (0,0,300mm)) with 4 M8×25mm expansion screws, and calibrate the level of the base with a laser level (deviation ≤0.1mm / m). Connect the control cable (USB communication) and power cable (12V DC, current ≤1A) of the robotic arm to the control board. Install encoders (resolution 1024 pulses / revolution) on each joint of the robotic arm to monitor the joint rotation angle. Motion test: The robotic arm was controlled to move from its initial position (all joint angles were 0°) to above the experimental cabin model, grab the simulation experimental box, and then move to the living cabin model. The entire movement was smooth, the gripper held firmly (the experimental box did not fall off), and the joint angle error was ≤0.5°, confirming that the robotic arm was functioning normally.

[0041] The integration of testing and demonstration platforms is formed: Integrating all components: The control lines of the lifting platform, robotic arm, plant cultivation chamber, and fixed handle are all connected to the central control board (model STM32F103, processing frequency 72MHz), and centralized control is achieved through control software (developed based on LabVIEW); Interactive functionality testing: Lifting platform linkage: The lifting platform is controlled to lift the astronaut model to the top of the node module. After the position sensor feedback is in place, the robotic arm moves synchronously to the front of the astronaut model to simulate "cooperative operation between astronaut and robotic arm". Component linkage: When the plant cultivation cabin is touched to trigger the display, the robotic arm moves to the vicinity of the experimental cabin, simulating "the robotic arm transporting experimental samples"; when the fixed handle is gripped and force is applied, the lifting platform slowly descends, simulating "astronauts moving under microgravity"; Stability test: After running continuously for 2 hours, all components were fault-free (e.g., the lifting platform did not jam, the robotic arm did not misalign, and the touch trigger had no delay). The interactive functions matched the teaching content (e.g., the plant cultivation cabin demonstrated the corresponding robotic arm experimental operation), and finally formed a display carrier with functional interactive attributes.

[0042] The dynamic projection module is used to generate teaching content images adapted to the model structure. Through cross-projection using dual projectors, a scene anchor point matching algorithm is employed to align the projected image with the model's functional zones in real time. When switching teaching themes, the module automatically adjusts the image scaling and content focus, generating dynamic projection content that is linked to the functions of the entity modules. Specifically, the dynamic projection module is used for: Based on the teaching focus of each functional area of ​​the space station, a multi-theme teaching content library was created, with each theme content associated with a corresponding area tag, and a structured teaching content library was output. The core of the dynamic projection module is to achieve "screen linkage with functional zones" by associating teaching content with functional zones of the physical model. Therefore, it is necessary to first build a multi-topic content library covering core knowledge around the teaching focus of each functional zone of the space station to ensure that the content matches the model function, thus laying the foundation for subsequent projection alignment and theme switching.

[0043] Teaching Focus and Theme Division: Referring to the popular science direction of "Tiangong Classroom" and the actual functions of the space station, the teaching themes are divided according to the core functional areas of the model (node ​​module, experimental module-plant cultivation area, living quarters-fixed handle area, storage module). Each theme focuses on 1-2 core knowledge points to avoid redundancy. Theme 1: "Experimental Module - Plant Cultivation Module Microgravity Growth": Focusing on the growth characteristics of plants in the microgravity environment of the space station, corresponding to the plant cultivation area of ​​the experimental module, the knowledge points include "Arabidopsis thaliana growth cycle" and "the effect of microgravity on root orientation". Topic 2: "Living Module - Use of Fixed Handles and Microgravity Movement": Focusing on how astronauts move using fixed handles in microgravity, corresponding to the fixed handle area of ​​the living module, knowledge points include "handle force feedback principle" and "safe movement posture"; Theme 3: "Node Module - Astronaut Ascent and Descendant and Module Docking": Focusing on the node module's function as a hub, corresponding to the node module's ascent and descendant area, key knowledge points include "ascent and descendant movement logic" and "module docking interface structure"; Theme 4: "Storage Modules - Material Storage and Robotic Arm Handling": Focusing on space station material management and robotic arm operation, corresponding to storage modules and the robotic arm on top of the model, knowledge points include "sealed storage design" and "robotic arm grasping principle".

[0044] Thematic content creation and association: Each thematic content includes 3 types of media formats to ensure intuitive teaching: Video content: 4K resolution (3840×2160), 30fps, MP4 format, 30-60 seconds in length, such as the video in Theme 1 showing "the microgravity growth process of Arabidopsis thaliana from germination to flowering", with the plant position in the picture aligned with the plant model in the simulated plant cultivation chamber inside the model; Image content: PNG format, resolution 2048×1536, including knowledge point annotations, such as the image annotations for Topic 2: "Fixed handle force feedback sensor location (red arrow)" and "Astronaut's hand gripping area (blue box)"; Text content: Use Microsoft YaHei font, font size 24-36 (to ensure clear readability from 5 meters away), white color (to match the dark surface of the model), such as the text description of Theme 3: "The maximum travel of the lifting platform is 30cm, simulating the movement of astronauts within a 0.5-meter range inside the node module (1:10 scale restoration)."

[0045] Structured teaching content repository construction: The content is stored using a MySQL database, with tables containing 5 core fields: Topic ID: A unique identifier (e.g., T001 corresponds to topic 1, T002 corresponds to topic 2); Subject name: such as "Experimental Chamber - Microgravity Growth of Plants"; Content types: differentiate between video (V), image (I), and text (T); Partition labels: associated entity model partitions (e.g., "Experimental Chamber - Plant Cultivation Area" "Living Chamber - Fixed Handle Area"); Resource path: Content storage path.

[0046] For example, the complete entry for T001 in the database table is: "T001, Experimental Chamber - Microgravity Growth of Plant Cultivation Chamber, V / I / T, Experimental Chamber - Plant Cultivation Area, / Content / T001 / ...". After the content library is built, it needs to pass a teaching adaptability test. For example, the plant growth stages in the video footage of Topic 1 should match the morphology of the simulated plants in the model, and the deviation between the image annotations of Topic 2 and the actual fixed handle positions should be ≤2mm. After confirming that there are no errors, the structured teaching content library will be output.

[0047] Install the two projectors at the preset positions on the outside of the model, adjust the projection angle so that the image overlaps and covers the surface of the model, calibrate the brightness and color consistency of the two projectors using a grayscale test card, and output the calibrated dual projector system. The installation and calibration of dual projectors are fundamental to ensuring the integrity of the projected image. Precise positioning and parameter calibration are required to avoid uneven brightness, color deviation, or incomplete coverage, thus providing stable hardware support for subsequent anchor point matching and image alignment.

[0048] Projector selection and preset location determination: Projector selection: Choose a high-brightness, short-throw projector (such as the BenQ MH560, with a brightness of 4000 lumens, a contrast ratio of 10000:1, and a throw ratio of 1.5-1.65:1) to adapt to the ambient light of the exhibition hall (illuminance ≤500 lux) and ensure clear images; support 1920×1080 resolution to match the resolution of the teaching content and avoid image stretching; Preset positions: Centered on the physical model, the projectors are symmetrically distributed on the left and right sides of the model's outer perimeter. The left projector is mounted on the ground support on the left side of the experimental cabin (1.2 meters high, 2.0 meters from the model), and the right projector is mounted on the ground support on the right side of the living cabin (1.2 meters high, 2.0 meters from the model). The projection angles are adjusted so that the left projector is at a 45° angle to the front of the model (covering the left side of the experimental cabin and the left side of the node cabin), and the right projector is at a 135° angle to the front of the model (covering the right side of the experimental cabin and the right side of the living cabin). Ensure that the images from both projectors overlap in the central area of ​​the model (front of the node cabin), with an overlap width of ≥10cm (to reserve space for subsequent image fusion).

[0049] Projection angle and coverage adjustment: Initial adjustments: Using the projector's built-in keystone correction function (vertical ±40°, horizontal ±30°), the left projector image covered the left side of the experimental chamber (179cm long, 42cm high) and the left side of the node chamber (30cm long, 42cm high), while the right projector image covered the right side of the experimental chamber and the right side of the living quarters chamber (50cm long, 42cm high), ensuring no obvious missed areas (missed area ≤ 5cm). 2 ); Cross-area confirmation: Project a white test screen (full white image) and observe the cross-area in the center of the model. Ensure that there is no obvious misalignment between the two projector images in the cross-area (misalignment ≤ 2mm). If misalignment exists, fine-tune the horizontal position of the projector stand (move left or right, adjusting by 5mm each time) until the images in the cross-area are continuous.

[0050] Brightness and color consistency calibration: Calibration tools: A 24-level grayscale test chart (A4 size, grayscale value from 0 (pure black) to 255 (pure white)) and ColorChecker color chart (containing 24 standard colors) were used to collect image parameters through professional calibration software (such as Datacolor SpyderX); Brightness calibration: Attach the grayscale test card to the front of the model test chamber (intersection area), and adjust the brightness knobs of both projectors until the brightness of both projectors at a grayscale value of 128, as measured by the calibration software, is 300 cd / m². 2(deviation ≤10cd / m) 2 To avoid one side being too bright and the other too dark, adjust the contrast to 80% to ensure that each gray level on the grayscale test card can be clearly distinguished (without gray level banding). Color calibration: Place the ColorChecker color chart in the same position, collect the color data of the color chart projected by the two projectors using calibration software, and adjust the RGB three-color gain of the projectors (red: 95, green: 98, blue: 92) so that the color difference value ΔE between the two projectors for the same standard color is ≤2 (ΔE is the color difference degree, ΔE≤2 is imperceptible to the human eye). For example, the actual RGB of standard red (RGB 255,0,0) after projection on the left projector is (250,5,3), and on the right projector it is (252,3,5), ΔE=1.8, which meets the requirements; Calibration verification: Project a color teaching video (such as the plant growth video of Theme 1), observe the cross-area image, if there is no obvious color banding or brightness change, the calibration is deemed qualified, and the calibrated dual projector system is output.

[0051] Physical anchor points are set at the edges of each functional partition of the model. The scene anchor point matching algorithm is used to identify the anchor point position, calculate the geometric deviation between the projected image and the partition, and adjust the output parameters of the projector in real time to make the image and the partition outline accurately aligned, and output the aligned projected image. Physical anchor points are "reference marks" for aligning the projected image with the physical model's partitions. The scene anchor point matching algorithm then identifies the anchor points to achieve geometric correction of the image, ensuring that the projected content accurately fits the partition outline, avoiding the image from exceeding the partition or not covering it completely, and solving the problems of "projected image exceeding the medium" and "ghosting at superposition".

[0052] Physical anchor point settings: Anchor point selection and placement: Black circular reflective stickers (5mm in diameter, 0.1mm thick, reflectivity ≥80%, facilitating algorithm recognition) are used as physical anchor points. Anchor points are set along the outline edge of each functional zone, with the principle of "at least 4 anchor points per zone, distributed at the four corners and the midpoint of the long side of the zone". Plant cultivation area in the experimental chamber (150mm×100mm): 1 anchor point is set at each of the upper left corner, upper right corner, lower left corner, lower right corner and the midpoint of the long side (midpoint of the upper long side and midpoint of the lower long side), for a total of 6 anchor points. The center of the anchor point is 2mm away from the edge of the partition (to avoid being covered by the screen). Living compartment fixed handle area (150mm×80mm): 4 anchor points are set at each of the four corners, and the anchor points correspond to the mounting holes of the fixed handles (to facilitate the alignment of the screen and the handles). Node cabin lifting platform area (100mm×80mm): One anchor point is set at each of the four corners, for a total of 4. The anchor points correspond to the movement trajectory boundary of the lifting platform. Anchor point coordinate recording: Use a laser rangefinder (accuracy ±0.1mm) to measure the physical coordinates of each anchor point (establish a three-dimensional coordinate system with the model center as the origin, and record the X, Y, and Z axis coordinates). For example, the physical coordinates of the anchor point at the upper left corner of the plant cultivation area in the experimental chamber are (80mm, 20mm, 150mm). After recording, store the coordinates in the anchor point coordinate database as the benchmark for the algorithm to calculate the deviation.

[0053] Implementation of scene anchor point matching algorithm: The scene anchor point matching algorithm is based on the SIFT (Scale Invariant Feature Transform) feature detection algorithm. Its core is to identify anchor point features in the projected image, compare them with the coordinates of physical anchor points, calculate the geometric deviation, and adjust the projector parameters. The specific process is as follows: Anchor point image acquisition: The test screen containing the anchor point recognition area is projected through the projector (only the area where the anchor points of each partition are located is illuminated), and the high-definition camera on the top of the model (resolution 1920×1080, frame rate 15fps) is activated at the same time to acquire images containing physical anchor points. SIFT Feature Extraction: The acquired image is converted to grayscale (to reduce computation), and the SIFT algorithm is used to detect anchor point features in the image (edge ​​gradient of black circles, scale invariance features), generating a feature vector (128 dimensions) for each anchor point, and eliminating non-anchor point interference (such as scratches on the model surface, where the difference between the feature vector and the anchor point is >0.5). Anchor point matching and deviation calculation: The extracted anchor point feature vectors are matched with the physical anchor point features in the anchor point coordinate database (using the nearest neighbor matching algorithm with a matching threshold of 0.8) to determine the physical anchor point corresponding to each image anchor point; the geometric deviation between the pixel coordinates of the anchor points in the image (e.g., the pixel coordinates of the upper left corner anchor point in the experimental cabin in the projected image are (320, 240)) and the pixel mapping coordinates of the corresponding physical anchor points (calculated according to the model scale and projector projection ratio, e.g., the physical coordinates (80mm, 20mm) correspond to pixel coordinates (315, 235)) is calculated. The deviation includes 3 types: Translation deviation: ΔX = 320 - 315 = 5 pixels, ΔY = 240 - 235 = 5 pixels (representing the image offset along the X and Y axes); Scaling deviation: A physical anchor point spacing of 150mm corresponds to an anchor point spacing of 145 pixels in the image, while the theoretical spacing is 150 pixels. Scaling deviation = 145 / 150 = 0.967 (the image is too small). Rotation deviation: The angle between the line connecting the anchor points in the image and the line connecting the physical anchor points is 3° (the image is rotated 3° clockwise). Projector parameter adjustment: Based on the calculated geometric deviation, adjustment commands are sent via the projector's RS232 communication interface to correct the output parameters in real time. Translation adjustment: Send the command "translate X = -5 pixels, translate Y = -5 pixels" to the projector to move the image to the lower left and eliminate ΔX and ΔY deviations; Scaling adjustment: Send the command "scaling ratio = 1 / 0.967≈1.034" to enlarge the image by 3.4% to match the anchor point spacing with the theoretical value; Rotation Adjustment: Send the command "Rotation Angle=-3°" to rotate the screen counterclockwise by 3° to eliminate rotation deviation; Edge cropping: For image edges that exceed the partition outline (e.g., the right side of the image exceeds the experimental chamber partition by 2mm after adjustment), the excess part is cut off by the software cropping function (the cropping range is calculated based on the anchor point coordinates) to avoid the problem of "image exceeding the medium". Alignment verification and image output: After adjustment, re-acquire anchor point images and calculate geometric deviations. The overall deviation of all anchor points should be ≤1 pixel (corresponding to an actual deviation of ≤0.5mm on the model surface). For example, the anchor point deviations in the plant cultivation area of ​​the experimental cabin are all 0.8 pixels, which meets the requirements. Project teaching content (such as plant growth pictures of Theme 1), observe the fit between the image and the partition outline, the gap between the image edge and the partition edge should be ≤1mm, there should be no ghosting or missing projection, and output the aligned projection image.

[0054] When a theme switching command is received, the system automatically calls up the teaching content of the corresponding theme, adjusts the screen scaling ratio according to the target partition size, and updates the anchor point matching parameters in sync to maintain alignment, generating dynamic projection content that is linked with the functions of the physical module.

[0055] Theme switching is key to achieving "dynamic updates of teaching content with audience interaction". It is necessary to ensure that the screen is always precisely aligned with the target area during the switching process, and that the content focus matches the function of the area. This will prevent screen misalignment or content disconnection caused by the switching, and achieve functional linkage with the physical modules.

[0056] Receiving and parsing topic switching commands: The trigger scenarios for theme switching commands include "interactive response commands sent by the behavior perception module" (such as when a visitor focuses on the fixed handle area of ​​the living cabin, triggering a switch to theme 2) and "manual trigger commands" (such as when a guide sends a switching command via a tablet). The command format is JSON, as shown in the example: {"commandType":"themeSwitch","targetThemeID":"T002","targetZone":"Living Pod - Fixed Handle Area","timestamp":"2024-10-10 14:30:00"} Where “targetThemeID” is the target theme ID and “targetZone” is the target functional zone. After receiving the instruction through the TCP / IP communication link (port 8080, transmission delay ≤100ms), the system parses out the target theme and target zone, and calls the corresponding content resources (such as videos, images, and text for theme 2) in the structured teaching content library.

[0057] Adjusting the screen zoom level: The scaling ratio is adjusted based on the actual size of the target area and the projector's throw ratio to ensure that the teaching content fills the target area without stretching. The calculation formula is: Scaling ratio = Actual width of target area (mm) × Projector throw ratio / Original width of projected image (pixels) × Model scale Wherein, the projector throw ratio = projection distance (mm) / screen width (mm), the projection distance in this system = 2000mm, the original screen width = 1920 pixels (1080P resolution), and the model ratio = 1:10; For example, if the target area is the fixed handle area of ​​the living quarters, with an actual width of 150mm (model size), corresponding to a width of 1500mm (150mm × 10) on the actual space station, then: Scaling ratio = 150mm × (2000mm / actual screen width) / 1920 pixels × 10 Further calculation: Actual screen width = Projection distance / Projection ratio = 2000mm / 1.6 ≈ 1250mm, substituting, we get: Scaling ratio = 150 × (2000 / 1250) / 1920 × 10 = 150 × 1.6 / 19200 = 240 / 19200 = 0.0125 (This is the mapping ratio between pixels and actual size. When actually adjusting, it is converted into a screen scaling factor. For example, scaling the original 1920 pixel wide screen to 1920 × 0.8 = 1536 pixels will make the screen width match the target partition). After adjustment, the width of the projected image of Theme 2 (showing how to use the fixed handle) is consistent with the width of the fixed handle area of ​​the living cabin (150mm), and the deviation between the projected position of the handle in the image and the actual handle position of the model is ≤1mm, ensuring that "the image is aligned with the physical part".

[0058] Anchor point matching parameters are updated synchronously: After the target partition is switched, the position and number of its physical anchor points are different from the original partition (e.g., the fixed handle area of ​​the living quarters has 4 anchor points, and the plant cultivation area of ​​the experimental compartment has 6 anchor points). The parameters of the scene anchor point matching algorithm need to be updated synchronously. Anchor point database switching: Retrieve the physical anchor point coordinates of the target partition (living quarter fixed handle area) from the anchor point coordinate database and replace the anchor point data of the original partition; Feature detection parameter adjustment: Based on the distribution density of the target partition anchor points (e.g., the distribution of 4 anchor points is relatively sparse), adjust the feature detection threshold of the SIFT algorithm (from the original 0.04 to 0.03) to ensure that the anchor point features can be accurately extracted; Deviation calculation benchmark update: Based on the anchor point coordinates of the target partition, recalculate the geometric deviation between the projected image and the partition. If there is a 2-pixel translation deviation in the image after switching, immediately send an adjustment command to correct it and maintain the alignment. Dynamic projection content generation: After completing content retrieval, scaling adjustment, and anchor point update, the teaching content of the target theme is projected. For example, after switching to theme 2, the left projector projects the left view of the fixed handle of the living module (including force feedback sensor labels), the right projector projects the right view (including astronaut grip animation), and the cross area projects a dynamic diagram of the force feedback of the handle. The image is completely aligned with the actual handle and fixed structure of the model, with no overshoot or undershoot, and generates dynamic projection content that is linked with the function of the physical module (such as when the audience touches the model handle, the projected image synchronously displays the text explanation of "handle force feedback principle").

[0059] The behavior perception module is used to capture audience interaction behavior. It identifies the audience's standing area and gaze direction using a microwave radar array, and determines the interaction intent based on the duration. When an audience focuses on a certain functional area for more than a preset time, it triggers the display of in-depth teaching content corresponding to that area and generates targeted interactive response commands. Specifically, the behavior perception module is used for: A microwave radar array is distributed in a fan shape around the model, dividing it into multiple monitoring areas. The radar sampling frequency is configured to capture the positional movement and micro-movement signals of the audience, and output the completed radar perception system. The core hardware foundation of the behavior perception module is a microwave radar array. Its deployment needs to cover the area around the model where the audience may stand to ensure no blind spots. At the same time, through reasonable sampling frequency settings, it can accurately capture the audience's positional changes and subtle movements, providing high-quality raw data for subsequent behavior analysis. This is consistent with the technical direction of "human microwave radar sensing device".

[0060] Microwave Radar Selection and Parameter Determination: A 24GHz millimeter-wave radar sensor (TI IWR6843) was selected. This radar model features a range resolution of 0.01m, an angular resolution of ±1°, and a maximum detection range of 8m, meeting the requirement of "accurate perception within 3 meters" in exhibition hall scenarios. It also supports micro-Doppler effect detection, capable of recognizing micro-movements of ≥0.1mm (such as the slight head rotation of a viewer when looking at a model), meeting the technical requirements for capturing interactive behavior. The radar operates on 5V DC with a working current ≤150mA, and can be centrally powered through the model's internal power module, avoiding the complexity of separate wiring.

[0061] Sector-shaped distribution and monitoring area delineation: Centered on the model of the scene-based entity module (coordinate origin (0,0,0)), six radar sensors are deployed within a 3-meter radius around the model according to the "uniform fan-shaped" principle, forming 360° coverage without blind spots. Distribution angle: The six radars are installed at positions at 0°, 60°, 120°, 180°, 240° and 300° from the center of the model, respectively. The horizontal detection angle of each radar is 50° (to cover the gap between adjacent radars and avoid blind spots). Installation height: All radars are fixed on ground supports 1.2 meters high (level with the audience's chest when standing, reducing interference from ground debris), and the bottom of the supports is equipped with a weighted base (weighing 2kg) to prevent tipping; Monitoring area division: Based on the radar distribution angle, the area within 3 meters is divided into 6 independent monitoring areas (Area 1: 0°-60°, Area 2: 60°-120°, ..., Area 6: 300°-360°). Each area corresponds to the main detection range of one radar. At the same time, there is a 10° overlap between adjacent areas (to ensure no missed detection at the boundary). For example, Area 1 and Area 2 overlap at 60°-70° to avoid the audience not being identified when they are at the boundary.

[0062] Radar sampling frequency configuration: Sampling frequency is a key parameter that determines whether micro-limb movements can be captured. Considering the characteristics of audience interaction behavior (position movement speed ≤ 0.5m / s, micro-limb movement period ≥ 0.1s), the radar sampling frequency is set to 10Hz (i.e., 10 signals are collected per second). Frequency selection criteria: A sampling frequency of 10Hz can improve the time resolution to 0.1 seconds, which can completely capture the audience's movement from "standing still" to "slightly turning their head" (this movement lasts about 0.2 seconds and can be captured by 2 samples); if the frequency is too low (such as 5Hz), fast micro-movements may be missed; if the frequency is too high (such as 20Hz), it will increase data redundancy and computational burden. Parameter synchronization: The sampling frequency of the six radars was uniformly set to 10Hz using the radar configuration software (TI mmWave Studio). At the same time, the "micro-Doppler mode" (a signal processing mode specifically for detecting micro-limb movements) was enabled, and the signal output format was set to "distance-angle-Doppler" three-dimensional data (to facilitate subsequent analysis of position and movement).

[0063] Deployment verification and system output: After completing the hardware deployment, conduct tests on the sensing range and signal quality: Blind spot test: Test personnel walked around the monitoring areas around the model and observed through radar data acquisition software. Personnel signals could be stably detected in all areas (signal strength ≥ -60dBm, no blind spots with continuous no signal). Micro-motion capture test: The tester stands in area 2 and performs the actions of "gazing model" and "turning head to look elsewhere". The radar can accurately capture the Doppler frequency shift corresponding to the action (shift +2Hz when gazing, shift -1.5Hz when turning head). System output: After confirming that all radars are working properly, the sampling frequency is consistent, and the monitoring area is reasonably divided, the system outputs the completed radar sensing system. The system includes configuration information such as the IP addresses of 6 radars, the coordinate range of the monitoring area, and the sampling frequency, which prepares for subsequent signal acquisition.

[0064] The radar sensing system collects signals in real time, identifies the standing area of ​​the audience through signal strength analysis, determines the gaze direction by combining the signal Doppler effect, and outputs raw data of audience behavior. The core function of radar perception systems is to convert physical signals into analyzable behavioral data, determine the location of the audience through the spatial distribution of signal intensity, and distinguish the direction of gaze using the Doppler effect. This solves the problem of "knowing only that the audience exists, but not which section they are paying attention to," and provides accurate behavioral evidence for subsequent intention judgment.

[0065] Real-time signal acquisition and preprocessing: Six radars synchronously acquire signals at a frequency of 10Hz. The raw signals acquired include four types of parameters: distance (straight-line distance between the audience and the radar), angle (azimuth angle of the audience relative to the radar), signal strength (reflecting the degree of obstruction between the audience and the radar), and Doppler frequency (reflecting the motion state of the audience). The system preprocesses the acquired signals: Noise filtering: The Kalman filter algorithm is used to remove environmental noise (such as signal fluctuations caused by air conditioning in the exhibition hall; after filtering, the signal strength fluctuation range is reduced from ±5dBm to ±1dBm). Invalid signal rejection: Signals with a distance > 3 meters (outside the monitoring range) or a signal strength < -70dBm (possibly due to clutter interference) are rejected, while valid personnel signals are retained. For example, if a radar collects a signal with a distance of 2.5 meters, an angle of 5°, a signal strength of -55dBm, and a Doppler frequency of +2Hz, it is determined to be a valid signal after preprocessing. If a signal with a distance of 3.5 meters and a signal strength of -75dBm is collected, it is determined to be invalid and rejected.

[0066] Signal strength analysis and standing area identification: The identification of the audience's standing area is based on "multi-radar signal strength collaborative analysis". The core logic is that "the radar signal strength in the audience's area is the highest, and the signal strength of adjacent radars decreases with increasing distance". Signal strength threshold setting: Set the effective signal strength threshold of a single radar to -60dBm (experimentally verified that this value can effectively distinguish between people and environmental debris). When the signal strength of a radar is ≥-60dBm, mark the radar coverage area as "suspected area with audience". Area positioning: By combining the signal strength distribution of multiple radars, the specific monitoring area of ​​the audience is determined. For example, if the signal strength of radar 2 (covering area 2: 60°-120°) is -52dBm, radar 1 (area 1: 0°-60°) is -65dBm, and radar 3 (area 3: 120°-180°) is -68dBm, and only radar 2 has a signal strength ≥-60dBm, and the signal strength of adjacent radars decreases, then the audience is determined to be standing in area 2. If the signal strengths of radar 1 and radar 2 are both ≥-60dBm (-58dBm and -55dBm respectively), then the audience is determined to be in the overlapping area of ​​area 1 and area 2 (60°-70°), and further, by using angle parameters (the angle of the audience relative to radar 1 is 55°, and the angle relative to radar 2 is 5°), the audience can be accurately located to the edge of area 2.

[0067] Doppler effect and gaze direction judgment: The direction of a viewer's gaze can be determined by the Doppler frequency shift generated by subtle body movements (such as head turning or leaning forward). The core principle of the Doppler effect is that "the signal frequency increases when the target moves toward the radar and decreases when it moves away." Reference frequency calibration: When there is no audience, the radar's natural frequency (approximately 24.125 GHz) is collected as a reference; when there is an audience, the difference between the actual collected frequency and the reference frequency (Doppler shift Δf) is calculated. Direction determination rules: If the audience looks at the model, their heads will tilt slightly toward the model, causing the radar on the side facing the model to receive a positive frequency shift (Δf=+1Hz~+3Hz). For example, in the audience looking at the experimental cabin in area 2 (located in the positive direction of the model's X-axis), radar 2 (corresponding to area 2) collected Δf=+2.2Hz and determined the direction of gaze to be "experimental cabin-plant cultivation area". If the viewer turns their head to look elsewhere, their head is away from the model, and the radar receives a negative frequency shift (Δf=-1Hz~-3Hz), for example, Δf=-1.8Hz, and determines the direction of gaze as "away from the model"; Directional verification: By combining the Doppler frequency shifts of multiple radars to make a comprehensive judgment and avoid the error of a single radar, for example, radar 2 Δf = +2.2Hz, radar 3 Δf = +0.5Hz (the audience's side is facing radar 3), the direction of gaze is further confirmed to be the experimental cabin.

[0068] Raw data output of audience behavior: The raw data is stored in structured text format, containing five types of information: timestamp, standing area, gaze direction, signal strength, and Doppler shift. For example: "2024-10-15 09:30:01, Area 2 (60°-120°), gaze direction = experimental cabin - plant cultivation area, signal strength = -52dBm, Doppler shift = +2.2Hz". The data output frequency is consistent with the radar sampling frequency (10Hz) and is transmitted in real time to the behavior analysis module to provide input for subsequent time series analysis.

[0069] Perform time-series analysis on raw audience behavior data to calculate the dwell time and gaze duration of an audience in a certain functional area. When the duration exceeds a preset value, it is determined that there is a deep interaction intention in that area, and the intention judgment result is output. Temporal analysis is the core step of extracting "interaction intent" from "raw behavioral data". By analyzing the duration of viewers' stay and gaze in a specific area, it distinguishes between "casual passersby" and "active attention", ensuring that in-depth teaching content is triggered only for viewers with in-depth interaction needs, thus avoiding waste of resources. This is consistent with the interaction logic of "letting nearby viewers watch the teaching video from the beginning".

[0070] Time series analysis window and data sliding processing: A sliding time window was used for time series analysis, with a window length of 5 seconds (to balance capturing continuous behavior and rapid response) and a window sliding step of 1 second (updating the analysis results once per second). Window data integration: Within each sliding window, integrate 50 raw data points at a sampling frequency of 10Hz (5 seconds × 10 data points / second), remove outliers (such as invalid data caused by momentary signal interruption), and retain windows with a valid data ratio of ≥80% (otherwise, the data is considered invalid and will not be analyzed); Data association: The standing area data in the window is associated with the functional areas of the model. For example, “Area 2” corresponds to the two functional areas of the model: “Experimental Chamber - Plant Cultivation Area” and “Living Chamber - Fixed Handle Area”. Furthermore, the gaze direction (such as “gazing at the experimental chamber”) is used to lock onto “Experimental Chamber - Plant Cultivation Area” to ensure that the behavioral data corresponds accurately with the functional areas.

[0071] Calculation of dwell time and fixation duration: Dwell time calculation: Within the sliding window, the time the audience remains in the same functional zone corresponding to the monitoring area is counted. If more than 80% of the valid data within the window shows that the audience is in area 2 corresponding to "Experimental Chamber - Plant Cultivation Area", then the dwell time = window length × 80% = 4 seconds; if the audience moves from area 2 to area 3 within the window, the dwell time is divided according to the dwell ratio of different areas (e.g., 3 seconds in area 2 and 2 seconds in area 3). Calculation of gaze duration: Within the dwell time, the time the audience gazes at this functional area is counted. If 70% of the effective data in the window shows that the gaze direction is "Experimental Pod - Plant Cultivation Area", then the gaze duration = dwell time × 70% = 4 × 0.7 = 2.8 seconds. If the gaze direction changes frequently (such as switching between the experimental pod and the living pod), then the longest single gaze time is taken as the gaze duration.

[0072] Preset value settings and depth intent determination: The preset values ​​need to be set based on the average dwell time of visitors in the exhibition hall. Taking into account the interactive requirement of "triggering video playback when a person approaches", the "preset value for gaze duration" is set to 3 seconds (that is, if a visitor gazes at a certain functional area for more than 3 seconds, it is judged as having a deep interactive intention), and the "preset value for dwell time" is set to 5 seconds (to avoid being misjudged when passing by for a short time). Judgment logic: If a viewer stays in functional area A for ≥5 seconds and the gaze duration is ≥3 seconds within a certain sliding window, it is judged as "having a deep interaction intention with functional area A"; if the stay time is ≥5 seconds but the gaze duration is <3 seconds (such as the viewer looking down at their phone in the area), or the gaze duration is ≥3 seconds but the stay time is <5 seconds (such as a quick glance), it is not judged as a deep intention. Example: If a visitor stays in area 2 of the "Experimental Pod - Plant Cultivation Area" for 6 seconds, with a gaze time of 3.5 seconds, both exceeding the preset value, it is determined that "there is a deep interaction intention with the Experimental Pod - Plant Cultivation Area"; if the visitor stays for 6 seconds but the gaze time is only 2.5 seconds, it is determined that "there is no deep interaction intention".

[0073] Intent determination result output: The results include four types of information: "functional area identifier, deep interaction intent determination, dwell time, and gaze duration." For example, "functional area = experimental cabin - plant cultivation area (corresponding to area 2), deep interaction intent = yes, dwell time = 6 seconds, gaze duration = 3.5 seconds." If no deep intent is determined, the output is "functional area = living cabin - fixed handle area, deep interaction intent = no, reason = gaze duration 2.2 seconds < 3 seconds," providing a clear basis for triggering deep teaching content in the future.

[0074] Based on the intent judgment result, the corresponding in-depth teaching content of the functional area is triggered, and an interactive response instruction containing area identifier and content type is generated.

[0075] The triggering of in-depth teaching content needs to be precisely matched with functional areas to ensure that the areas that the audience focuses on can receive targeted knowledge supplementation. At the same time, through structured interactive response instructions, the intent information is transmitted to multi-dimensional linkage modules to realize the "perception-response" closed loop. This is the key link connecting behavioral perception and multi-component linkage.

[0076] Matching and calling up in-depth teaching content: The system has a built-in "Functional Zone - In-depth Content" mapping table, designed based on the requirement that "space classrooms need in-depth scientific knowledge." Each functional zone corresponds to 1-2 types of in-depth content, including "principle explanation videos (30-60 seconds), 3D structural animations, and textual knowledge points," ensuring that the content depth is higher than the basic projection content. Mapping rule example: Experimental Chamber - Plant Cultivation Area: The corresponding depth content is "Arabidopsis microgravity root growth mechanism" (video, 45 seconds, explaining how microgravity affects the geotropism of plant roots) and "Animation of the sealed structure of the plant cultivation chamber" (3D animation, showing how to maintain the temperature and humidity inside the chamber). Living Pod - Fixed Handle Area: The corresponding depth content is "The Mechanics of Human Movement under Microgravity" (text + diagrams, explaining why handles are needed for movement) and "The Working Principle of Handle Force Feedback Sensors" (short video, 25 seconds, showing how the sensor detects the force applied by the hand). Content retrieval: When the intent judgment result is "there is a deep interactive intent for the experimental cabin - plant cultivation area", the system retrieves the corresponding deep content of the partition from the structured teaching content library (built in step three), giving priority to video content (strong intuitiveness). If the video fails to load (such as network fluctuations), it will automatically switch to text + image content.

[0077] Interactive response command structure design: The instruction must contain "key information that can be parsed by the multi-dimensional linkage module," in JSON format (for easy cross-module data transfer), and the core fields include: Area Identifier: Uniquely identifies the target functional area, in the format of "functional area name + corresponding monitoring area", such as "experimental cabin - plant cultivation area (area 2)", to ensure that the multi-dimensional linkage module can locate the specific area; Content type: Mark the deep content type triggered, such as "video + 3D animation", to facilitate the dynamic projection module in preparing the corresponding playback resources; Content ID: A unique ID associated with in-depth content in the structured teaching content library, such as "DC001" (corresponding to the video "Microgravity Root Growth Mechanism of Arabidopsis"), to ensure unambiguous content retrieval; Trigger timestamp: Records the time when the instruction is generated, in the format "YYYY-MM-DD HH:MM:SS.fff" (accurate to milliseconds), used for multi-module synchronization (such as time alignment between dynamic projection and robotic arm movements). Priority: Set the command priority (level 1-5, level 1 is the highest). The command priority triggered by deep interaction intent is set to level 3 (higher than level 2 of manual switching command, lower than level 5 of emergency stop command) to ensure that important interaction needs are responded to first.

[0078] Command generation and transmission verification: Example of instruction generation: Based on the judgment result of "deep interaction intent between the experimental cabin and the plant cultivation area", generate instructions, such as: {"regionID":"Experimental Chamber - Plant Cultivation Area (Region 2)", "contentType":"Video + 3D Animation", "contentID":"DC001;DC002", "timestamp":"2024-10-15 09:30:06.123","priority":3}.

[0079] Transmission verification: The command is transmitted to the multi-dimensional linkage module and dynamic projection module via the LAN UDP protocol. The transmission timeout is set to 200ms, and the command is automatically retransmitted after the timeout (up to 3 times) to ensure that the command is not lost. After receiving the command, the receiver returns a "command received" confirmation signal. After the sender confirms, the command transmission is completed, avoiding the problem of "no response due to command not arriving".

[0080] Interactive response command output: The instructions are ultimately output in the form of "executable text + confirmation signal". On the one hand, they are passed to the multi-dimensional linkage module to control the movement of the lifting platform and robotic arm; on the other hand, they are passed to the dynamic projection module to switch in-depth teaching content, realizing the initial connection of "behavior perception - content response - component linkage" and laying the foundation for the subsequent generation of scenario-based interactive effects.

[0081] The multi-dimensional linkage module coordinates the collaborative actions of various components. Based on interactive response commands, it controls the lifting platform to move the simulated astronaut model to the corresponding functional area, simultaneously driving the robotic arm to simulate operation demonstrations. This ensures that the state changes and mechanical movements of the simulated operating components are matched in real time with the dynamically projected content, generating a scenario-based interactive space classroom display effect. Specifically, the multi-dimensional linkage module is used for: Receive interactive response commands, parse the target functional area location in the commands, generate the lifting platform movement path and the robotic arm operation task, and output motion control parameters; As the core of system collaboration, the multi-dimensional linkage module's primary task is to accurately receive and parse the interactive response commands sent by the behavior perception module, transforming abstract commands into executable hardware action parameters to ensure that the movements of the lifting platform and the robotic arm can accurately match the target functional areas. This process needs to be combined with the design requirements of "electromechanical linkage function" to achieve unbiased conversion of "command-action".

[0082] Mechanism for receiving interactive response commands: The module receives commands via a TCP / IP communication link (port 8081, transmission rate 1Mbps, latency ≤100ms). The command format is JSON (consistent with the output format of the behavior-aware module). Typical command example: {"regionID":"Experimental Chamber - Plant Cultivation Area (Region 2)", "contentType":"Video + 3D Animation", "contentID":"DC001;DC002", "timestamp":"2024-10-15 10:00:00.500","priority":3}.

[0083] Upon receiving the command, the system first performs integrity checks (checking whether it contains required fields such as "regionID" and "timestamp") and timeliness checks (the difference between the current time and the command timestamp should be ≤500ms to avoid delayed commands causing misaligned actions). If the checks pass, the system proceeds to the parsing stage. If the checks fail (e.g., fields are missing or timeout), a "command resend request" is sent to the behavior awareness module to ensure the command is valid.

[0084] Target function partition location analysis: The module has a built-in "Region ID - Physical Coordinates" mapping table. This table is constructed based on the partitioning structure blueprint of the scenario-based entity module (output of step two), converting "regionID" into three-dimensional coordinates in the model coordinate system (with the model center as the origin, the X-axis along the experimental module-living module direction, the Y-axis perpendicular to the X-axis, and the Z-axis vertical). For example: The target coordinates for “Experimental Module - Plant Cultivation Area (Area 2)” are (X=150mm, Y=0mm, Z=200mm), where X=150mm represents 150mm from the origin along the positive X-axis (center of the plant cultivation area in the experimental module), Y=0mm represents the model's symmetry plane, and Z=200mm represents 200mm from the bottom of the model (the height the lifting platform needs to reach to ensure the simulated astronaut model is aligned with the plant cultivation module); During the analysis process, coordinate deviations need to be eliminated. For example, by comparing the coordinates with the blueprint of the model partition structure, the ±5mm deviation caused by installation error is corrected, and the precise target coordinates (X=150mm, Y=0mm, Z=200mm) are finally determined.

[0085] Platform movement path generation: The path generation adopts the A* (A-Star) path planning algorithm, which finds the optimal path by combining "estimated cost + actual cost" to balance efficiency and safety (avoiding fixed structures within the model, such as node cabin support columns).

[0086] Initial position: The lifting platform is docked at the bottom of the node compartment by default, with coordinates (X=0mm, Y=0mm, Z=50mm). Path node planning: The algorithm generates 3 key nodes, namely (0,0,50)→(80,0,120)→(150,0,180)→(150,0,200), where node (80,0,120) is the turning point to avoid the node cabin support column, and node (150,0,180) is the pre-deceleration point to approach the target; Path constraint parameters: The maximum moving speed of the lifting platform is set to 5 mm / s (based on a model scale of 1:10, corresponding to the actual space station's moving speed of 0.05 m / s, consistent with the characteristics of slow movement in a microgravity environment), and the acceleration is set to 1 mm / s². 2 (To avoid the simulated astronaut model shaking due to start-stop impact), the turning radius is set to 30mm (to adapt to the steering capability of the lifting platform's mechanical structure).

[0087] Robotic arm operation task generation: Based on the teaching content of the target functional zones (such as "sample handling in the plant cultivation chamber"), generate the operation tasks for the robotic arm. The tasks include three types of information: "action type", "target position", and "action parameters". Action type: "Grab-Transport-Place", that is, the robotic arm grabs the simulated plant sample box (30mm×20mm×15mm, made of ABS plastic) from the top standby position, transports it to the sample slot (32mm×22mm×16mm) in the plant cultivation chamber of the experimental cabin, and completes the placement; Target location: The coordinates of the robotic arm's grasping position (X=150mm, Y=50mm, Z=250mm) (the sample box storage location on the top of the model), and the coordinates of the placement position (X=150mm, Y=0mm, Z=200mm) (the center of the sample tank in the plant cultivation chamber). Action parameters: The joint speed of the robotic arm is set to 30° / s (to ensure smooth movement and prevent the sample box from falling off), the gripping force of the end gripper is set to 1N (to accommodate a sample box weight of 50g; too much gripping force will damage the sample box, while too little force will result in unstable gripping), and the action completion time limit is set to 10 seconds (to match the movement time of the lifting platform to ensure coordination).

[0088] Motion control parameter output: The lifting platform path and the robotic arm task are integrated into structured motion control parameters, which include: Lifting platform parameters: "Initial coordinates (0,0,50), target coordinates (150,0,200), velocity 5mm / s, acceleration 1mm / s²" 2 , path nodes (80,0,120), (150,0,180); Robotic arm parameters: "Motion type: gripping-carrying-placing, gripping coordinates (150,50,250), placement coordinates (150,0,200), joint speed 30° / s, gripping force 1N"; The parameters are output to the lifting platform controller and the robotic arm controller in binary format (for easy hardware parsing), with an output frequency of 10Hz (to ensure that the controller receives updated parameters in real time).

[0089] Based on motion control parameters, the lifting platform is driven to move the simulated astronaut model along the path. The current position is fed back in real time through position sensors. When the target area is reached, a stop signal is triggered and an astronaut arrival signal is output. The lifting platform must strictly follow the motion control parameters. The deviation is corrected through real-time feedback from the position sensor to ensure that the simulated astronaut model can accurately reach the target functional area and avoid alignment failure due to mechanical errors. This is consistent with the design requirement of "lifting platform and application linkage" and realizes closed-loop control of "precise movement - real-time feedback - smooth stop".

[0090] Lifting platform drive components and working principle: The lifting platform adopts a ball screw drive structure (model ZOLIX TGA100, consistent with the lifting platform installed in the scene-based physical module), and the drive components include: Drive motor: 12V DC stepper motor (step angle 1.8°, reduction ratio 1:64, maximum torque 0.5N・m), the rotation angle is controlled by pulse signal, which in turn drives the ball screw to lift and lower; Transmission mechanism: ball screw (5mm lead, C7 precision, ensuring 5mm lifting per revolution without deviation), the screw nut is rigidly connected to the lifting platform plate, and when the motor rotates, the nut drives the plate to move up and down along the screw; Guiding mechanism: Two parallel optical axes (8mm in diameter and 300mm in length) are installed on both sides of the lifting platform to limit the radial sway of the load plate and ensure that the vertical deviation is ≤0.1mm / m during movement.

[0091] Drive process based on control parameters: The module sends pulse signals and direction signals to the lifting platform controller. The frequency of the pulse signal determines the moving speed, and the direction signal determines the moving direction (positive pulses correspond to upward movement, and negative pulses correspond to downward movement). From the initial coordinates (0,0,50) to the node (80,0,120): The controller outputs positive pulses at a frequency of 500Hz (corresponding to a speed of 5mm / s; since the lead screw has a lead of 5mm, a lead of 5mm means the motor lifts and lowers by 5mm per revolution, and a speed of 5mm / s corresponds to 1 revolution / s of the motor; the step angle is 1.8°, and 200 steps are needed per revolution (360 / 1.8); the reduction ratio is 1:64, so 200 × 64 = 12800 pulses are actually needed per revolution, and the pulse frequency is 12800Hz to ensure accurate speed). This drives the lifting platform to move 80mm along the positive X-axis and rise 70mm along the Z-axis, reaching the node (80,0,120). From node (80,0,120) to (150,0,180): Maintain a pulse frequency of 12800Hz, continue to move 70mm along the positive X-axis and rise 60mm along the Z-axis to reach the pre-deceleration node; From (150,0,180) to the target (150,0,200): Reduce the pulse frequency to 6400Hz (speed 2.5mm / s), slowly rise 20mm, and avoid impact when reaching the target.

[0092] Real-time feedback and deviation correction from position sensors: Two position sensors (model E3Z-LS63, photoelectric, detection distance 5-30mm, accuracy ±0.1mm) are installed on the side of the lifting platform, respectively for X-axis and Z-axis position feedback: X-axis sensor: Installed on the side of the lifting platform's carrying plate, aligned with the scale bar in the X-axis direction of the model (accuracy 0.1mm), it collects the X-axis coordinates every 100ms. If the current X=148mm (target 150mm) is collected, there is a 2mm deviation, and the controller immediately increases the number of pulses in the X-axis direction to correct the deviation. Z-axis sensor: Installed at the bottom of the lifting platform, aligned with the Z-axis scale bar, to collect Z-axis coordinates. If the current Z=199mm (target 200mm), and the deviation is 1mm, the controller increases the number of Z-axis pulses to ensure that the final position deviation is ≤0.5mm.

[0093] Stop signal triggering and astronaut arrival signal output: When the position sensor reports X-axis coordinate = 150mm ± 0.5mm, Z-axis coordinate = 200mm ± 0.5mm, and Y-axis coordinate = 0mm ± 0.5mm, the lifting platform is determined to have reached the target zone. Stop signal trigger: The controller sends a "stop pulse" (frequency 0Hz) to the drive motor and cuts off the motor power at the same time. The electromagnetic brake (built into the motor, braking torque 0.3N・m) fixes the position of the lead screw to prevent the lifting platform from sliding down due to gravity. Arrival Signal Generation: Generate an astronaut arrival signal containing "arrival status (success), target coordinates (150,0,200), actual coordinates (150.2,0,199.8), and deviation value (0.2mm,0,0.2mm)". The signal format is a CAN bus frame (ID=0x18FF6001, data field=0x01,0x96,0x00,0xC8,0x00,0x00,0x02,0x02), where 0x96=150mm, 0xC8=200mm, and 0x02 represents a deviation of 0.2mm. The signal is sent to the multi-dimensional linkage module and the robotic arm controller via the CAN bus to inform the lifting platform that it has arrived in position.

[0094] After receiving the astronaut's arrival signal, the robotic arm is synchronously driven to move along a preset trajectory. The joint encoder ensures the accuracy of the movement, simulates the equipment handling or operation process, and outputs a signal that the robotic arm has completed its movement. The robotic arm's movements need to be initiated synchronously after the lifting platform is in place. Through real-time precision control of the joint encoder, it simulates the operation scenarios of real equipment on the space station, such as sample handling and component installation. This process must strictly match the preset trajectory to avoid collisions with the lifting platform or model structure, while ensuring the relevance of the movements to the teaching content. For example, "handling plant samples" corresponds to the popular science theme of "material transfer under microgravity".

[0095] Hardware structure and preset trajectory planning of the robotic arm: The robotic arm is a 6-joint industrial robotic arm (model DOBOT Magician Lite, consistent with the robotic arm mounted on the top of the scene-based physical module). The joints are numbered from the base to the end in the following order: J1 (base rotation), J2 (arm swing), J3 (arm swing), J4 (wrist rotation), J5 (wrist swing), and J6 (end-effector rotation). Each joint is equipped with a joint encoder (resolution 1024 pulses / revolution, accuracy ±0.1°). The preset trajectory is based on the "grab-transport-place" task planning and includes 5 key attitude nodes, each corresponding to the angle values ​​(unit: °) of 6 joints: Standby posture: J1=0, J2=-30, J3=60, J4=0, J5=-30, J6=0 (end point is located at the top of the model, coordinates (150, 50, 250), away from other structures); Grasping posture before grasping: J1=0, J2=-45, J3=75, J4=0, J5=-45, J6=0 (the end descends to 5mm above the sample box, coordinates (150,50,255)); Grasping posture: J1=0, J2=-50, J3=80, J4=0, J5=-50, J6=0 (end attached to the top of the sample box, coordinates (150,50,260)); Handling posture: J1=0, J2=-30, J3=60, J4=0, J5=-30, J6=0 (the end of the sample box rises to a safe height, coordinates (150, 50, 250)). Placement posture: J1=0, J2=-45, J3=75, J4=0, J5=-45, J6=0 (the end descends to 5mm above the sample trough of the plant cultivation chamber, coordinates (150,0,205)); Post-position posture: J1=0, J2=-30, J3=60, J4=0, J5=-30, J6=0 (the end rises to the standby position and completes the action).

[0096] Synchronization drive based on position signal: After receiving the astronaut's arrival signal (ID=0x18FF6001, data field=0x01) via the CAN bus, the robotic arm controller immediately initiates the action sequence, with the following drive logic: From the ready position to the pre-grabbing position: send rotation commands to joints J2, J3, and J5. J2 rotates from -30° to -45° (rotation angle -15°), J3 rotates from 60° to 75° (+15°), and J5 rotates from -30° to -45° (-15°). The joint speed is 30° / s, and the time taken is 0.5 seconds. From the pre-grasping posture to the grasping posture: J2 continues to rotate to -50° (-5°), J3 to 80° (+5°), J5 to -50° (-5°), and at the same time sends a "grasping command" to the end gripper (driving the gripper motor to rotate and generate a 1N gripping force). After contacting the sample box, the gripper position sensor (built-in, accuracy 0.1mm) reports "grasping successful", taking 0.3 seconds. From grasping posture to transport posture: J2 rotates from -50° back to -30° (+20°), J3 rotates from 80° back to 60° (-20°), and J5 rotates from -50° back to -30° (+20°), raising the sample box to a height of 250mm in 0.7 seconds; From transport posture to placement posture: J1 remains at 0°, J2 rotates to -45° (-15°), J3 to 75° (+15°), J5 to -45° (-15°), while J4 rotates to 0° (no need to adjust direction), moving the sample box above the sample slot, taking 0.5 seconds; From placement posture to post-placement posture: Send a "release command" to the gripper (gripping force drops to 0N), the sample box falls into the sample slot, and then J2, J3, and J5 rotate back to the standby angle, taking 0.5 seconds; The entire process took a total of 2.5 seconds, which meets the requirement of "10-second completion time limit" and is completely synchronized with the time node after the lifting platform is in place (the lifting platform starts immediately after it is in place, without delay).

[0097] Precision control and collision avoidance of joint encoders: The encoder at each joint acquires the joint angle in real time (sampling frequency 100Hz), compares it with the angle value of the preset trajectory, and corrects deviations. If the preset angle of joint J2 is -45°, and the encoder acquires an actual angle of -44.5° (deviation +0.5°), the controller immediately sends a "correction pulse" to the joint motor to reduce the angle to -45°, ensuring that the joint angle deviation is ≤0.5°; Collision avoidance: The robotic arm controller has a built-in 3D coordinate library of the model structure. If it is calculated that the end of a joint will approach the model support column when it rotates (distance ≤ 5mm), the action will be paused immediately and the trajectory will be finely adjusted (e.g., the J1 joint will rotate 5° to avoid the support column) to ensure the safety of the action.

[0098] Robotic arm completes action signal output: The action is considered complete when the robotic arm returns to the standby position, and the gripper release signal and the sample slot position sensor (installed inside the sample slot in the plant cultivation chamber to detect whether the sample box is in place) report "sample placed". Completion signal generation: The robotic arm action completion signal includes "Action status (success), Task type (grasping-carrying-placing), Action time (2.5 seconds), Maximum joint deviation (0.3°)", and is in CAN bus frame format (ID=0x18FF6002, data field=0x01,0x03,0x09,0x50,0x00,0x00,0x03,0x00), where 0x03 represents the task type, and 0x0950=2384ms≈2.5 seconds; Signal transmission: The signal is sent to the multi-dimensional linkage module and dynamic projection module via the CAN bus to inform the robotic arm that the action has been completed, thus preparing for the subsequent simulation of component state changes and projection synchronization.

[0099] By integrating the astronaut arrival signal and the robotic arm action completion signal, the system triggers the state changes of the corresponding simulated operation components and sends synchronous commands to the dynamic projection module, enabling the projected content to match the physical actions in real time and generating a scenario-based interactive display effect for the space classroom.

[0100] This step is the final link in system collaboration. It requires integrating the preceding hardware action signals, driving the simulation operation components to move in tandem, and updating the projected content synchronously. This achieves a high degree of coordination among "physical actions - component status - projected image", creating an immersive scene-based effect, solving the problem of "lack of in-depth interactive experience", and allowing the audience to intuitively experience the operation process and scientific knowledge of the space classroom.

[0101] Signal integration and trigger condition determination: The multi-dimensional linkage module receives and analyzes the astronaut's arrival signal (ID=0x18FF6001) and the robotic arm's action completion signal (ID=0x18FF6002), and integrates the signals: Integration logic: Check whether the "status" field of both signals is "success" and whether the timestamp difference is ≤1 second (to ensure action synchronization). If both are met, the trigger condition is determined to be met. If the status of a signal is "failure" (such as failure of robotic arm grasping), send an "action retry instruction" to re-execute steps two and three, retrying up to 3 times. If it still fails, output an "action abnormality" prompt. Example: Astronaut arrival signal timestamp = 10:00:00.500, robotic arm completion signal timestamp = 10:00:02.800, difference of 2.3 seconds (because the robotic arm action takes 2.5 seconds, which is in line with expectations), and both statuses are "successful", so the trigger condition is determined to be met.

[0102] Simulation triggering of state changes in operating components: Based on the target functional zoning (experimental chamber - plant cultivation area), trigger the state changes of the corresponding simulation operation components. The state changes of the components must match the robotic arm actions (placing sample boxes) and the teaching content (plant sample detection). Status changes of the OLED screen in the plant cultivation chamber: The module sends a "content switching command" (via 485 communication, baud rate 9600bps) to the control board (model Arduino Nano) of the plant cultivation chamber. The command content is "switch to sample detection interface". After receiving the command, the control board drives the 2.4-inch OLED screen to switch from the "plant growth cycle" screen to the "sample box detection data" screen (displaying "sample temperature 25℃, humidity 60%, light intensity 500 lux", simulating the plant cultivation environment parameters of the space station). Sample cell indicator light status change: A red LED indicator light (0.5W power, 5V voltage) is installed inside the sample cell. When the module sends a "light-up command", the control board drives the indicator light to change from "off" to "always on" (brightness 50cd / m²). 2 This visually indicates that the sample has been placed in place; Force feedback fixed handle (living module) status change: Since the current target area is the experimental module, the fixed handle of the living module will not trigger a status change. If the target area is the living module, the force feedback motor driving the handle will generate 1N reverse resistance to simulate the astronaut's gripping action.

[0103] Synchronization command sent and projected content matched: The module sends synchronization commands to the dynamic projection module to ensure that the projected content matches the actual actions (lifting platform positioning, robotic arm placing the sample, component state changes) in real time. The command format is JSON. {"syncType":"action-content","actionState":"done","targetRegion":"Experimental Chamber - Plant Cultivation Area","contentUpdate":"Sample Detection Data + Indicator Light On Animation","timestamp":"2024-10-15 10:00:03.000"} After receiving the instruction, the dynamic projection module performs the following operations: Content switching: Switch from the original "Plant Growth Video" (DC001) to "Sample Detection 3D Animation" (DC002). The animation content is "robotic arm places sample box → sample slot indicator light lights up → OLED screen displays data", which is completely synchronized with the physical action; Image alignment adjustment: Due to changes in the status of physical components (indicator lights on), the brightness of the corresponding area of ​​the projected image needs to be adjusted. The brightness of the sample slot projection area should be adjusted from 300 cd / m². 2 Increased to 350 cd / m 2 This highlights the effect of the indicator light being lit, while the scene anchor point matching algorithm corrects the image deviation (ensuring that the projected sample slot is aligned with the physical sample slot, with a deviation of ≤1mm). Timing Matching: The keyframes of the projection animation strictly correspond to the timing of the entity's actions. For example, the robotic arm places the sample box (t=10:00:02.8) → the projection animation shows the placement action (t=10:00:02.8) → the entity's indicator light turns on (t=10:00:02.9) → the projection animation shows the indicator light turning on (t=10:00:02.9) → the OLED screen switches (t=10:00:03.0) → the projection displays the data screen (t=10:00:03.0). The time difference is ≤100ms, with no obvious delay.

[0104] Generation of contextualized interactive display effects: The final generated scene-based effect involves collaboration across four dimensions: Physical actions: The lifting platform, carrying the simulated astronaut model, stops next to the plant cultivation area of ​​the experimental module (Z=200mm), the robotic arm returns to the standby position, and the sample box is located in the sample slot; Component status: The OLED screen displays sample test data, and the sample slot indicator light remains constantly on; Projected screen: Dynamically displays a 3D animation of "sample placement-detection", with the image precisely aligned with the physical components and the brightness highlighting key areas; Teaching guidance: The projected screen simultaneously plays audio narration ("Now we see that the robotic arm has placed the plant sample into the culture chamber. The indicator light in the sample slot is lit up to indicate that the sample is in place. The screen displays the current culture environment parameters of the sample"), complementing the visual effects; Standing in front of the model, visitors can directly see the complete process of "astronaut arrival → robotic arm handling → component response → projection explanation," transforming abstract space classroom knowledge into a concrete interactive scenario. This achieves a deep integration of "operation-display-popularization," meeting the core requirement of "enhancing the interactive experience of space classrooms."

[0105] Another embodiment of the present invention provides a space station simulation interactive display method for use in space classrooms, see [link to relevant documentation]. Figure 2 The method may include: S201: Construct a scaled-down model of the space station's functional zones. The model's interior is equipped with simulation operation components corresponding to the teaching content. The node module contains a simulated astronaut model with a lifting platform, and the top is fitted with a robotic arm that can simulate equipment handling, forming a display carrier with functional interactive attributes. S202 generates teaching content images adapted to the model structure. Through cross projection by dual projectors, the projected images are aligned with the model's functional areas in real time using a scene anchor point matching algorithm. When switching teaching themes, the scaling ratio and content focus are automatically adjusted to generate dynamic projection content that is linked with the functions of the entity modules. S203 captures audience interaction behavior, identifies the audience's standing area and gaze direction through a microwave radar array, and judges the interaction intention based on the duration. When the audience focuses on a certain functional area for more than a preset time, it triggers the display of in-depth teaching content corresponding to that area and generates targeted interactive response instructions. S204 coordinates the coordinated actions of various components. Based on interactive response commands, it controls the lifting platform to move the simulated astronaut model to the corresponding functional area, and synchronously drives the robotic arm to simulate operation demonstration. This allows the state changes and mechanical actions of the simulated operation components to match the dynamic projection content in real time, generating a scenario-based interactive display effect for the space classroom.

[0106] Figure 3 This is a hardware structure block diagram of a computer terminal for a secure communication method for big data provided in an embodiment of the present invention. (See diagram below.) Figure 3 As shown, the computer device includes a processor, memory, and network interface connected via a system bus, wherein the memory may include non-volatile storage media and internal memory.

[0107] Non-volatile storage media can store operating systems and computer programs. These computer programs include program instructions that, when executed, cause the processor to perform any method of secure communication involving large amounts of data.

[0108] The processor provides computing and control capabilities, supporting the operation of the entire computer device.

[0109] Internal memory provides an environment for the execution of computer programs stored in non-volatile storage media. When these computer programs are executed by a processor, the processor can perform any method of secure communication for large amounts of data.

[0110] This network interface is used for network communication, such as sending assigned tasks. Those skilled in the art will understand that... Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0111] It should be understood that the processor can be a Central Processing Unit (CPU), but it can also 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. Among these, a general-purpose processor can be a microprocessor or any conventional processor.

[0112] This invention also provides a storage medium storing a computer program, wherein the computer program is configured to execute the steps in any of the above method embodiments when running.

[0113] This invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

[0114] Specifically, the aforementioned electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the aforementioned processor, and the input / output device is connected to the aforementioned processor.

[0115] The above description, based on the embodiments shown in the figures, details the structure, features, and effects of the present invention. The above description is only a preferred embodiment of the present invention, but the present invention is not limited to the scope of implementation shown in the figures. Any changes made in accordance with the concept of the present invention, or equivalent embodiments modified to have equivalent changes, that do not exceed the spirit covered by the specification and figures, should be within the protection scope of the present invention.

Claims

1. A space station simulation interactive display system for use in space classrooms, characterized in that, The system includes: a scenario-based entity module, a dynamic projection module, a behavior perception module, and a multi-dimensional linkage module; The scenario-based entity module is used to construct a scaled-down model of the space station's functional zones. The model's interior is equipped with simulation operation components corresponding to the teaching content. The node module is equipped with a simulated astronaut model carrying a lifting platform, and the top is fitted with a robotic arm that can simulate equipment handling, forming a display carrier with functional interactive attributes. The dynamic projection module is used to generate teaching content images that adapt to the model structure. It uses dual projectors to project the image crosswise and employs a scene anchor point matching algorithm to align the projected image with the model's functional areas in real time. When switching teaching topics, it automatically adjusts the image scaling ratio and content focus to generate dynamic projection content that is linked to the functions of the physical module. The behavior perception module is used to capture audience interaction behavior. It identifies the audience's standing area and gaze direction through a microwave radar array, and judges the interaction intention based on the duration. When the audience focuses on a certain functional area for more than a preset time, it triggers the display of in-depth teaching content corresponding to that area and generates targeted interactive response instructions. The multi-dimensional linkage module is used to coordinate the collaborative actions of various components. Based on interactive response commands, it controls the lifting platform to move the simulated astronaut model to the corresponding functional area, and synchronously drives the robotic arm to simulate operation demonstration. This allows the state changes and mechanical actions of the simulated operation components to match the dynamic projection content in real time, generating a scenario-based interactive display effect for the space classroom.

2. The system according to claim 1, characterized in that, The scenario-based entity module is specifically used for: Based on the actual structural parameters of the space station, the core functional areas are divided according to a preset ratio, the size, relative position and connection method of each area are determined, and the structural blueprint of the area is output. Create a solid model framework based on the partition structure blueprint, reserve installation interfaces for simulation operation components at corresponding positions inside the model, and output a model framework with interfaces. Install simulation operation components: assemble a triggerable display plant cultivation chamber at the experimental chamber interface, install a force feedback fixed handle at the living chamber interface, and output the model of the basic assembly components. A lifting platform carrying a simulated astronaut model is installed in the node module. A multi-jointed robotic arm is assembled on top of the model. All components are integrated and the interactive functions are tested to form a display carrier with functional interactive attributes.

3. The system according to claim 2, characterized in that, The dynamic projection module is specifically used for: Based on the teaching focus of each functional area of ​​the space station, a multi-theme teaching content library was created, with each theme content associated with a corresponding area tag, and a structured teaching content library was output. Install the two projectors at the preset positions on the outside of the model, adjust the projection angle so that the image overlaps and covers the surface of the model, calibrate the brightness and color consistency of the two projectors using a grayscale test card, and output the calibrated dual projector system. Physical anchor points are set at the edges of each functional partition of the model. The scene anchor point matching algorithm is used to identify the anchor point position, calculate the geometric deviation between the projected image and the partition, and adjust the output parameters of the projector in real time to make the image and the partition outline accurately aligned, and output the aligned projected image. When a theme switching command is received, the system automatically calls up the teaching content of the corresponding theme, adjusts the screen scaling ratio according to the target partition size, and updates the anchor point matching parameters in sync to maintain alignment, generating dynamic projection content that is linked with the functions of the physical module.

4. The system according to claim 3, characterized in that, The behavior perception module is specifically used for: A microwave radar array is distributed in a fan shape around the model, dividing it into multiple monitoring areas. The radar sampling frequency is configured to capture the positional movement and micro-movement signals of the audience, and output the completed radar perception system. The radar sensing system collects signals in real time, identifies the standing area of ​​the audience through signal strength analysis, determines the gaze direction by combining the signal Doppler effect, and outputs raw data of audience behavior. Perform time-series analysis on raw audience behavior data, calculate the audience's dwell time and gaze duration in a certain functional area, and when the duration exceeds a preset value, determine that there is a deep interaction intention in that area, and output the intention judgment result; Based on the intent judgment result, the corresponding in-depth teaching content of the functional area is triggered, and an interactive response instruction containing area identifier and content type is generated.

5. The system according to claim 4, characterized in that, The multi-dimensional linkage module is specifically used for: Receive interactive response commands, parse the target functional area location in the commands, generate the lifting platform movement path and the robotic arm operation task, and output motion control parameters; Based on motion control parameters, the lifting platform is driven to move the simulated astronaut model along the path. The current position is fed back in real time through position sensors. When the target area is reached, a stop signal is triggered and an astronaut arrival signal is output. After receiving the astronaut's arrival signal, the robotic arm is synchronously driven to move along a preset trajectory. The joint encoder ensures the accuracy of the movement, simulates the equipment handling or operation process, and outputs a signal that the robotic arm has completed its movement. By integrating astronaut arrival signals and robotic arm action completion signals, the system triggers state changes of corresponding simulated operation components and simultaneously sends synchronization commands to the dynamic projection module, enabling real-time matching of projected content with physical actions and generating a scenario-based interactive space classroom display effect.

6. A method for interactive simulation of a space station in a space classroom, characterized in that, The method includes: Construct a scaled-down model of the space station's functional zones, with simulation operation components corresponding to the teaching content set up inside the model. The node module is equipped with a simulated astronaut model carrying a lifting platform, and the top is fitted with a robotic arm that can simulate equipment handling, forming a display carrier with functional interactive attributes. The system generates teaching content images that are adapted to the model structure. It uses dual projectors to project the images crosswise and employs a scene anchor point matching algorithm to align the projected images with the model's functional areas in real time. When switching teaching topics, it automatically adjusts the image scaling and content focus to generate dynamic projection content that is linked to the functions of the entity modules. Capture audience interaction behavior, identify the audience's standing area and gaze direction through a microwave radar array, and judge the interaction intention based on the duration. When the audience focuses on a certain functional area for more than a preset time, trigger the display of the corresponding in-depth teaching content in that area and generate targeted interactive response instructions. By coordinating the coordinated actions of various components and controlling the lifting platform to move the simulated astronaut model to the corresponding functional area based on interactive response commands, the robot arm is simultaneously driven to simulate operation demonstrations. This allows the state changes and mechanical movements of the simulated operation components to match the dynamic projection content in real time, generating a scenario-based interactive display effect for the space classroom.

7. The method according to claim 6, characterized in that, The construction of a scaled-down model of the space station's functional zones includes simulated operation components corresponding to the teaching content on the inner side of the model, a simulated astronaut model with a lifting platform inside the node module, and a robotic arm on top that can simulate equipment handling, forming a display carrier with functional interactive attributes, including: Based on the actual structural parameters of the space station, the core functional areas are divided according to a preset ratio, the size, relative position and connection method of each area are determined, and the structural blueprint of the area is output. Create a solid model framework based on the partition structure blueprint, reserve installation interfaces for simulation operation components at corresponding positions inside the model, and output a model framework with interfaces. Install simulation operation components: assemble a triggerable display plant cultivation chamber at the experimental chamber interface, install a force feedback fixed handle at the living chamber interface, and output the model of the basic assembly components. A lifting platform carrying a simulated astronaut model is installed in the node module. A multi-jointed robotic arm is assembled on top of the model. All components are integrated and the interactive functions are tested to form a display carrier with functional interactive attributes.

8. The method according to claim 7, characterized in that, The generated teaching content screen adapted to the model structure is projected through dual projectors and uses a scene anchor point matching algorithm to align the projected screen with the model's functional areas in real time. When switching teaching topics, the screen scaling and content focus are automatically adjusted to generate dynamic projection content that is linked to the functions of the entity modules, including: Based on the teaching focus of each functional area of ​​the space station, a multi-theme teaching content library was created, with each theme content associated with a corresponding area tag, and a structured teaching content library was output. Install the two projectors at the preset positions on the outside of the model, adjust the projection angle so that the image overlaps and covers the surface of the model, calibrate the brightness and color consistency of the two projectors using a grayscale test card, and output the calibrated dual projector system. Physical anchor points are set at the edges of each functional partition of the model. The scene anchor point matching algorithm is used to identify the anchor point position, calculate the geometric deviation between the projected image and the partition, and adjust the output parameters of the projector in real time to make the image and the partition outline accurately aligned, and output the aligned projected image. When a theme switching command is received, the system automatically calls up the teaching content of the corresponding theme, adjusts the screen scaling ratio according to the target partition size, and updates the anchor point matching parameters in sync to maintain alignment, generating dynamic projection content that is linked with the functions of the physical module.

9. A storage medium, characterized in that, The storage medium stores a computer program, wherein the computer program is configured to execute the method of any one of claims 6-8 when it is run.

10. An electronic device comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to run the computer program to perform the method of any one of claims 6-8.

Images