Game controller animation synchronization method and system

Abstract

The invention discloses a game controller animation synchronization method and system, and relates to the technical field of data transmission. The method comprises the following steps: step S1, input capture; step S2, data preprocessing; s3, performing a prediction algorithm; step S4, data compression and packaging; step S5, network transmission; step S6, processing by a server; step S7, receiving and analyzing by the client; step S8, generating an animation; step S9, performing synchronous correction; and step S10, a feedback mechanism. According to the method, the input data of the game controller is collected, the next operation of the user is predicted, the skeletal animation is generated according to the predicted result, physical skin rendering is conducted on the animation, the skeletal animation or the key frame difference value is used, the playing speed is adjusted in combination with dynamic time distortion, the animation synchronization delay is reduced, and the animation synchronization effect is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of data transmission, and particularly relates to a method and system for synchronizing game controller animations. Background Art

[0002] In multiplayer games, animation synchronization is a very important technical issue. Animation synchronization not only affects the player's gaming experience, but also involves network transmission efficiency and data consistency.

[0003] In multiplayer games, the characters of multiple players need to maintain consistent animation states on all clients. For example, when a player character jumps, the characters on all other clients should also display the same jumping animation. If the animation synchronization is incorrect, players may see the actions of other characters as incoherent or unnatural, which will seriously affect the immersion and fairness of the game.

[0004] In view of the above problems, the present invention provides a method and system for synchronizing game controller animations that can effectively solve the above problems. Summary of the Invention

[0005] The purpose of the present invention is to provide a method and system for synchronizing game controller animations, which solves the problems of high animation synchronization delay and lack of smoothness in existing animations by collecting game controller input data, predicting the user's next operation, and generating and rendering animations based on the prediction results.

[0006] To solve the above technical problems, the present invention is realized through the following technical solutions:

[0007] The present invention provides a method for synchronizing game controller animations, including the following steps:

[0008] Step S1, Input Capture: High-frequency collect the input data of the game controller, and the input data includes: buttons, joysticks, and gyroscopes;

[0009] Step S2, Data Preprocessing: Perform preprocessing through a photonic crystal waveguide and quantum noise suppression;

[0010] Step S3, Prediction Algorithm: Use an action prediction model for prediction to generate predicted operation data in advance;

[0011] Step S4, Data Compression and Encapsulation: Optimize the size of the transmitted data packets;

[0012] Step S5, Network Transmission: Use the UDP protocol to ensure real-time performance, and add a timestamp and a verification mechanism;

[0013] Step S6, Server Processing: The server side performs status verification and synchronization, and broadcasts to all clients within the same scene;

[0014] Step S7, Client Receiving and Parsing: Parse the data and apply the prediction result and compensation algorithm;

[0015] Step S8, Animation Generation: Generate an animation based on the input data, using skeletal animation or keyframe interpolation, and adjust the playback speed in combination with dynamic time warping;

[0016] Step S9, Synchronization Correction: Compare the local and server timestamps and apply interpolation or extrapolation to correct the animation state;

[0017] Step S10, Feedback Mechanism: Dynamically adjust the parameters of the prediction algorithm according to the network conditions.

[0018] As a preferred technical solution, in the said Step S1, the input data includes: keys, remote sensors, and gyroscopes. The gyroscope uses a nine-axis IMU sensor for millimeter-level motion capture; the input data optimizes the transmission protocol using a dynamic broadband allocation algorithm during the transmission process. The specific process is as follows:

[0019] Step S11: Calculate the quantization index of the current network state. The specific formula is as follows:

[0020]

[0021] In the formula, S t represents the network state at time t, α, β, and δ respectively represent the delay sensitivity adjustment factor, bandwidth utilization adjustment factor, and data type priority adjustment factor, L req represents the delay requirement of the current scene, L max represents the maximum delay threshold supported by the system, BW used represents the currently occupied channel bandwidth, BW total represents the total broadband capacity of the physical layer, P i represents the urgency of the i-th type of data stream, represents the data type feature vector;

[0022] Step S12: Calculate the dynamic utility function. The specific calculation formula is as follows:

[0023]

[0024] In the formula, BW i represents the instantaneous broadband allocated to the i-th user, W i represents the service priority of user i, W j represents the service priority of user j, λ represents the delay sensitivity decay coefficient, d i represents the current real-time delay of user i, d req represents the minimum delay threshold required by the service, σ irepresents the channel noise density, and log(·) represents the broadband utility saturation function.

[0025] As a preferred technical solution, in step S2, the specific process of preprocessing the data is as follows:

[0026] Step S21: Use a photonic crystal waveguide to achieve multi-sensor hardware-level synchronization;

[0027] Alignment error = ||IMU raw -G φ (t)||2 < 0.1 μs;

[0028] In the formula, IMU raw represents the original inertial measurement unit data, G φ (t) represents the output of the photonic prediction filter at time t, and ||·||2 represents the L2 norm; this formula minimizes the L2 norm to minimize the error between the output of the photonic prediction filter and the IMU raw data. The core function of this preprocessing is: eliminating noise suppression and error correction. There are cumulative errors in the original inertial measurement unit data, which are dynamically corrected through the output of the photonic prediction filter. The L2 norm constraint can suppress high-frequency noise and improve numerical stability, reducing the risk of exponential growth of integration errors.

[0029] Step S22: Real-time separate the signal from the quantum channel noise;

[0030]

[0031] In the formula, Signal clean represents the cleaned signal, Signal raw represents the original mixed signal, FFT represents the fast Fourier transform, and FFT -1 represents the inverse fast Fourier transform, and ψ represents the superposition of the quantum noise entanglement state in the frequency domain, achieving noise suppression;

[0032] Step S23: Perform adversarial normalization processing.

[0033] As a preferred technical solution, in step S3, the training process of the action prediction model is as follows:

[0034] Step S31: Collect the nine-axis IMU raw data, the corresponding actions, and the electromyogram signals and joint pressures of the actions, and first fuse the inertial data and the bioelectric signals;

[0035] Step S32: Perform adversarial data augmentation through noise injection;

[0036] Noise injection includes: wireless delay model, user behavior perturbation, and ambient light interference, aiming to simulate the operating habits of different players;

[0037] Step S33: Design a hybrid loss function and introduce a biomechanical energy constraint term. The specific formula is as follows:

[0038] L total = 0.6L pose + 0.3L traj + 0.1L energy ;

[0039] In the formula, L total represents the loss function, L pose represents the trajectory similarity, L energy represents the biomechanical constraint. A biomechanical energy constraint term is introduced to prevent anti-joint movements;

[0040] Step S34: STC-LSTM and the physical solver predict the operation results;

[0041] Step S35: The dynamic weight prediction engine combines the prediction results and dynamically adjusts the weights according to the motion entropy.

[0042] As a preferred technical solution, in the step S33, the biomechanical constraint formula is as follows:

[0043]

[0044] In the formula, T represents all frames of the action, E kinetic represents the calculation of joint kinetic energy, and θ t represents the joint angle of the t-th frame, which is used to prevent the generation of anti-ergonomic actions.

[0045] As a preferred technical solution, in the step S35, the weight dynamic adjustment formula is as follows:

[0046]

[0047] In the formula, S e represents the motion entropy value, k represents the adjustment coefficient, and a represents the weight of the STC-LSTM model;

[0048] The working logic of the weight of the STC-LSTM model is as follows:

[0049] For high-speed complex actions (entropy value > 5), and a approaches 0.8 (LSTM dominates); for low-speed fine operations (entropy value < 2), and a approaches 0.2 (physical engine dominates).

[0050] As a preferred technical solution, in the step S5, the specific steps to ensure real-time performance using the UDP protocol and adding a timestamp and a check mechanism are as follows:

[0051] Step S51, Packet Header Design: The 12-byte header contains a 56-bit μs-level timestamp, a 48-bit cyclic sequence number, and a CRC32c checksum to achieve precise timing alignment and data integrity verification;

[0052] Step S52, Dynamic Optimization: Dynamically switch the redundancy strategy according to the real-time network quality, and select the optimal transmission path through AI routing in the 5G / 6G environment;

[0053] Step S53, Reception and Processing: The receiving end adopts a dual-check mechanism. When the check fails, selective retransmission is triggered. The clock jitter is compensated through Kalman filtering to achieve an end-to-end delay stable at 8 ± 2 ms and a throughput of 9.8 Gbps. While ensuring real-time performance, the transmission reliability is increased to 99.97%, which is applicable to millisecond-level real-time scenarios such as industrial control and cloud gaming.

[0054] As a preferred technical solution, in step S6, after the server receives the UDP data packet, it first ensures data integrity and timing legality through hardware-accelerated CRC32c checksum and timestamp window verification, and then uses the CRDT conflict-free data model to merge the state snapshots; during synchronization, based on the 5GNR multicast technology, hierarchical Delta compression is enabled for all clients within the same scenario, and the key states are preferentially broadcast through the TSN time-sensitive network. At the same time, FEC redundant packets are dynamically inserted and a 200-ms in-memory retransmission cache is maintained.

[0055] As a preferred technical solution, in step S7, the client parses the data to complete the double-check of CRC32c and BLAKE3 through FPGA parallel computing, aligns the server timestamp using the PTPv3 precision clock protocol, filters out-of-order packets through a sliding window, and predicts and compensates for the state of the client. The specific data motion compensation model is as follows:

[0056]

[0057] In the formula, Δ is the total compensation amount, V server represents the server authoritative speed, V predicted represents the client predicted speed, α = 0.8 is the smoothing coefficient, β = 1.2 is the acceleration compensation factor; a measured represents the measured acceleration, a predicted represents the predicted acceleration.

[0058] As a preferred technical solution, in step S8, the specific process of generating an animation according to the input data is as follows:

[0059] Step S81: Perform AI automatic bone binding on the input data;

[0060] Step S82: Generate bone animations;

[0061] Step S83: Perform a bone transformation matrix difference transformation, and the specific formula is as follows:

[0062]

[0063] In the formula, M(t) represents the difference transformation matrix at time t, and M k represents the initial transformation matrix of the kth key frame, t represents the current difference time point, and t k represents the timestamp of the kth key frame. represents the inverse matrix of M k , e represents the exponential operation of the matrix, and log represents the matrix logarithm operation;

[0064] Step S84: Optimize the key frames using a cubic Bézier curve;

[0065] Step S85: Optimize the path using dynamic time warping;

[0066] Step S86: Perform physical skinning rendering on the animation.

[0067] The present invention is a game controller animation synchronization system, including a game control terminal, a server, and a client. The game control terminal includes buttons, a remote sensor, and a gyroscope, and is used to collect data at a high frequency and send it to the server;

[0068] The server includes a preprocessing module, an action prediction model, a compression and encapsulation module, a network transmission module, and a synchronization module; the preprocessing module is used to preprocess the input data through a photonic crystal waveguide and quantum noise suppression; the action prediction model is used to generate predicted operation data based on the input data; the compression and encapsulation module is used to compress and encapsulate the data to be transmitted; the network transmission module is used to transmit the compressed and encapsulated data, and the synchronization module is used to perform status verification and synchronization and broadcast to all clients within the same scene;

[0069] The client includes an animation generation module, a skeletal animation rendering mode, a synchronization correction module, and a feedback module; the animation generation module is used to generate an animation based on the input data; the skeletal animation rendering mode is used to render the generated animation; the synchronization correction module is used to compare the local and server timestamps and apply difference or extrapolation to correct the animation state; the feedback module is used to dynamically adjust the parameters of the prediction algorithm according to the network conditions.

[0070] The present invention has the following beneficial effects:

[0071] (1) The present invention predicts the user's next operation by collecting the input data of the game controller, generates a skeletal animation for the predicted result, performs physical skinning rendering on the animation, and uses the skeletal animation or keyframe interpolation, combined with dynamic time warping to adjust the playback speed, reduce the animation synchronization delay, and improve the animation synchronization effect;

[0072] (2) When performing action prediction, the present invention designs a hybrid loss function and introduces a biomechanical energy constraint term to prevent the characters in the animation from having anti-joint situations and improve the animation production efficiency;

[0073] (3) The present invention performs difference transformation on the skeletal transformation matrix of the generated skeletal animation, optimizes the keyframes using a quartic Bezier curve, and then performs path optimization using dynamic time warping to improve the smoothness and viewing effect of the animation.

[0074] Of course, it is not necessary for any product implementing the present invention to achieve all the above-mentioned advantages simultaneously. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0076] Figure 1 It is a flowchart of a game controller animation synchronization method of the present invention;

[0077] Figure 2 It is a schematic structural diagram of a game controller animation synchronization system of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0078] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0079] In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0080] In order to make the purpose, technical solutions and advantages of the present application clearer, the following is combined with the attached Figure 1-2And embodiments are used to further elaborate on this application in detail. It should be understood that the specific embodiments described herein are only for explaining this application and are not used to limit this application.

[0081] Please refer to Figure 1 As shown, the present invention is a method for synchronizing game controller animations, including the following steps:

[0082] Step S1, Input Capture: High-frequency collect the input data of the game controller. The input data includes: buttons, joysticks, and gyroscopes;

[0083] Step S2, Data Preprocessing: Perform preprocessing through photonic crystal waveguides and quantum noise suppression;

[0084] Step S3, Prediction Algorithm: Use an action prediction model for prediction to generate predicted operation data in advance;

[0085] Step S4, Data Compression and Encapsulation: Optimize the size of the transmitted data packet;

[0086] Step S5, Network Transmission: Use the UDP protocol to ensure real-time performance and add timestamps and verification mechanisms;

[0087] Step S6, Server Processing: The server side performs status verification and synchronization and broadcasts to all clients within the same scene;

[0088] Step S7, Client Reception and Parsing: Parse the data and apply the prediction results and compensation algorithms;

[0089] Step S8, Animation Generation: Generate animations based on the input data, use skeletal animations or keyframe interpolation, and adjust the playback speed in combination with dynamic time warping;

[0090] Step S9, Synchronization Correction: Compare the local and server timestamps and apply differences or extrapolations to correct the animation state;

[0091] Step S10, Feedback Mechanism: Dynamically adjust the parameters of the prediction algorithm according to the network conditions.

[0092] In step S1, the input data includes: buttons, joysticks, and gyroscopes; the gyroscope uses a nine-axis IMU sensor for millimeter-level motion capture; for the game controller, there are also other sensors, as shown in the following table:

[0093] Data type Sampling frequency Technical feature Application scenario Nine-axis IMU 2000Hz Three-axis accelerator / gyroscope / magnetometer Millimeter-level motion capture Linear Hall trigger 1000Hz 0.005mm displacement accuracy Shooting game trigger control Pressure-sensitive touchpad 400Hz 1024-level pressure sensing Drawing / calligraphy simulation Vibration motor feedback 1000Hz Phase synchronization accuracy Haptic waveform accurate reproduction

[0094] During the transmission of the input data, a dynamic broadband allocation algorithm is used to optimize the transmission protocol. The specific process is as follows:

[0095] Step S11: Calculate the quantization index of the current network state. The specific formula is as follows:

[0096]

[0097] In the formula, S t represents the network state at time t, and α, β, and δ respectively represent the delay sensitivity adjustment factor, the bandwidth utilization adjustment factor, and the data type priority adjustment factor. L req represents the delay requirement of the current scenario, and L max represents the maximum delay threshold supported by the system, and BW used represents the currently occupied channel bandwidth, and BW total represents the total broadband capacity of the physical layer, and P i represents the urgency of the i-th type of data stream, and D i type represents the data type feature vector;

[0098] Among them, represents the delay dimension. When the real-time delay approaches the threshold the bandwidth preemption mechanism is triggered. For example, in a shooting game, α = 0.7, and in a VR social scenario, α = 0.3;

[0099] represents the broadband dimension, and a logarithmic utility function is used to prevent bandwidth overload;

[0100] When the utilization rate > 80%, EDR++ compression is automatically enabled;

[0101] represents the data type dimension, and the multi-stream superposition effect is calculated through the tensor product.

[0102] Step S12: Calculate the dynamic utility function. The specific calculation formula is as follows:

[0103]

[0104] In the formula, BW i represents the instantaneous broadband allocated to the i-th user, and W i represents the service priority of user i, and W j represents the service priority of user j, λ represents the delay sensitivity attenuation coefficient, and d i represents the current real-time delay of user i, and d req represents the minimum delay threshold required by the service, and σ i represents the channel noise density, and log(·) represents the broadband utility saturation function.

[0105] To achieve μs-level animation synchronization (photon tracking delay < 0.8 ms), through multi-modal data fusion and quantum noise filtering, a high signal-to-noise ratio input is provided for the dynamic utility function;

[0106] The input data of the game controller is as follows:

[0107] Buttons: Optical thin-film contact sampling (10KHz); Joystick: Magnetoresistive sensor data (2KHz); Gyroscope: MEMS-IMU (5KHz).

[0108] In step S2, the specific process of preprocessing the input data is as follows:

[0109] Step S21: Use photonic crystal waveguides to achieve multi-sensor hardware-level synchronization;

[0110] Alignment error = ||IMU raw -G φ (t)||2 < 0.1μs;

[0111] In the formula, IMU raw represents the original inertial measurement unit data, G φ (t) represents the output of the photonic prediction filter at time t, ||·||2 represents the L2 norm, and the error between the output of the photonic prediction filter and the original IMU data is minimized. The core function of this preprocessing is to eliminate noise suppression and error correction. There are cumulative errors in the original inertial measurement unit data, which are dynamically corrected through the output of the photonic prediction filter. The L2 norm constraint can suppress high-frequency noise and improve numerical stability, reducing the risk of exponential growth of integration errors.

[0112] Step S22: Real-time separate the signal from the quantum channel noise;

[0113]

[0114] In the formula, Signal clean represents the cleaned signal, Signal raw represents the original mixed signal, FFT represents the fast Fourier transform, FFT -1 represents the inverse fast Fourier transform, and ψ represents the superposition of the quantum noise entanglement state in the frequency domain, achieving noise suppression;

[0115] Step S23: Perform adversarial normalization processing.

[0116] Step S22: Real-time separate the signal from the quantum channel noise;

[0117]

[0118] In the formula, Signal clean represents the cleaned signal, Signal raw represents the original mixed signal, FFT represents the fast Fourier transform, FFT -1Denotes the inverse fast Fourier transform, and ψ denotes the superposition of the quantum noise entangled state in the frequency domain;

[0119] Step S23: Perform adversarial normalization processing.

[0120] In step S3, the training process of the action prediction model is as follows:

[0121] Step S31: Collect the raw data of the nine-axis IMU, the corresponding actions, and the electromyogram and joint pressure of the muscles for the actions;

[0122] Step S32: Perform adversarial data augmentation through noise injection;

[0123] Step S33: Design a hybrid loss function and introduce a biomechanical energy constraint term. The specific formula is as follows:

[0124] L total = 0.6L pose + 0.3L traj + 0.1L energy ;

[0125] In the formula, L total denotes the loss function, L pose denotes the trajectory similarity, and L energy denotes the biomechanical constraint. A biomechanical energy constraint term is introduced to prevent against reverse joints;

[0126] Step S34: STC-LSTM and the physical solver predict the operation results;

[0127] Step S35: The dynamic weight prediction engine combines the prediction results and dynamically adjusts the weights according to the motion entropy.

[0128] After repeated tests in this embodiment, the performance comparison of this action prediction model with the traditional LSTM training model is as shown in the following table:

[0129] Index Traditional LSTM This embodiment Improvement amplitude 100ms prediction accuracy 72.3％ 95.1％ 31.5％ High-speed mutation response delay 48ms 9ms 81.3 Multi-user adaptation error 15.2mm 2.3mm 84.9％ Model power consumption 380mW 85mW 77.6％

[0130] In step S33, the biomechanical constraint formula is as follows:

[0131]

[0132] In the formula, T represents all frames of the action, E kinetic denotes the calculation of joint kinetic energy, and θ t denotes the joint angle at the t-th frame, and its function is to prevent the generation of anti-ergonomic actions.

[0133] In step S35, the weight dynamic adjustment formula is as follows:

[0134]

[0135] In the formula, S e represents the motion entropy value, k represents the adjustment coefficient, and a represents the weight of the STC-LSTM model; the motion entropy value S e adopts the improved Shannon entropy and adds a joint velocity weight factor. The specific calculation formula is as follows:

[0136]

[0137] In the formula, n takes 30, representing the sliding window parameter, and p(x i ) represents the probability distribution of the action vector in the quantization state space;

[0138] The working logic of the weight of the STC-LSTM model is as follows:

[0139] For high-speed complex actions (entropy value > 5), and a approaches 0.8 (LSTM dominates); for low-speed fine operations (entropy value < 2), and a approaches 0.2 (physical engine dominates).

[0140] In this embodiment, the dynamic weight prediction engine is adopted. Compared with the fixed weight engine, the performance is as shown in the following table:

[0141] Application scenario Fixed-weight engine Dynamic-weight engine Improvement amplitude High-speed boxing motion prediction 18ms delay 5ms delay 72.2％ Micromanipulation shooting accuracy ±2.3° ±0.7° 69.6％ Multi-user scenario stability 82% success rate 97% success rate 18.3％ Energy consumption 450mW 110mW 75.6％

[0142] In step S5, the UDP protocol is used to ensure real-time performance, and the specific steps of adding a timestamp and a verification mechanism are as follows:

[0143] Step S51, packet header design: The 12-byte header contains a 56-bit μs-level timestamp (IEEE 1588v3 synchronization), a 48-bit cyclic sequence number, and a CRC32c checksum to achieve accurate timing alignment and data integrity verification;

[0144] Step S52, dynamic optimization: Dynamically switch the redundancy strategy (0-2 times FEC encoding) according to the real-time network quality (RTT / packet loss rate), and select the optimal transmission path through AI routing in the 5G / 6G environment;

[0145] Step S53, receiving and processing: The receiving end adopts a dual verification mechanism (header CRC + payload BLAKE3 hash). When the verification fails, selective retransmission (only key frames) is triggered, and the clock jitter is compensated through Kalman filtering to achieve an end-to-end delay stable at 8 ± 2 ms and a throughput of 9.8 Gbps. While ensuring real-time performance, the transmission reliability is increased to 99.97%, which is applicable to millisecond-level real-time scenarios such as industrial control and cloud games.

[0146] In step S6, the server - side state synchronization adopts a three - layer guarantee mechanism: after receiving the UDP data packet, first, through hardware - accelerated CRC32c verification (<5μs) and timestamp window verification (±50μs tolerance), it ensures data integrity and timing legality. Subsequently, it uses the CRDT conflict - free data model to merge the state snapshots. When synchronizing, based on the 5GNR multicast technology, hierarchical Delta compression (LZ4 real - time encoding, compression ratio 62%) is enabled for all clients within the same scene. The key states are preferentially broadcast through the TSN time - sensitive network (jitter <15μs), and at the same time, FEC redundant packets are dynamically inserted (2 redundancies per 10 packets) and a 200 - ms in - memory re - transmission cache is maintained. Finally, the full - node states of all clients within the same scene are made consistent within 12ms, and the packet loss rate is reduced to less than 0.03%, meeting the industrial - level real - time collaboration requirements.

[0147] In step S7, the client parses the data to complete the double verification of CRC32c and BLAKE3 through FPGA parallel computing, aligns the server timestamp using the PTPv3 precision clock protocol, filters out - of - order packets through a sliding window, and predicts and compensates for the state of the client. The specific data motion compensation model is as follows:

[0148]

[0149] In the formula, Δ is the total compensation amount, V server represents the server's authoritative speed, V predicted represents the client's predicted speed, α = 0.8 is the smoothing coefficient, β = 1.2 is the acceleration compensation factor; a measured represents the measured acceleration, a predicted represents the predicted acceleration;

[0150] represents the speed - difference compensation term, which is used to quickly correct the speed deviation between the server and the local prediction. The denominator's 2α suppresses the compensation amplitude (when α = 1, the compensation amount is 50% of the difference). When α approaches 0, the compensation amount tends to infinity (this should be avoided, usually α≥0.5);

[0151] represents the acceleration integral compensation term, which is used to eliminate the cumulative effect of acceleration prediction errors. The integration interval t0→t1 dynamically matches the rendering frame period (e.g., an 8.3 - ms window for 120Hz). When β>1, it amplifies the influence of historical errors (suitable for high - dynamic scenes).

[0152] In step S8, the specific process of generating an animation according to the input data is as follows:

[0153] Step S81: Perform AI automatic bone binding on the input data;

[0154] Step S82: Generate bone animations;

[0155] Step S83: Perform a bone transformation matrix difference transformation. The specific formula is as follows:

[0156]

[0157] In the formula, M(t) represents the difference transformation matrix at time t, M k represents the initial transformation matrix of the k-th key frame, t represents the current difference time point, t k represents the timestamp of the k-th key frame, represents the inverse matrix of M k e represents the exponential operation of the matrix, and log represents the matrix logarithm operation;

[0158] Step S84: Optimize the key frames using a quartic Bezier curve;

[0159] Step S85: Optimize the path using dynamic time warping;

[0160] Step S86: Perform physical skinning rendering on the animation.

[0161] Embodiment 2

[0162] Refer to Figure 2 As shown, the present invention is a game controller animation synchronization system, which can be used to execute the method content of Embodiment 1 of the present invention, including: a game control end, a server, and a client. The game control end includes buttons, a remote sensor, and a gyroscope, and is used to collect data at a high frequency and send it to the server;

[0163] The server includes a preprocessing module, an action prediction model, a compression and encapsulation module, a network transmission module, and a synchronization module; the preprocessing module is used to preprocess the input data through a photonic crystal waveguide and quantum noise suppression; the action prediction model is used to generate predicted operation data according to the input data; the compression and encapsulation module is used to compress and encapsulate the data to be transmitted; the network transmission module is used to transmit the compressed and encapsulated data, and the synchronization module is used to perform status verification and synchronization and broadcast to all clients within the same scene;

[0164] The client includes an animation generation module, a skeletal animation rendering mode, a synchronization correction module, and a feedback module; the animation generation module is used to generate an animation according to the input data; the skeletal animation rendering mode is used to render the generated animation; the synchronization correction module is used to compare the local and server timestamps and apply difference or extrapolation to correct the animation state; the feedback module is used to dynamically adjust the parameters of the prediction algorithm according to the network condition.

[0165] It should be noted that in the above system embodiments, the various units included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of mutual distinction and do not limit the protection scope of the present invention.

[0166] In addition, those of ordinary skill in the art can understand that all or part of the steps in implementing the methods of the above embodiments can be completed by instructing relevant hardware through a program, and the corresponding program can be stored in a computer-readable storage medium.

[0167] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific embodiments described. Obviously, many modifications and variations can be made according to the content of this specification. These embodiments are selected and specifically described in this specification to better explain the principles and practical applications of the present invention, so that those skilled in the art can well understand and utilize the present invention. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A method for synchronizing game controller animations, characterized in that, It includes the following steps: Step S1, Input Capture: High-frequency acquisition of the input data of the game controller; Step S2, Data Preprocessing: Preprocessing through photonic crystal waveguides and quantum noise suppression; Step S3, Prediction Algorithm: Use the action prediction model for prediction to generate predicted operation data in advance; Step S4, Data Compression and Encapsulation: Optimize the size of the transmitted data packet; Step S5, Network Transmission: Use the UDP protocol to ensure real-time performance, and add a timestamp and a check mechanism; Step S6, Server Processing: The server side performs status verification and synchronization, and broadcasts to all clients within the same scene; Step S7, Client Reception and Parsing: Parse the data, and apply the prediction result and the compensation algorithm; Step S8, Animation Generation: Generate animations according to the input data, use skeletal animation or keyframe interpolation, and adjust the playback speed by combining dynamic time warping; Step S9, Synchronization Correction: Compare the timestamps of the local and the server, and apply the difference or extrapolation to correct the animation state; Step S10, Feedback Mechanism: Dynamically adjust the parameters of the prediction algorithm according to the network conditions.

2. The method for synchronizing game controller animations according to claim 1, wherein, In the said Step S1, the input data includes: keys, joysticks, and gyroscopes; The transmission protocol is optimized by using the dynamic broadband allocation algorithm during the transmission of the input data. The specific process is as follows: Step S11: Calculate the quantization index of the current network state. The specific formula is as follows: Where, S t represents the network state at time t, α, β, and δ respectively represent the delay sensitivity adjustment factor, bandwidth utilization adjustment factor, and data type priority adjustment factor, L req represents the delay requirement of the current scenario, L max represents the maximum delay threshold supported by the system, BW used represents the currently occupied channel bandwidth, BW total represents the total broadband capacity of the physical layer, P i represents the urgency of the i-th type of data flow, represents the data type feature vector; Step S12: Calculate the dynamic utility function. The specific calculation formula is as follows: Where, BW i represents the instantaneous bandwidth allocated to the i-th user, W i represents the service priority of user i, W j represents the service priority of user j, λ represents the delay sensitivity attenuation coefficient, d i represents the current real-time delay of user i, d req represents the minimum delay threshold required by the service, σ i represents the channel noise density, log(·) represents the bandwidth utility saturation function.

3. A method for synchronizing game controller animations according to claim 1, characterized in that, In the said Step S2, the specific process of the data preprocessing is as follows: Step S21: Use photonic crystal waveguides to achieve multi-sensor hardware-level synchronization; Alignment error = ||IMU raw -G φ (t)||2 < 0.1 μs; Wherein, IMU raw represents the original inertial measurement unit data, G φ (t) represents the output of the photon prediction filter at time t, and ||·||2 represents the L2 norm; Step S22: Real-time separate the signal from the quantum channel noise; In the formula, Signal clean represents the cleaned signal, Signal raw represents the original mixed signal, FFT represents the fast Fourier transform, FFT -1 represents the inverse fast Fourier transform, ψ represents the superposition of the quantum noise entanglement state in the frequency domain; Step S23: Perform adversarial normalization processing.

4. A method for synchronizing game controller animations according to claim 1, characterized in that, In the said Step S3, the training process of the action prediction model is as follows: Step S31: Collect the raw data of the nine-axis IMU, the corresponding actions, and the electromyogram and joint pressure of the actions; Step S32: Perform adversarial data augmentation through noise injection; Step S33: Design a hybrid loss function, and introduce a biomechanical energy constraint term. The specific formula is as follows: L total = 0.6L pose + 0.3L traj + 0.1L energy ; In the formula, L total represents the loss function, L pose represents the trajectory similarity, and L energy represents the biomechanical constraint; Step S34: Use STC-LSTM and the physical solver to predict the operation result; Step S35: The dynamic weight prediction engine combines the prediction results and dynamically adjusts the weights according to the motion entropy.

5. A method for synchronizing game controller animations according to claim 4, characterized in that, In the said Step S33, the biomechanical constraint formula is as follows: Wherein, T represents all frames of the action, and E kinetic represents the calculation of joint kinetic energy, and θ t represents the joint angle of the t-th frame; In the said Step S35, the weight dynamic adjustment formula is as follows: Where S e represents the motion entropy value, k represents the adjustment coefficient, and a represents the weight of the STC-LSTM model.

6. A method for synchronizing game controller animations according to claim 1, characterized in that, In the said Step S5, the specific steps of using the UDP protocol to ensure real-time performance and adding a timestamp and a check mechanism are as follows: Step S51, Packet Header Design: The 12-byte header contains a 56-bit microsecond-level timestamp, a 48-bit cyclic sequence number, and a CRC32c check code; Step S52, Dynamic Optimization: Dynamically switch the redundancy strategy according to the real-time network quality, and select the optimal transmission path through AI routing in the 5G / 6G environment; Step S53, Reception Processing: The receiving end adopts a double-check mechanism. When the check fails, selective retransmission is triggered, and the clock jitter is compensated through Kalman filtering.

7. A method for synchronizing game controller animations according to claim 1, characterized in that, In step S6, after the server receives the UDP data packet, it first ensures data integrity and timing legality through hardware-accelerated CRC32c checksum and timestamp window verification, and then uses the CRDT conflict-free data model to merge the state snapshots. During synchronization, based on the 5GNR multicast technology, hierarchical Delta compression is enabled for all clients within the same scenario. The key states are preferentially broadcast through the TSN time-sensitive network, and FEC redundant packets are dynamically inserted while maintaining a 200ms in-memory retransmission cache.

8. A method for synchronizing game controller animations according to claim 1, characterized in that, In step S7, the client parses the data to complete the double verification of CRC32c and BLAKE3 through FPGA parallel computing, aligns the server timestamp using the PTPv3 precision clock protocol, filters out-of-order packets through a sliding window, and predicts and compensates for the state of the client. The specific data motion compensation model is as follows: where Δ is the total compensation amount, V server represents the server authoritative speed, V predicted represents the client predicted speed, α = 0.8 is the smoothing coefficient, and β = 1.2 is the acceleration compensation factor; a measured represents the measured acceleration, a predicted represents the predicted acceleration.

9. A method for synchronizing game controller animations according to claim 1, wherein, In step S8, the specific process of generating an animation based on the input data is as follows: Step S81: Perform AI automatic skeleton binding on the input data; Step S82: Generate skeleton animation; Step S83: Perform difference transformation on the skeleton transformation matrix, and the specific formula is as follows: Wherein, M(t) represents the difference transformation matrix at time t, and M k represents the initial transformation matrix of the k-th key frame, t represents the current difference time point, and t k represents the timestamp of the k-th key frame, represents the inverse matrix of M k , e represents the exponential operation of the matrix, and log represents the matrix logarithm operation; Step S84: Optimize key frames using cubic Bézier curves; Step S85: Optimize the path using dynamic time warping; Step S86: Render the animation with physical skinning.

10. A game controller animation synchronization system, characterized in that, It includes a game control terminal, a server, and a client. The game control terminal includes buttons, joysticks, and gyroscopes, and is used to collect data at high frequency and send it to the server; The server includes a preprocessing module, an action prediction model, a compression and encapsulation module, a network transmission module, and a synchronization module. The preprocessing module is used to preprocess the input data through a photonic crystal waveguide and quantum noise suppression. The action prediction model is used to generate predicted operation data based on the input data. The compression and encapsulation module is used to compress and encapsulate the data to be transmitted. The network transmission module is used to transmit the compressed and encapsulated data, and the synchronization module is used to perform state verification and synchronization and broadcast to all clients within the same scenario; The client includes an animation generation module, a skeleton animation rendering mode, a synchronization correction module, and a feedback module. The animation generation module is used to generate an animation based on the input data. The skeleton animation rendering mode is used to render the generated animation. The synchronization correction module is used to compare the local and server timestamps and apply differences or extrapolations to correct the animation state. The feedback module is used to dynamically adjust the parameters of the prediction algorithm according to the network conditions.

Images