Group exercise formation intelligent marker based on dynamic light positioning

Abstract

The invention discloses a group exercise formation intelligent marker based on dynamic light positioning, and relates to the technical field of sports performance auxiliary equipment, the system is composed of a plurality of function modules, and the system comprises a sensing execution module which takes a wearable device as a carrier, integrates LED and UWB positioning functions, and performs positioning by sending a dynamic light signal; after the receiving unit receives the instruction, a light spot is projected to guide an actor to operate; when the position of the actor deviates, timely feedback is performed through the vibration motor; the data fusion module is used for acquiring related position data on the basis of the data acquired by the sensing execution module, performing fusion processing by applying a Kalman filtering algorithm, and outputting real-time coordinates; the wireless networking module is used for constructing a Mesh network according to the real-time coordinates and carrying out efficient transmission of positioning and control instructions; and the edge control module is used for processing the positioning and control instruction transmitted by the wireless networking module by utilizing the parallel processing capability of an FPGA (Field Programmable Gate Array).

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of sports performance auxiliary equipment, in particular to a group exercise formation intelligent identifier based on dynamic light positioning. BACKGROUND

[0002] Group exercises form dynamic patterns (such as characters, emblems, natural landscapes, etc.) through the spatial arrangement of performers. Precise formation is the basis for realizing artistic creativity. The choreography of group exercises is closely coordinated with the rhythm of music and the effect of lighting. Precise formation can ensure the synchronization of performer movements. In the face of large-scale performances with hundreds to thousands of performers, precise control of formation is the key to maintaining order on site. By pre-setting positions and movement paths, problems such as performer collisions and disordered walking can be avoided, especially in complex stage scheduling (such as multiple performers crossing), which requires positioning technology to achieve dynamic path planning to ensure safe and orderly performances.

[0003] Traditional group exercise formation management relies on manual rehearsals and performer experience, which has obvious limitations. In terms of positioning, manual or simple marking has low precision, and it is difficult to determine the real-time position of performers in large-scale performances, which can easily lead to formation chaos. Lighting control and positioning are separate, only serving as background decoration, and cannot be adjusted in real time according to performer positions, making it difficult to effectively guide and feedback deviations. In complex environments, the reliability and precision of positioning are significantly reduced, and the system deployment and debugging are tedious, requiring a large amount of manual intervention. Although there are positioning technologies based on sensors, in the context of group exercises, the data fusion efficiency, communication delay, and collaborative control are insufficient, and there is an urgent need for an intelligent identification system with high-precision positioning, dynamic light guidance, and low-delay collaborative control. SUMMARY

[0004] To achieve the above purpose, the present application realizes the following technical solutions:

[0005] The perception and execution module: taking wearable devices as carriers, integrating LED and UWB positioning functions, positioning through sending dynamic light signals; after receiving instructions at the receiving unit, projecting light spots to guide performer operations; when there is a deviation in the position of the performer, timely feedback through the vibration motor;

[0006] The data fusion module: on the basis of data obtained by the perception and execution module, collecting relevant position data, using Kalman filtering algorithm for fusion processing, and outputting real-time coordinates;

[0007] The wireless networking module: according to the real-time coordinates, constructing a Mesh network for efficient transmission of positioning and control instructions;

[0008] The edge control module: using the parallel processing capability of FPGA to process the positioning and control instructions transmitted by the wireless networking module; generating projection parameters and developing a hierarchical early warning strategy, while linking the cloud path optimization algorithm for low-delay collaborative control.

[0009] Further, the process of positioning by sending dynamic light signals is:

[0010] Taking the wearable device as the carrier, the device integrates a micro LED array and a UWB tag, the LED array flashes at a unique frequency within a certain range as an optical positioning active marker, and the UWB tag communicates with the base station to provide centimeter-level position data, and the positioning is realized by combining the data.

[0011] Further, the process of projecting the light spot to guide the actor operation is:

[0012] The real-time coordinates of the actor are output by a positioning fusion algorithm based on Kalman filtering fusion of UWB, IMU and LiDAR data; a dynamic projection planning module calculates the target position according to a preset formation program, generates light identification parameter control documents, and generates instructions accordingly; the execution layer laser projector projects light spots according to the instructions, and the adaptive light adjustment module adjusts the brightness, thereby realizing guidance and feedback of the execution state.

[0013] Further, the process of outputting real-time coordinates is:

[0014] The control layer uses a multi-sensor fusion strategy to dynamically allocate weights of optical, UWB, IMU, LiDAR and other sensors according to the motion state, adopts robust Kalman filtering to remove abnormal data, and outputs real-time coordinates through Kalman filtering algorithm fusion processing.

[0015] Further, the specific process of constructing a Mesh network is:

[0016] A wireless Mesh ad hoc network of mesh topology is constructed, and each wearable device acts as a node that can receive and forward data.

[0017] Further, the process of efficient transmission of positioning and control instructions is:

[0018] A low-delay communication protocol, a multiple access protocol combining time division multiple access and carrier sense multiple access, and a dynamic routing protocol are used to select the optimal path according to the node state and communication quality.

[0019] Further, the process of processing positioning and control instructions is:

[0020] An edge computing platform with FPGA as the core, combined with an AI coprocessor, realizes data processing and instruction interaction through a specific software architecture.

[0021] Further, the process of the hierarchical early warning mechanism is:

[0022] When the deviation is less than or equal to the distance y, only the vibration motor of the wearable device is used for reminding.When the deviation is between the distance y and the distance y1, the vibration reminding and the local light flickering are triggered.When the deviation is greater than the distance y1, the global red light warning is started, and the alarm signal is sent to the command end.

[0023] Further, the process of low-delay cooperative control is:

[0024] Based on the edge computing platform, real-time positioning data of the perception layer is received and fused to obtain real-time coordinates and deviation values of the actors, a corresponding hierarchical early warning mechanism is triggered according to the deviation values, projection control instructions are generated to adjust the spot shape and color parameters, and the actor position data and deviation information are sent to the cloud layer for generating a three-dimensional heat map and VR training simulation and other training aids and data analysis.

[0025] The group exercise formation intelligent identifier based on dynamic light positioning provided by the application has the following beneficial effects:

[0026] (1) The application uses a multi-sensor fusion strategy (optical, UWB, IMU, LiDAR) and a Kalman filter algorithm, and the positioning accuracy in a static scene is ±3cm and the positioning accuracy in a dynamic scene is ±5cm, which is significantly better than traditional single-sensor technology.Meanwhile, the system can dynamically adjust the sensor weight according to the motion state (such as increasing the IMU weight to 50% when moving at high speed), and combine the robust filter to remove abnormal data such as occlusion, so as to ensure the positioning reliability in a complex performance environment (such as fast movement and occlusion scene), and solve the problems of low positioning accuracy and poor environmental adaptability of traditional methods.

[0027] (2) The perception execution module integrates LED and UWB positioning through the wearable device, and combines the dynamic projection planning of the edge control module to realize the closed-loop control of "positioning-projection-feedback".For example, the system generates a light spot to guide the actor to walk (a green circle represents a target point, and a red arrow indicates a direction), and automatically enlarges the light spot radius when the deviation is greater than 30cm, and real-time feedback is provided through the vibration motor.In addition, the adaptive light adjustment module can automatically adjust the projection brightness according to the ambient light (0.1-2000lux) to keep the light identification highly visible in different scenes, and break through the limitation of traditional separation of light and positioning.

[0028] (3)The application constructs a wireless Mesh ad hoc network by adopting ZigBee 3.0, supports dynamic access of 500+ wearable devices, does not need to pre-deploy base stations, and can automatically complete network configuration and sensor calibration through a self-organizing algorithm in an outdoor site, so that rapid adaptive deployment is realized. Meanwhile, the edge computing platform utilizes the parallel processing capability of FPGA, in combination with the IEEE 1588 clock synchronization protocol (error ≤1 ms), so that the data transmission delay is ≤50 ms, the cooperative control algorithm delay is <10 ms, and the low-delay requirement of large-scale performances is met. In addition, the three-dimensional heat map and the VR training system (shortening the rehearsal period by more than 50%) assist the targeted training of actors, and further improve the rehearsal efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0029] Figure 1 It is a system flowchart of the application. DETAILED DESCRIPTION

[0030] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the application.

[0031] EMBODIMENT

[0032] Please refer to the drawings, the embodiment of the application provides a group exercise formation intelligent identifier based on dynamic light positioning, and the system comprises:

[0033] A perception and execution module: taking a wearable device as a carrier, integrating LED and UWB positioning functions; positioning is performed by sending dynamic light signals; after receiving a command at a receiving unit, a light spot is projected to guide personnel operation; when there is a deviation, a vibration motor is used for timely feedback;

[0034] Wearable device:

[0035] Mainly composed of a lightweight positioning terminal, including a micro LED array (embedded in a shoe pad or a wristband), a UWB tag, an IMU sensor (a three-axis gyroscope and an accelerometer), and a vibration motor.

[0036] Function implementation: the micro LED array flashes at a specific frequency, serving as an active marker for optical positioning, and its flashing frequency coding adopts a unique frequency allocation scheme, such as different frequencies in the range of 10 Hz to 50 Hz, to avoid interference between multiple devices, and is denoted as:

[0037] f i ∈[10,50]Hz

[0038] Wherein, i represents different wearable devices.

[0039] UWB tags communicate with base stations to provide centimeter-level location data using ultra-wideband technology, with positioning accuracy of ±3 cm in static scenarios and ±5 cm in dynamic scenarios.

[0040] IMU sensors collect motion trajectories in real-time to compensate for transient errors in optical and UWB positioning. By integrating acceleration and angular velocity, the device's motion state parameters are obtained.

[0041] By sending dynamic light signals:

[0042] Composed of a multi-modal sensor network and an optical positioning grid, including high-density optical sensors deployed on the stage floor, lightweight LiDAR on the stage edge, and vision cameras on the top.

[0043] Function implementation: Through high-density optical sensors, the flashing frequency and position of LEDs are captured to achieve centimeter-level positioning with an accuracy of ±5 cm. Lightweight LiDAR scans actor profiles to assist in positioning actor positions in occluded areas, obtaining three-dimensional profile information of actors through laser radar point cloud data. Vision cameras capture global formation images to verify the consistency of multi-sensor data and perform global monitoring of the entire stage formation through image processing algorithms.

[0044] Through the perception layer, the collected LED signals, UWB data, LiDAR point cloud, and vision images are preprocessed to extract key features, and these data are transmitted to the control layer for fusion processing.

[0045] Projecting light spots to guide actor operations:

[0046] Based on an edge computing platform, the positioning fusion algorithm and dynamic projection planning unit are integrated. Through the positioning fusion algorithm based on Kalman filtering, UWB, IMU, and LiDAR data are fused to output real-time coordinates of actors, with a data output frequency of 100 Hz and a delay of less than 10 ms. Its mathematical model can be represented as:

[0047] X k =F k X k-1 +K k (Z k -H k X k-1 )

[0048] Where X k is the state vector at time k (including position and velocity), F k is the state transition matrix, K k is the Kalman gain, Z k is the sensor measurement, and Hk is the observation matrix. The dynamic projection planning module calculates the next target position of the actor according to the preset formation program, and generates the corresponding light identifier, including the parameters such as spot shape and color. Decision control: the control layer generates projection control instructions and feedback warning instructions according to the fused positioning data and the preset formation scheme, and sends them to the execution layer and the wearable device layer.

[0049] It mainly includes a dynamic projection device and an adaptive dimming module. The dynamic projection device uses a laser projector to project dynamic spots on the stage floor according to the control instructions, with a resolution of 1080p and a response time less than or equal to 20ms. The adaptive dimming module automatically adjusts the projection brightness according to the ambient light intensity, supporting scenes from 0.1 to 2000 lux, and its brightness adjustment formula is:

[0050]

[0051] where L is the adjusted projection brightness, L0 is the initial brightness, I env is the ambient light intensity, I ref is the reference light intensity. The execution layer receives the projection instructions from the control layer, accurately controls the working state of the projector, realizes the projection of dynamic spots, and feeds back the execution state to the control layer.

[0052] Through the vibration motor:

[0053] The vibration motor receives the instructions from the control layer and triggers the tactile warning when the actor deviates from the target position, realizing real-time feedback.

[0054] Data interaction: the wearable device layer sends LED flashing signals and UWB positioning signals to the perception layer, and receives feedback control instructions from the control layer.

[0055] Data fusion module: based on the data obtained by the perception and execution module, relevant position data is collected, and Kalman filtering algorithm is used for fusion processing to output real-time coordinates;

[0056] Collect relevant position data:

[0057] Multi-sensor data fusion strategy, through sensor weight distribution, dynamically adjust the weight of each sensor according to the motion state. In static scenes, the weight of optical positioning accounts for 70%, and the weight of UWB accounts for 30%; in high-speed motion scenes, the weight of IMU is increased to 50%, and LiDAR supplements contour data. Its weight distribution function can be expressed as:

[0058] The weight of optical positioning is 0.7 in static scenes and 0.3 in high-speed motion scenes;

[0059] The weight of UWB is 0.3 in static scenes and 0.2 in high-speed motion scenes;

[0060] IMU has a weight of 0.0 in static scenes and 0.5 in high-speed motion scenes;

[0061] LiDAR has a weight of 0.0 in static scenes and 0.3 in high-speed motion scenes.

[0062] And perform abnormal data rejection, using robust Kalman filter algorithm, automatically reject abnormal positioning data, such as LiDAR noise or UWB jump data caused by temporary occlusion. By introducing robust factor, the abnormal data is weighted and processed, reducing its impact on the positioning result.

[0063] Output real-time coordinates:

[0064] Through the above multi-sensor fusion strategy, the positioning accuracy of the system in static scenes can reach ±3cm, and in dynamic scenes it is ±5cm, meeting the demand of high-precision positioning for group performance. Compared with traditional single sensor positioning technology, this fusion positioning technology significantly improves the reliability and accuracy of positioning, especially in complex performance environments, such as scenes with occlusion or rapid motion.

[0065] Different types of markers are distinguished by color and shape, green circles represent target points, red arrows represent moving directions, and yellow areas represent dangerous avoidance zones, intuitively conveying instructions for formation transformation to performers, improving their understanding and response speed to formation adjustment.

[0066] Each marker type has specific parameters, such as spot radius, color brightness, arrow length, etc. The initial value of the target point circle radius is 30cm, and when the performer deviates from the target position by more than 30cm, the circle radius automatically increases to 50cm, and its adjustment formula is: r = {30, 50, deviation ≤ 30cm deviation > 30cm where r is the circle radius.

[0067] Perform brightness self-adaptive adjustment, automatically adjust the projection brightness according to the intensity of the ambient light, support 0.1 to 2000 lux scene, ensure that the projection marker has good visibility under different lighting conditions.

[0068] According to the position deviation of the performer, dynamically adjust the shape and range of the projected marker. When the deviation is small, the projected marker maintains normal size and shape; when the deviation is large, the projection range is appropriately expanded or the shape is adjusted to enhance the performer's perception of the deviation.

[0069] The position data of the actors is collected by the perception layer, and after being processed by the fusion positioning algorithm of the control layer, the real-time coordinates of the actors are obtained. According to the preset formation scheme and real-time coordinates, the control layer calculates the next target position of the actors and generates the corresponding projection identification instruction. After receiving the instruction, the execution layer controls the projector to project dynamic light spots to guide the actors to move. At the same time, the perception layer continuously monitors the actual position of the actors. If a position deviation is detected, the control layer triggers the vibration reminder of the wearable device and dynamically adjusts the projection identification, forming a complete closed-loop control process. The mathematical model of closed-loop control can be expressed as:

[0070]

[0071] where u k is the control output (such as the adjustment parameter of the projection identification) at time k, e k is the position deviation at time k, K p , K i , and K d are the proportional, integral, and derivative gain coefficients, respectively. Through real-time feedback and control of the position deviation, the position of the actors is accurately adjusted to ensure the accuracy of the formation.

[0072] Wireless networking module: according to the real-time coordinates, a Mesh network is constructed for efficient transmission of positioning and control instructions;

[0073] Constructing a Mesh network:

[0074] A ZigBee 3.0 protocol is used to construct a wireless Mesh ad hoc network, and the network topology is a mesh topology. Each wearable device acts as a network node, which can both send and receive data, and also act as a relay node to forward data. This topology structure has good fault tolerance and scalability. When a node fails, data can be transmitted through other nodes to ensure the stability of the network.

[0075] The network architecture supports dynamic access of 500+ wearable devices without the need for pre-deployed fixed base stations, greatly improving the flexibility and adaptability of the system.

[0076] Efficient transmission of positioning and control instructions:

[0077] A low-latency communication protocol is used to ensure the real-time nature of data transmission. The delay time of data transmission is less than or equal to 50ms, meeting the real-time requirements of group exercise performances. A combination of time division multiple access and carrier sense multiple access multiple access protocol is used to improve the throughput and channel utilization of the network. A dynamic routing protocol is used to automatically select the optimal routing path based on the state of the network nodes and the communication quality, ensuring reliable data transmission.

[0078] The synchronization protocol adopts IEEE 1588 precision clock protocol to realize clock synchronization of multiple devices. Through communication between master and slave clocks, the time of the slave clock is adjusted to keep the clock of all devices in the network synchronized. The error of clock synchronization is less than or equal to 1 ms, ensuring the consistency of low-delay communication and data acquisition of large-scale nodes.

[0079] The key of the clock synchronization algorithm is to determine the time difference between the master clock and the slave clock and the time spent on signal transmission, and then to calibrate the time of the slave clock according to these information. The specific steps are as follows: the master clock first sends a synchronization message to the slave clock, which marks the time point of sending. After receiving the synchronization message, the slave clock records the time point of receiving, and then sends a response message to the master clock, which contains the time point of receiving the synchronization message and the time point of sending the synchronization message by the master clock. After receiving the response message from the slave clock, the master clock records the time point of receiving. Then, the master clock calculates the time deviation between the master and slave clocks and the signal transmission delay. The master clock sends the calculated time deviation and transmission delay information to the slave clock, and the slave clock adjusts its time according to the received time deviation information, thus realizing the synchronization of the master and slave clocks.

[0080] Fast deployment process:

[0081] When deploying in outdoor or temporary sites, first start the wearable devices and sensor nodes of the perception layer, which automatically form a wireless Mesh network. Then, the system automatically configures network parameters such as channel selection, node ID allocation, etc. through self-organizing algorithm. Next, the sensor nodes of the perception layer perform self-calibration and positioning to establish the coordinate system of the stage. Finally, the control layer and cloud layer are started, and the preset formation scheme is loaded, and the system enters the working state. The entire deployment process does not require manual intervention, realizing fast, convenient and adaptive deployment.

[0082] Environmental adaptability optimization:

[0083] Adopting frequency hopping technology and spread spectrum technology, the network's anti-interference ability is improved, and the influence of external electromagnetic interference on communication is reduced. According to the distance and communication quality between nodes, the transmission power is automatically adjusted to reduce power consumption and prolong the working time of the device while ensuring communication quality. When the number of nodes in the network changes or a node fails, the network automatically reconfigures and selects a new routing path to ensure normal operation of the network.

[0084] Edge control module: using the parallel processing capability of FPGA, the positioning and control instructions transmitted by the wireless networking module are processed; projection parameters are generated and hierarchical early warning strategies are developed, while the cloud path optimization algorithm is linked to realize low-delay collaborative control;

[0085] Processing of positioning and control instructions:

[0086] The edge computing platform uses FPGA chips as the core processor, utilizing the parallel processing capabilities of FPGA to achieve high-speed processing of positioning data. The platform's throughput is greater than or equal to 1000 points per second, capable of meeting the needs of large-scale simultaneous positioning of actors. At the same time, the platform integrates AI co-processors, running lightweight path planning models such as TensorFlowLite models, to achieve real-time optimization of actor movement paths.

[0087] An embedded real-time operating system is used to ensure real-time scheduling and response of tasks. Algorithm library: integrates core algorithm libraries such as positioning fusion algorithm, dynamic projection planning algorithm, and path optimization algorithm, providing support for the system's intelligent control. Interface layer: provides interfaces with other layers, enabling data interaction and instruction transmission.

[0088] Hierarchical warning mechanism:

[0089] The system uses a hierarchical warning mechanism based on the size of the actor's position deviation, as follows: when the deviation is less than or equal to 10 cm, only the vibration motor of the wearable device is used for reminders. When the deviation is between 10 and 30 cm, vibration reminders and local light flashing are triggered. When the deviation is greater than 30 cm, global red light warning is activated, and an alarm signal is sent to the command end. This hierarchical warning mechanism can provide different levels of reminders according to the severity of the deviation, improving the actor's perception efficiency of the deviation, while avoiding excessive interference.

[0090] Collaborative control algorithm:

[0091] The collaborative control algorithm is based on the edge computing platform, which realizes the collaborative processing of positioning data, projection control, and feedback warning. Its algorithm flow is as follows: the edge computing platform receives real-time positioning data from the perception layer, performs fusion processing, and obtains the real-time coordinates and deviation values of the actors. According to the deviation value, the corresponding warning mechanism is triggered, and the projection control instruction is generated to adjust the parameters of the projection identifier. At the same time, the actor's position data and deviation information are sent to the cloud layer for subsequent training assistance and data analysis. The delay time of this algorithm is less than 10 ms, ensuring the real-time and collaboration of the system.

[0092] The positioning data and projection instructions of the actors during each group exercise rehearsal are recorded. Using these historical trajectory data, the system can generate a three-dimensional heat map. In the three-dimensional heat map, the gradient of color represents the frequency of actor position deviation. Red areas mean that actor position deviation occurs frequently, while blue areas indicate that deviation occurs rarely.

[0093] A three-dimensional heat map is generated by kernel density estimation. In short, the bias of the actor at different positions is statistically analyzed, and the density of each position bias is calculated. The final three-dimensional heat map intuitively shows the position bias of the actor during the rehearsal process. According to the situation presented by the heat map, targeted training and improvement can be carried out, such as strengthening the exercise in the red high-frequency bias area corresponding to the action or walking position.

[0094] System architecture of the VR training simulation system:

[0095] The VR training simulation system is built based on the Unity engine, including stage three-dimensional scene reconstruction, actor model import, formation scheme loading and interaction module.

[0096] The system reconstructs the three-dimensional scene of the stage using the Unity engine, and the actor observes the spatial relationship of the formation through the head-mounted display. The system supports perspective switching and bias labeling functions, and the actor can observe the formation from different angles and intuitively see his position bias.

[0097] Through the VR training simulation system, the actor can better understand the complex three-dimensional formation and improve the rehearsal efficiency. Experimental data shows that the system can shorten the formation rehearsal cycle by more than 50%.

[0098] The above embodiments can be realized wholly or partially by software, hardware, firmware or any combination thereof. When realized by software, the above embodiments can be realized wholly or partially in the form of a computer program product. Those skilled in the art can realize that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution.

[0099] The units described as separate components can or can not be physically separated, and the components displayed as units can or can not be physical units, which can be located in one place or distributed on multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0100] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any skilled person in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered within the protection scope of the present application.

Claims

1. A group calisthenics formation intelligent marker based on dynamic light positioning, characterized in that, The system comprises: The perception execution module: taking the wearable device as the carrier, integrating the LED and UWB positioning functions, positioning by sending dynamic light signals; after receiving the instructions at the receiving unit, projecting the light spot to guide the actor operation; when the actor position deviates, timely feedback through the vibration motor; The data fusion module: on the basis of the data obtained by the perception execution module, collecting relevant position data, using Kalman filtering algorithm for fusion processing, and outputting real-time coordinates; The wireless networking module: according to the real-time coordinates, constructing a Mesh network for efficient transmission of positioning and control instructions; The edge control module: using the parallel processing capability of FPGA, processing the positioning and control instructions transmitted by the wireless networking module; generating projection parameters and formulating a hierarchical early warning strategy, while linking the cloud path optimization algorithm for low-delay collaborative control.

2. The dynamic light positioning based group exercise formation intelligent marker according to claim 1, characterized in that, The process of positioning by sending dynamic light signals is: Taking the wearable device as the carrier, the device integrates a micro LED array and a UWB tag, the LED array flashes with a unique frequency within a certain range as an active marker for optical positioning, while the UWB tag communicates with the base station to provide centimeter-level position data, combining to achieve positioning.

3. The dynamic light positioning based group exercise formation intelligent marker of claim 2, wherein, The process of projecting the light spot to guide the actor operation is: Through the positioning fusion algorithm, the UWB, IMU and LiDAR data are fused based on Kalman filtering to output the real-time coordinates of the actor; the dynamic projection planning module calculates the target position based on the preset formation program, generates light identification parameter control documents, and the laser projector of the execution layer projects the light spot according to the instructions, and the self-adaptive light adjustment module adjusts the brightness to realize guidance and feedback of the execution state.

4. The dynamic light positioning based group exercise formation intelligent marker of claim 1, wherein, The process of outputting real-time coordinates is: The control layer uses a multi-sensor fusion strategy to dynamically allocate the weights of optical, UWB, IMU and LiDAR sensors according to the motion state, uses robust Kalman filtering to remove abnormal data, and outputs real-time coordinates through Kalman filtering algorithm fusion processing.

5. The dynamic light positioning based group exercise formation intelligent marker of claim 1, wherein, The specific process of constructing a Mesh network is: A wireless Mesh ad hoc network with a mesh topology is constructed, and each wearable device acts as a node that can receive and forward data.

6. The dynamic light positioning based group exercise formation intelligent marker of claim 1, wherein, The process of efficient transmission of positioning and control instructions is: Low-delay communication protocols, time division multiple access and carrier sense multiple access combined multiple access protocols, and dynamic routing protocols are used to select the optimal path according to the node state and communication quality.

7. The dynamic light positioning based group exercise formation intelligent marker of claim 1, wherein, The process of processing positioning and control instructions is: An edge computing platform with FPGA as the core, combined with an AI coprocessor, realizes data processing and instruction interaction through a specific software architecture.

8. The dynamic light positioning based group exercise formation intelligent marker of claim 1, wherein, The process of the hierarchical early warning mechanism is: When the deviation is less than or equal to the distance y, only the vibration motor of the wearable device is used for warning; when the deviation is between the distance y and the distance y1, the vibration warning and local light flashing are triggered; when the deviation is greater than the distance y1, the global red light warning is started, and an alarm signal is sent to the command end.

9. The dynamic light positioning based group exercise formation intelligent marker of claim 1, wherein, The process of low-delay collaborative control is: Based on the edge computing platform, real-time positioning data of the perception layer is received and fused to obtain real-time coordinates and deviation values of the actor; according to the deviation values, a corresponding hierarchical early warning mechanism is triggered to generate projection control instructions to adjust the spot shape and color parameters; the actor position data and deviation information are sent to the cloud layer to generate a three-dimensional heat map and VR training simulation training assistance and data analysis.

Images