Gamepad power management method and system

Abstract

The invention discloses a gamepad power management method which comprises the following steps: acquiring power consumption data of a gamepad in different game scenes, and constructing a power consumption model; according to the power consumption model, analyzing the influence of the player operation habit on power consumption, and generating an operation habit classification label; dynamically adjusting a time interval threshold value of a standby mode according to the operation habit classification label and a game scene change trend; vibration function use, light flicker frequency and key operation frequency of the handle during game scene switching are monitored, and the power consumption trend of the current scene is judged; if the power consumption trend of the current scene is lower than a preset threshold value, triggering a low power consumption mode; according to the time interval threshold value of the standby mode, whether the low-power-consumption standby mode is quitted or not is judged by combining the information of the environment where the handle is located; in the low-power-consumption standby mode, an operation signal of the handle is monitored, and power source resources are distributed according to the current game scene and operation habits of players.

Description

Technical Field

[0001] The present invention relates to the technical field of entertainment device management, and in particular to a power management method and system for a game controller. Background Art

[0002] The power management of a game controller is a complex technical issue involving multiple aspects. First, the power consumption of a game controller varies greatly in different game scenarios. For example, in some large 3D games, the game controller needs to vibrate and flash lights frequently, which will consume a large amount of power; while in some casual puzzle games, the game controller may only need to press a few buttons occasionally, and the power consumption is relatively low. Second, different players' gaming habits and operation methods will also affect the power consumption of the game controller. Some players like to play games continuously for a long time, and some players like to turn the game controller on and off frequently. These behavior patterns pose challenges to power management.

[0003] In addition, the power management of the game controller also needs to consider the design of the standby mode. Since players do not always play games, the game controller must have an intelligent way to automatically enter the low-power standby mode after the player pauses the game for a period of time to extend the battery life. However, the design of this standby mode also faces a dilemma: if the time interval for switching to the standby mode is set too short, the player will feel that the controller responds slowly; if the switching time interval is set too long, it will waste power. Therefore, finding a suitable time interval that can save power without affecting the gaming experience is a technical difficulty. Summary of the Invention

[0004] The present invention provides a power management method for a game controller to solve the above-mentioned existing technical problems.

[0005] The technical solution of the present invention is realized as follows:

[0006] A power management method for a game controller includes the following steps:

[0007] S1. Obtain the power consumption data of the game controller in different game scenarios and construct a power consumption model;

[0008] S2. Analyze the influence of the player's operation habits on the power consumption according to the power consumption model and generate operation habit classification labels;

[0009] S3. Dynamically adjust the time interval threshold of the standby mode according to the operation habit classification labels and the change trend of the game scenario;

[0010] S4. Monitor the vibration function usage, light flashing frequency, and button operation frequency of the controller when the game scenario switches, and judge the power consumption trend of the current scenario;

[0011] S5. If the power consumption trend of the current scenario is lower than the preset threshold, trigger the low-power mode and reduce the power supply for the vibration function and the light flashing.

[0012] S6. According to the time interval threshold of the standby mode, combined with the environmental information of the handle, determine whether to exit the low-power standby mode.

[0013] S7. In the low-power standby mode, monitor the operation signals of the handle and allocate power resources according to the current game scenario and the player's operation habits.

[0014] Furthermore, the process of constructing the power consumption model in step S1 is specifically as follows:

[0015] Obtain the vibration frequency, light flashing frequency, and key operation frequency data of the game handle in different game scenarios, and preprocess the data.

[0016] According to the preprocessed power consumption data of the game handle, construct an initial power consumption model. Take the vibration frequency, light flashing frequency, and key operation frequency as independent variables and the power consumption as the dependent variable, and fit the coefficients of the regression model by the least squares method.

[0017] Evaluate the initial power consumption model. By calculating the mean square error and determination coefficient indicators, judge the fitting degree and prediction ability of the model. If the model performance does not meet the requirements, introduce a regularization term to control the model complexity, and try to use the support vector machine. By comparing the performance of different algorithms, select the modeling method.

[0018] Furthermore, the process of generating the operation habit classification label in step S2 is specifically as follows:

[0019] According to the power consumption model, obtain the operation record data of the player. For the operation record data of each player, count the frequency of various operations to obtain the operation frequency distribution.

[0020] According to the operation frequency distribution, judge whether each operation is a high-frequency operation or a low-frequency operation. If the operation frequency is higher than the preset threshold, it is determined as a high-frequency operation; otherwise, it is determined as a low-frequency operation.

[0021] For the high-frequency operations and low-frequency operations, extract the operation features respectively to obtain the high-frequency operation feature set and the low-frequency operation feature set.

[0022] According to the high-frequency operation feature set and the low-frequency operation feature set, classify the players' operation habits to obtain the operation habit classification result.

[0023] For each operation habit classification, count the average power consumption of the players in this classification to obtain the power consumption levels of each operation habit classification.

[0024] Further, the process of dynamically adjusting the time interval threshold of the standby mode in step S3 is specifically as follows:

[0025] According to the pre-established correspondence between the operation frequency and the time interval threshold, obtain the real-time operation data of the player, and determine the current operation frequency of the player through data analysis;

[0026] Judge whether the operation frequency of the player belongs to high-frequency operation or low-frequency operation. If it is high-frequency operation, set the time interval of the standby mode to a longer interval. If it is low-frequency operation, set the time interval of the standby mode to a shorter interval;

[0027] Use the clustering algorithm to classify the operation habits of the player to obtain typical behavior patterns under different operation habits;

[0028] For each typical behavior pattern, analyze its correlation with the game scene change through the association rule mining algorithm to obtain the scene change trend under different behavior patterns;

[0029] Dynamically adjust the time interval threshold of the standby mode according to the scene change trend.

[0030] Further, the process of judging the power consumption trend of the current scene in step S4 is specifically as follows:

[0031] Obtain the trigger signal for game scene switching. If a scene switching signal is detected, start real-time monitoring of the handle vibration frequency, light flashing frequency, and button operation frequency;

[0032] According to the preset vibration frequency threshold, light flashing frequency threshold, and button operation frequency threshold, judge whether the current handle vibration frequency, light flashing frequency, and button operation frequency exceed the threshold;

[0033] If at least one of the handle vibration frequency, light flashing frequency, and button operation frequency exceeds the preset threshold, judge that the current game scene is a high power consumption scene;

[0034] Among them, for the game scene judged to be high power consumption, use the support vector machine algorithm to train a high power consumption scene prediction model according to the handle vibration frequency, light flashing frequency, and button operation frequency data of the historical high power consumption scene.

[0035] Further, the process of triggering the low power consumption mode in step S5 is specifically as follows:

[0036] Obtain the power consumption data in the current scene, compare it with the preset threshold. If the power consumption is lower than the threshold, trigger the low power consumption mode;

[0037] Determine the reduction amplitudes of the vibration function and the light flashing effect according to the preset low-power strategy, and achieve function degradation by controlling the power supply;

[0038] Gradually reduce the vibration frequency and the light brightness until the lowest power consumption level is reached, while monitoring the change trend of power consumption;

[0039] Obtain the frequency reduction and dimming schemes by fitting the functional relationship between power consumption, vibration frequency, and light brightness through the least squares method;

[0040] Utilize the decision tree algorithm to dynamically adjust the trigger threshold of the low-power mode by comprehensively considering the current scene characteristics and user usage habit factors;

[0041] Construct a Markov model, predict the power consumption trend in the future period based on historical power consumption data, and make energy-saving decisions.

[0042] Further, the process of determining whether to exit the low-power standby mode by combining the environmental information where the handle is located in step S6 is specifically as follows:

[0043] Obtain the environmental information where the handle is currently located, including temperature, humidity, light intensity, and noise data;

[0044] Judge whether the current environment where the handle is located meets the normal working conditions according to the preset threshold ranges of environmental data in each dimension;

[0045] If the current environment does not meet the normal working conditions of the handle, start the low-power standby mode of the handle;

[0046] In the low-power standby mode, periodically collect the handle environmental data, and predict the environmental state of the handle in the future period according to the change trend of the environmental data;

[0047] Establish an association model between the environmental data and the handle power consumption through a machine learning algorithm, and dynamically adjust the standby time threshold of the low-power mode;

[0048] When the continuous standby duration of the handle exceeds the standby time threshold, and the environmental prediction result indicates that the handle will still be in an unavailable environment in the future period, control the handle to exit the low-power standby mode and enter the sleep state.

[0049] Further, the process of allocating power resources in step S7 is specifically as follows:

[0050] Obtain the operation signal of the handle, judge whether the handle is in the low-power standby mode, and if so, enter the continuous monitoring state, otherwise maintain the normal working state;

[0051] In the continuous monitoring state, the convolutional neural network algorithm is used to extract the feature of the handle operation signal to obtain the operation feature vector;

[0052] According to the pre-established game scene model and player operation habit model, the operation feature vector is classified to determine the current game scene type and player operation habit type;

[0053] For different game scene types and player operation habit types, the decision tree algorithm is used to generate the corresponding power resource allocation strategy;

[0054] The power resource allocation strategy is converted into a control instruction and sent to the power management module of the handle through the communication interface of the handle;

[0055] The power management module adjusts the working voltage and clock frequency of each hardware module according to the received control instruction;

[0056] When the handle ends the low-power standby mode, the default working parameters of each hardware module are restored;

[0057] The historical operation data and real-time power consumption data of the device are obtained to construct a training data set for the continuous training and optimization of the machine learning algorithm;

[0058] Through the continuous training and iterative optimization of the machine learning algorithm, the intelligent level of the power management strategy is continuously improved.

[0059] Furthermore, the method also includes:

[0060] S8. Dynamically adjust the time interval threshold of the standby mode, and combine the historical operation data and the current power consumption trend to optimize the power management strategy;

[0061] The process of optimizing the power management strategy in step S8 is specifically as follows:

[0062] The historical operation data and real-time power consumption data of the device are obtained to construct a training data set;

[0063] The decision tree algorithm is used to dynamically adjust the time interval threshold of the standby mode according to the historical operation data and the real-time power consumption trend to obtain the optimized threshold parameter;

[0064] Through the support vector machine algorithm, the historical data and real-time trend are comprehensively analyzed to judge the usage state and power consumption level of the current device, and determine whether to trigger the standby mode;

[0065] If the idle time of the current device exceeds the optimized time interval threshold and the power consumption level is low, it will automatically enter the standby mode to reduce power consumption;

[0066] In the standby mode, continuously monitor the operation behavior and power consumption changes of the device. If a user operation or a sudden increase in power consumption is detected, wake up the device and exit the standby mode.

[0067] A game controller power management system includes a power management module, a data acquisition and processing module, and a data analysis and model construction module.

[0068] The power management module controls the turning on, turning off, and switching of different power consumption states of the controller power supply according to the trigger conditions and exit conditions of the low-power mode.

[0069] According to the preset low-power strategy, reduce the power supply of the vibration function and the light flashing, and adjust the working voltage and clock frequency of each hardware module.

[0070] Dynamically adjust the time interval threshold of the standby mode to optimize the power management strategy.

[0071] The data acquisition and processing module acquires the data related to the power consumption of the game controller in different game scenarios, as well as the operation record data of the player and the environmental information where the controller is located, and preprocesses the acquired data.

[0072] Transmit the processed data to the power management module and the data analysis and model construction module to provide data support for the formulation and optimization of the power management strategy.

[0073] The data analysis and model construction module constructs an initial power consumption model based on the preprocessed power consumption data of the game controller and evaluates the performance of the model.

[0074] Based on the power consumption model, analyze the influence of the player's operation habits on the power consumption, and statistically analyze the power consumption levels of each category.

[0075] Analyze the correlation between the player's operation habits and the changes in the game scenarios.

[0076] Construct a Markov model to assist the power management module in making energy-saving decisions.

[0077] Establish a correlation model between the environmental data and the power consumption of the controller to provide support for dynamically adjusting the standby time threshold of the low-power mode.

[0078] Compared with the prior art, the present invention has the following beneficial effects:

[0079] The present invention obtains the power consumption data of the game controller in different game scenarios, establishes a power consumption model, analyzes the impact of players' operating habits on power consumption, classifies tags according to operating habits, analyzes the players' behavior patterns and the changing trends of game scenarios in real time, pre-sets the time interval threshold of the standby mode, and monitors the vibration, lighting and key operation frequencies of the game controller in real time when the game scenario changes, determines the power consumption trend, and triggers the low-power mode when the consumption trend is lower than the preset threshold to reduce the power supply of vibration and lighting;

[0080] The present invention also dynamically adjusts the time interval threshold of the standby mode through a machine learning algorithm and adjusts the power supply intensity according to the real-time data of the battery life to ensure that the game experience is not affected, realizes the intelligent management of the game controller power supply, effectively extends the battery life, and at the same time ensures a good game experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] Figure 1 It is a flowchart of the steps of a game controller power management method in Embodiment 1;

[0082] Figure 2 It is a framework diagram of a game controller power management in Embodiment 2. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0083] To make the objectives, features and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the embodiments described below are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0084] Embodiment 1

[0085] As Figure 1 shown, this embodiment provides a game controller power management method, including the following steps:

[0086] S1. Obtain the power consumption data of the game controller in different game scenarios and construct a power consumption model;

[0087] S2. Analyze the impact of players' operating habits on power consumption according to the power consumption model and generate operating habit classification tags;

[0088] S3. Dynamically adjust the time interval threshold of the standby mode according to the operating habit classification tags and the changing trend of the game scenario;

[0089] S4. Monitor the vibration function usage, light flashing frequency, and button operation frequency of the gamepad when the game scene switches, and judge the power consumption trend of the current scene;

[0090] S5. If the power consumption trend of the current scene is lower than the preset threshold, trigger the low-power mode and reduce the power supply for the vibration function and light flashing;

[0091] S6. According to the time interval threshold of the standby mode, combine the environmental information where the gamepad is located to judge whether to exit the low-power standby mode;

[0092] S7. In the low-power standby mode, monitor the operation signals of the gamepad, and allocate power resources according to the current game scene and the player's operation habits.

[0093] Further, the process of constructing the power consumption model in step S1 is specifically as follows:

[0094] Obtain the vibration frequency, light flashing frequency, and button operation frequency data of the gamepad in different game scenes, and preprocess the data;

[0095] According to the preprocessed power consumption data of the gamepad, construct an initial power consumption model, take the vibration frequency, light flashing frequency, and button operation frequency as independent variables, and power consumption as the dependent variable, and fit the coefficients of the regression model by the least squares method;

[0096] Evaluate the initial power consumption model, judge the fitting degree and prediction ability of the model by calculating the mean square error and determination coefficient indicators. If the model performance does not meet the requirements, introduce a regularization term to control the model complexity, and try to use a support vector machine. By comparing the performance of different algorithms, select the modeling method;

[0097] Obtain the power consumption data of the gamepad in different game scenes, including the vibration function usage frequency, light flashing frequency, and button operation frequency, and preprocess the obtained data, including operations such as data cleaning and normalization, for subsequent modeling use;

[0098] According to the preprocessed power consumption data, establish a power consumption model of the gamepad, obtain the model parameters through training, and determine the power consumption law of the gamepad in different game scenes;

[0099] Obtain the trigger signal for game scene switching. If a scene switching signal is detected, start to monitor the vibration frequency, light flashing frequency, and button operation frequency of the gamepad in real time, and input the real-time monitoring data into the power consumption model;

[0100] According to the preset vibration frequency threshold, light flashing frequency threshold, and key operation frequency threshold, determine whether the current gamepad vibration frequency, light flashing frequency, and key operation frequency exceed the thresholds. If at least one exceeds the preset threshold, determine that the current game scenario is a high power consumption scenario;

[0101] For the game scenario determined to be a high power consumption scenario, use the support vector machine algorithm. Based on the gamepad vibration frequency, light flashing frequency, and key operation frequency data of historical high power consumption scenarios, train a high power consumption scenario prediction model to obtain the prediction model parameters;

[0102] When the game scenario switches, input the current gamepad vibration frequency, light flashing frequency, and key operation frequency into the high power consumption scenario prediction model, and obtain the probability that the current scenario is a high power consumption scenario through model prediction;

[0103] According to the probability of the high power consumption scenario, set different power consumption trend levels, calculate the power consumption trend level of the current game scenario through the membership function, and feedback the power consumption trend level to the game. The game, based on the power consumption trend level, dynamically optimizes the power consumption of the game by adjusting parameters such as the special effect rendering intensity and background music sound quality in the game scenario;

[0104] Continuously monitor the power consumption data of the gamepad, continuously update the power consumption model and the high power consumption scenario prediction model based on the new data, and improve the accuracy and adaptability of the model through incremental learning to achieve dynamic optimization of the gamepad power consumption;

[0105] Specifically, as described in the following example, when obtaining the gamepad power consumption data, data such as the gamepad vibration frequency, light flashing frequency, and key operation frequency can be collected through sensors, for example, collecting 10 times per second and continuously collecting for 5 minutes to obtain 3000 data points;

[0106] Then clean the data, remove outliers and invalid data, and perform normalization processing on the data to scale the numerical range to between 0 and 1;

[0107] When establishing the power consumption model, the support vector machine algorithm can be selected, the Gaussian kernel function can be used, the penalty coefficient C can be set to 10, and the model parameters can be optimized through grid search to finally obtain a power consumption model with an accuracy rate of 95%;

[0108] When the game scenario switches, detect the scene switching characteristics through the image recognition algorithm. For example, when the scene brightness change exceeds 30%, it is determined that the scene has switched. According to the preset thresholds, such as the vibration frequency exceeding 5 times per second, the light flashing frequency exceeding 3 times per second, and the key operation frequency exceeding 10 times per second, it is determined that it is a high power consumption scenario;

[0109] Extract features from the data of high power consumption scenarios, select features such as the mean vibration frequency, the mean light flashing frequency, and the mean key operation frequency, train a prediction model for high power consumption scenarios, use the support vector machine algorithm, and obtain a prediction model with an accuracy of 90% through 5-fold cross-validation;

[0110] When switching scenarios, input the current features into the prediction model to obtain the probability of high power consumption scenarios. If the probability exceeds 0.8, it is determined as a high power consumption scenario. According to the probability of high power consumption scenarios, set fuzzy logic rules, such as a high power consumption trend when the probability is between 0.8 and 1, a medium power consumption trend when the probability is between 0.6 and 0.8, and a low power consumption trend when the probability is between 0.4 and 0.6. Calculate the power consumption trend level through the membership function. Finally, the game dynamically adjusts game parameters according to the power consumption trend level, such as reducing the special effect rendering intensity by 30% and the background music sound quality by 20% in the case of a high power consumption trend to optimize the power consumption of the game.

[0111] Further, the process of generating operation habit classification labels in step S2 is specifically as follows:

[0112] According to the power consumption model, obtain the operation record data of players. For the operation record data of each player, count the frequencies of various operations to obtain the operation frequency distribution;

[0113] According to the operation frequency distribution, judge whether each type of operation is a high-frequency operation or a low-frequency operation. If the operation frequency is higher than the preset threshold, it is determined as a high-frequency operation; otherwise, it is determined as a low-frequency operation;

[0114] For the high-frequency operations and low-frequency operations, extract operation features respectively to obtain a high-frequency operation feature set and a low-frequency operation feature set;

[0115] According to the high-frequency operation feature set and the low-frequency operation feature set, classify the operation habits of players to obtain the operation habit classification result;

[0116] For each operation habit classification, count the average power consumption of players in this classification to obtain the power consumption levels of each operation habit classification;

[0117] According to the power consumption levels of each operation habit classification, use the decision tree algorithm to generate operation habit classification labels to obtain an operation habit classification label set, and match the operation habit classification results of players with the operation habit classification label set to obtain the operation habit classification labels of each player;

[0118] Classify tags according to players' operation habits, use the association rule algorithm to mine frequent operation patterns under different operation habits, obtain the mapping relationship between operation habits and frequent operation patterns, store the mapping relationship between operation habits and frequent operation patterns in the knowledge base. When the operation record data of new players arrives, it can quickly judge the type of their operation habits and predict their possible frequent operation patterns, so as to realize the real-time analysis of the impact of operation habits on power consumption;

[0119] Specifically, as described in the following example, according to the pre-established power consumption model, obtain the operation record data of 1000 players for a continuous week. For the operation record data of each player, count the frequency of various operations to obtain the operation frequency distribution;

[0120] According to the operation frequency distribution, set the frequency threshold to 0.6. If the operation frequency is higher than 0.6, it is determined as a high-frequency operation, otherwise it is determined as a low-frequency operation. For high-frequency operations and low-frequency operations, extract 100-dimensional operation feature vectors respectively to obtain the high-frequency operation feature set and the low-frequency operation feature set;

[0121] According to the high-frequency operation feature set and the low-frequency operation feature set, use the K-Means clustering algorithm to classify players' operation habits, set the number of clusters K = 4, and obtain 4 types of operation habit classification results. For each operation habit classification, count the average power consumption of players in this classification to obtain the power consumption levels of each operation habit classification;

[0122] According to the power consumption levels of each operation habit classification, use the CART decision tree algorithm to generate operation habit classification tags to obtain the operation habit classification tag set, including heavy players, medium players, light players and casual players. Match the operation habit classification results of players with the operation habit classification tag set to obtain the operation habit classification tags of each player;

[0123] According to the operation habit classification tags of players, use the Apriori association rule algorithm to mine frequent operation patterns under different operation habits, set the minimum support to 0.5 and the minimum confidence to 0.8, and obtain the mapping relationship between operation habits and frequent operation patterns, such as heavy players are used to playing continuously for more than 3 hours, medium players are used to playing 1-2 hours each time, etc.;

[0124] Store the mapping relationship between operation habits and frequent operation patterns in the Redis knowledge base. When the operation record data of new players arrives, by quickly judging the type of their operation habits and combining the mapping relationship in the knowledge base, predict their possible frequent operation patterns, so as to realize the real-time analysis of the impact of operation habits on power consumption.

[0125] Furthermore, the process of dynamically adjusting the time interval threshold of the standby mode in step S3 is specifically as follows:

[0126] Obtain the real-time operation data of the player according to the pre-established correspondence between the operation frequency and the time interval threshold, and determine the current operation frequency of the player through data analysis;

[0127] Judge whether the operation frequency of the player belongs to high-frequency operation or low-frequency operation. If it is high-frequency operation, set the time interval of the standby mode to a longer interval. If it is low-frequency operation, set the time interval of the standby mode to a shorter interval;

[0128] Use a clustering algorithm to classify the operation habits of the player to obtain typical behavior patterns under different operation habits;

[0129] For each typical behavior pattern, analyze its correlation with the change of the game scene through the association rule mining algorithm to obtain the scene change trend under different behavior patterns;

[0130] Dynamically adjust the time interval threshold of the standby mode according to the scene change trend;

[0131] Apply the adjusted time interval setting to the standby mode to realize the adaptive adjustment of the standby mode time interval according to the dynamic changes of the player's operation habits and the game scene. In the standby mode, continuously monitor the player's operation behavior and the change of the game scene. If it is detected that the player re-enters high-frequency operation or the scene changes significantly, the device will be immediately awakened and the standby mode will be exited;

[0132] According to the actual operation situation of the player and the game experience feedback, use the reinforcement learning algorithm to continuously optimize the trigger conditions and time interval thresholds of the standby mode to form an adaptive game energy-saving strategy, which can minimize the device power consumption while ensuring the player experience;

[0133] Specifically, as described in the following example, the players are divided into three categories: casual, moderate, and heavy by the K-means clustering algorithm. Among them, the operation frequency of casual players is less than 10 times per minute, the operation frequency of moderate players is between 10 and 30 times per minute, and the operation frequency of heavy players is higher than 30 times per minute;

[0134] Use the Apriori association rule mining algorithm to analyze the operation behaviors of different types of players in different game scenes, and find that the average operation interval of casual players is 15 seconds in the exploration scene and is shortened to 5 seconds in the combat scene; the average operation interval of moderate players is 10 seconds in the exploration scene and is shortened to 3 seconds in the combat scene; the average operation interval of heavy players is 5 seconds in the exploration scene and is shortened to 1 second in the combat scene;

[0135] According to the real-time operation data of the player, calculate the current operation frequency of the player. If the operation frequency is lower than 15 times per minute, it is determined as low-frequency operation, and the standby time interval is set to 30 seconds; if the operation frequency is higher than 15 times per minute, it is determined as high-frequency operation, and the standby time interval is set to 60 seconds.

[0136] Through the decision tree algorithm, comprehensively analyze the player's operation frequency, behavior pattern and scene change trend. When the player is in the low-frequency operation state for 5 consecutive minutes and the scene has not changed, extend the standby time interval to 2 minutes; when the player's operation frequency suddenly increases and the scene changes frequently, shorten the standby time interval to 10 seconds.

[0137] In the standby mode, if it is detected that the player's operation frequency suddenly increases by more than 20 times per minute or the scene changes, immediately wake up the device and reset the standby time interval to the default value. Adopt the Q-learning reinforcement learning algorithm, and continuously adjust the trigger threshold and time interval of the standby mode according to the player's actual operation situation and game experience feedback. Through 1000 rounds of iterative training, obtain the optimal standby strategy, while ensuring the experience of 95% of the players, reduce the device power consumption by 20%.

[0138] Furthermore, the process of judging the power consumption trend of the current scene in step S4 is specifically as follows:

[0139] Obtain the trigger signal for game scene switching. If a scene switching signal is detected, start real-time monitoring of the handle vibration frequency, light flashing frequency and button operation frequency.

[0140] According to the preset vibration frequency threshold, light flashing frequency threshold and button operation frequency threshold, judge whether the current handle vibration frequency, light flashing frequency and button operation frequency exceed the threshold.

[0141] If at least one of the handle vibration frequency, light flashing frequency and button operation frequency exceeds the preset threshold, judge that the current game scene is a high power consumption scene.

[0142] Among them, for the game scene determined to be a high power consumption, adopt the support vector machine algorithm, and train the high power consumption scene prediction model according to the handle vibration frequency, light flashing frequency and button operation frequency data of the historical high power consumption scene.

[0143] When the game scene switches, input the current handle vibration frequency, light flashing frequency and button operation frequency into the high power consumption scene prediction model to obtain the probability that the current scene is a high power consumption scene.

[0144] Set different power consumption trend levels according to the probability of high power consumption scenarios, obtain the power consumption trend level of the current game scenario, and feedback the power consumption trend level to the game. By adjusting the special effect rendering intensity, background music sound quality, etc. in the game scenario, dynamically optimize the power consumption of the game;

[0145] Obtain the power consumption data of the game controller in different game scenarios, including the usage frequency of the vibration function, the flashing frequency of the light, and the key operation frequency, establish a power consumption model, and dynamically adjust the vibration intensity, light brightness, and key sensitivity of the game controller according to the power consumption model to reduce the power consumption of the controller and extend the usage time of the controller;

[0146] Specifically, as described in the following example, when the game scenario switches, the vibration frequency, light flashing frequency, and key operation frequency of the controller are monitored in real time and compared with the preset thresholds. For example, the vibration frequency threshold is set to 100 times per minute, the light flashing frequency threshold is set to 50 times per minute, and the key operation frequency threshold is set to 200 times per minute. If it is detected that any frequency exceeds the threshold, it is determined as a high power consumption scenario. According to historical data, such as a scenario with a vibration frequency of 120 times per minute, a light flashing frequency of 60 times per minute, and a key operation frequency of 250 times per minute, use the support vector machine algorithm to train a high power consumption scenario prediction model;

[0147] After the frequency data of the current scenario is input into the model, the probability of being a high power consumption scenario is obtained, such as 0.8. Then, using the fuzzy logic algorithm, map 0.8 to the preset power consumption trend level, such as "high". This level is feedback to the game, triggering optimization measures such as reducing the special effect rendering intensity by 20% and reducing the background music sound quality by 30%. At the same time, the power consumption data in different scenarios will also be collected to establish a power consumption model. For example, in a high power consumption scenario, the vibration intensity is reduced by 15%, the light brightness is reduced by 10%, and the key sensitivity is reduced by 5% to extend the usage time of the controller.

[0148] Further, the process of triggering the low power consumption mode in step S5 is specifically as follows:

[0149] Obtain the power consumption data in the current scenario, compare it with the preset threshold. If the power consumption is lower than the threshold, trigger the low power consumption mode;

[0150] According to the preset low power consumption strategy, determine the reduction amplitude of the vibration function and the light flashing effect, and achieve function degradation by controlling the power supply;

[0151] Gradually reduce the vibration frequency and light brightness until the lowest power consumption level is reached, and at the same time monitor the change trend of power consumption;

[0152] The functional relationships between power consumption, vibration frequency, and light brightness are fitted by the least squares method to obtain the frequency reduction and dimming schemes;

[0153] Using the decision tree algorithm, the triggering threshold of the low-power mode is dynamically adjusted by comprehensively considering the current scene characteristics and user usage habit factors;

[0154] A Markov model is constructed. Based on historical power consumption data, the power consumption trend in the next period of time is predicted to make energy-saving decisions;

[0155] The user experience feedback in the low-power mode is continuously recorded, and the frequency reduction and dimming strategies are continuously optimized using the reinforcement learning algorithm to seek a balance between energy saving and experience. If a game scene switching signal is detected, the vibration frequency, light flashing frequency, and button operation frequency of the handle are monitored in real time to determine whether the current scene is a high-power consumption scene;

[0156] According to the probability of the high-power consumption scene, different power consumption trend levels are set, and the power consumption trend levels are fed back to the game. By adjusting the special effect rendering intensity, background music sound quality, etc. in the game scene, the power consumption of the game is dynamically optimized;

[0157] Specifically, as described in the following example, when the power consumption data is lower than the preset 80% threshold, the low-power mode is triggered;

[0158] According to the preset strategy, the vibration frequency is reduced by 20%, and the light brightness is reduced by 30%. The function degradation is achieved through PWM control;

[0159] The vibration frequency and light brightness are gradually reduced at a learning rate of 0.05 until the power consumption drops below 50%. At the same time, the power consumption change trend is monitored every 100 ms. The functional relationships between power consumption, vibration frequency, and light brightness are fitted by the least squares method to obtain the optimal scheme when the vibration frequency drops to 60 Hz and the light brightness drops to 40%;

[0160] Using the decision tree algorithm, by comprehensively considering factors such as the current scene brightness and user operation frequency, the triggering threshold of the low-power mode is dynamically adjusted to between 75% and 85%;

[0161] A Markov model is constructed. Based on the power consumption data in the past 1 hour, the power consumption trend in the next 30 minutes is predicted. If the predicted power consumption drops below 60%, the energy-saving decision is started 5 minutes in advance;

[0162] Continuously record the user operation feedback in the low-power mode. Use the Q-Learning reinforcement learning algorithm, with energy consumption and user experience as the reward functions, to continuously optimize the frequency reduction and dimming strategies, and seek a balance between energy conservation and experience. If a game scene switch signal is detected, monitor the vibration frequency of the handle, the flashing frequency of the light, and the key operation frequency in real time. When any one of the vibration frequency exceeds 5Hz, the flashing frequency of the light exceeds 2Hz, and the key operation frequency exceeds 1 time per second, it is determined as a high power consumption scene. According to the probability of the high power consumption scene, set three power consumption trend levels: low, medium, and high, and feedback the level to the game. By adjusting the special effect rendering intensity, background music sound quality, etc. in the game scene, dynamically optimize the power consumption of the game.

[0163] Further, the process of combining the environmental information of the handle to determine whether to exit the low-power standby mode in step S6 is specifically as follows:

[0164] Obtain the environmental information where the handle is currently located, including temperature, humidity, light intensity, and noise data;

[0165] According to the threshold ranges of the environmental data in each dimension set in advance, determine whether the current environment where the handle is located meets the normal working conditions;

[0166] If the current environment does not meet the normal working conditions of the handle, start the low-power standby mode of the handle;

[0167] In the low-power standby mode, periodically collect the handle environmental data, and predict the environmental state of the handle in the next period of time according to the change trend of the environmental data;

[0168] Establish an association model between the environmental data and the handle power consumption through a machine learning algorithm, and dynamically adjust the standby time threshold of the low-power mode;

[0169] When the continuous standby duration of the handle exceeds the standby time threshold, and the environmental prediction result indicates that the handle will still be in an unavailable environment in the next period of time, control the handle to exit the low-power standby mode and enter the sleep state;

[0170] If the environmental prediction result meets the normal working conditions of the handle, and the current standby duration does not exceed the threshold, exit the low-power standby mode and restore the normal working state of the handle;

[0171] Obtain the historical operation data and real-time power consumption data of the handle, and construct a training data set for the training and optimization of the machine learning algorithm;

[0172] According to the actual usage of the handle and the feedback of power consumption, an enhanced learning algorithm is adopted to continuously optimize the trigger conditions and time interval thresholds of the standby mode, forming an adaptive power management strategy to maximize the battery life of the handle while ensuring the usage experience;

[0173] Specifically, as described in the following example, the handle collects environmental data in real time through built-in temperature and humidity sensors, light sensors, and noise sensors. For example, the temperature range is -10°C to 50°C, the humidity range is 20% to 80%, the light intensity range is 100 lux to 1000 lux, and the noise range is 30 dB to 80 dB;

[0174] According to the preset temperature threshold of -5°C to 45°C, humidity threshold of 30% to 70%, light intensity threshold of 200 lux to 800 lux, and noise threshold of 40 dB to 70 dB, it is judged whether the current environment meets the normal working conditions of the handle. If the environmental data exceeds the threshold range, such as the temperature is -8°C, the humidity is 85%, the light intensity is 50 lux, and the noise is 90 dB, the low-power standby mode is triggered;

[0175] In the standby mode, the handle collects environmental data every 10 minutes, analyzes the data change trend through the moving average algorithm, and predicts the environmental state within the next 30 minutes. At the same time, a correlation model between environmental data and the power consumption of the handle is established using the decision tree algorithm. According to the statistical analysis of historical data, in an environment where the temperature is below -5°C, the humidity is above 80%, the light intensity is below 100 lux, and the noise is above 80 dB, the average power consumption of the handle is 0.5 W. Accordingly, the standby time threshold is dynamically adjusted to 30 minutes;

[0176] When the continuous standby duration of the handle exceeds 30 minutes and the environmental prediction result shows that the handle will still be in an unavailable environment within the next 30 minutes, the handle is controlled to enter the deep sleep state, and the power consumption is reduced to 0.1 W. If the environmental prediction result shows that the handle can resume normal operation within the next 30 minutes and the current standby duration does not exceed 30 minutes, the low-power standby mode is exited and the normal working state is restored;

[0177] By sampling the historical operation data and real-time power consumption data of the handle, a training set and a test set are constructed, and the gradient boosting decision tree algorithm is used for training to obtain an optimized low-power standby strategy;

[0178] According to the actual usage of the handle and the feedback of power consumption, the Q-Learning enhanced learning algorithm is adopted. With the goal of extending the battery life and ensuring the usage experience, the low-power standby strategy is dynamically adjusted through continuous attempts and learning to achieve the adaptive optimization of the power management of the handle.

[0179] Further, the process of allocating power resources in step S7 is specifically as follows:

[0180] Obtain the operation signal of the handle, and determine whether the handle is in the low-power standby mode. If so, enter the continuous monitoring state; otherwise, maintain the normal working state.

[0181] In the continuous monitoring state, use the convolutional neural network algorithm to extract features from the handle operation signal to obtain the operation feature vector.

[0182] According to the pre-established game scene model and player operation habit model, classify the operation feature vector to determine the current game scene type and player operation habit type.

[0183] For different game scene types and player operation habit types, use the decision tree algorithm to generate corresponding power resource allocation strategies.

[0184] Convert the power resource allocation strategy into a control instruction and send it to the power management module of the handle through the communication interface of the handle.

[0185] The power management module adjusts the working voltage and clock frequency of each hardware module according to the received control instruction.

[0186] When the handle ends the low-power standby mode, restore the default working parameters of each hardware module.

[0187] Obtain the historical operation data and real-time power consumption data of the device, and construct a training data set for continuous training and optimization of the machine learning algorithm.

[0188] Through continuous training and iterative optimization of the machine learning algorithm, continuously improve the intelligent level of the power management strategy.

[0189] Specifically, as described in the following example, the operation signal of the handle can be obtained by a sensor with a sampling frequency of 1000Hz. If the average value of the sampling data is lower than 0.1, it is determined that the handle is in the low-power standby mode and enters the continuous monitoring state. In the continuous monitoring state, use a 5-layer convolutional neural network to extract features from the handle operation signal to obtain a 128-dimensional operation feature vector.

[0190] According to the pre-established model containing 100 game scenes and 50 player operation habits, classify the operation feature vector to determine the current game scene type and player operation habit type. For different game scene types and player operation habit types, use a decision tree algorithm with a maximum depth of 5 to generate corresponding power resource allocation strategies. For example, for shooting games and aggressive players, allocate 80% of the power resources to the GPU and CPU.

[0191] Convert the power resource allocation strategy into a 16-byte control instruction and send it to the power management module of the handle through a 2.4GHz wireless communication interface. According to the received control instruction, the power management module adjusts the working voltage of the GPU to 1.2V and the clock frequency to 1.5GHz to achieve intelligent allocation of power resources. When the handle exits the low-power standby mode, the working voltage and clock frequency of each hardware module are restored to the default values. For example, the default working voltage of the GPU is 0.9V and the default clock frequency is 1GHz;

[0192] Obtain the operation data and real-time power consumption data of the device in the past week, and construct a training data set containing 10,000 records to support the continuous training and optimization of the support vector machine and reinforcement learning algorithms;

[0193] Through the continuous training and iterative optimization of the machine learning algorithm, while ensuring that the response time of the handle is less than 50 milliseconds, reduce the average standby power consumption of the handle by 20%, maximize the battery life of the device, and ensure the user experience at the same time;

[0194] Furthermore, the method of the present invention further includes:

[0195] S8. Dynamically adjust the time interval threshold of the standby mode, and optimize the power management strategy by combining historical operation data and current power consumption trends;

[0196] The process of optimizing the power management strategy in step S8 is specifically as follows:

[0197] Obtain the historical operation data and real-time power consumption data of the device, and construct a training data set;

[0198] Adopt a decision tree algorithm to dynamically adjust the time interval threshold of the standby mode according to the historical operation data and real-time power consumption trend, and obtain optimized threshold parameters;

[0199] Through a support vector machine algorithm, comprehensively analyze historical data and real-time trends, judge the usage status and power consumption level of the current device, and determine whether to trigger the standby mode;

[0200] If the idle time of the current device exceeds the optimized time interval threshold and the power consumption level is low, automatically enter the standby mode to reduce power consumption;

[0201] In the standby mode, continuously monitor the operation behavior and power consumption changes of the device. If a user operation or a sudden increase in power consumption is detected, wake up the device and exit the standby mode;

[0202] According to the actual usage of the device and the feedback of power consumption, an enhanced learning algorithm is adopted to continuously optimize the triggering conditions and time interval thresholds of the standby mode, forming an adaptive power management strategy;

[0203] Through continuous training and iterative optimization of the machine learning algorithm, the intelligence level of the power management strategy is continuously improved, maximizing the battery life of the device while ensuring the usage experience;

[0204] A clustering algorithm is used to classify the operation habits of users, obtaining typical behavior patterns under different operation habits. For each typical behavior pattern, the association rule mining algorithm is used to analyze its correlation with the changes in the device usage scenarios, obtaining the scenario change trends under different behavior patterns;

[0205] The time interval threshold of the standby mode is dynamically adjusted according to the scenario change trend. When the trend indicates that the user may enter a high-frequency operation scenario, the time interval is extended accordingly. When the trend indicates that the user may enter a low-frequency operation scenario, the time interval is shortened accordingly, realizing the adaptive adjustment of the standby mode time interval according to the dynamic changes of the user operation habits and usage scenarios, and further optimizing the power management strategy;

[0206] Specifically, as described in the following example, by collecting the operation data and power consumption records of the device in the past month, sampling once every 5 minutes, a training data set containing 10,000 samples is constructed. Using the ID3 decision tree algorithm, with the operation frequency and power consumption level as features and whether to enter standby as the target variable, the optimal splitting attribute is selected through the information gain ratio, and the decision tree is recursively constructed to obtain the optimal parameters when the standby time interval threshold is 10 minutes;

[0207] The SVM support vector machine algorithm is adopted, with the operation data and power consumption data in the recent 1 hour as the input. The data is mapped to a high-dimensional space through the Gaussian kernel function, and the maximum margin hyperplane is searched to judge the current device state. If the idle time exceeds 10 minutes and the power consumption level is lower than 20%, the device will automatically enter the standby mode;

[0208] In the standby mode, the device state is detected once every 30 seconds. If there is a user operation or a sudden increase in power consumption exceeding 30%, the device will be immediately awakened;

[0209] According to the device usage situation and power consumption feedback, the Q-Learning enhanced learning algorithm is adopted, with the standby time interval and triggering conditions as actions and the device usage experience and battery life as rewards. Through continuous trial and error and learning, the standby strategy is optimized, such as shortening the time interval to 5 minutes and adjusting the triggering condition to the idle time exceeding 5 minutes and the power consumption level being lower than 15%;

[0210] Through 7 consecutive days of algorithm training and parameter tuning, the accuracy rate of the power management strategy has increased from 85% to 95%, the average standby time of the device has been extended by 20%, and the battery life has been improved by 10%. The K-Means clustering algorithm is used to cluster the operation habits of 1000 users with the operation frequency and duration within a week as features, and three typical behavior patterns of light use, moderate use, and heavy use are obtained. For each behavior pattern, through the Apriori association rule mining algorithm, the relevance between operation habits and usage scenarios is analyzed. It is found that in the light use pattern, users often enter long-term standby at night, while in the heavy use pattern, users frequently operate the device during the day and rarely enter standby;

[0211] According to the association rules, the standby time interval is dynamically adjusted. For example, in the light use pattern, the interval is extended to 30 minutes, and in the heavy use pattern, the interval is shortened to 2 minutes to achieve adaptive power management.

[0212] In this embodiment, it further includes:

[0213] S9. According to the real-time data of the battery life, combined with the power management strategy, dynamically adjust the power supply intensity of the vibration function and the light flash to ensure that the gaming experience is not affected;

[0214] Among them, the process of dynamically adjusting the power supply intensity of the vibration function and the light flash is specifically as follows:

[0215] Obtain the real-time battery life data, including parameters such as the current battery level, voltage, and current. Through machine learning algorithms such as support vector machines or neural networks, combined with historical data, predict the battery life in different usage scenarios;

[0216] According to the predicted battery life, dynamically adjust the supply voltage and current of the vibration motor and the LED lamp. When the battery life is short, reduce the supply intensity to weaken the vibration and light effects; when the battery life is sufficient, increase the supply intensity to enhance the vibration and light effects to ensure the gaming experience;

[0217] Through machine learning algorithms such as decision trees or association rule mining, analyze the preferences of players for vibration and light effects in different gaming scenarios, and adjust the power supply strategy accordingly to appropriately enhance the effects in key scenarios and improve the gaming experience;

[0218] Establish a power management strategy knowledge base, set corresponding vibration and light power supply strategies for different battery life states and gaming scenarios, and achieve rapid matching and dynamic adjustment of strategies;

[0219] Real-time monitor the battery level change during the game process. When the battery level is lower than the preset threshold, trigger the adjustment of the power management strategy, reduce the vibration and light effects, and extend the game time until the battery runs out;

[0220] Record the battery life data and power management strategies during each game session. Through machine learning algorithms such as reinforcement learning, continuously optimize the strategies to maximize the battery life while ensuring the gaming experience.

[0221] Evaluate the effectiveness of the power management strategy and its impact on vibration and lighting effects through user feedback and game data analysis. Continuously improve the algorithm model and strategy knowledge base to achieve continuous optimization of power management. If the idle time of the current device exceeds the optimized time interval threshold and the power consumption level is low, automatically enter the standby mode to reduce power consumption.

[0222] In the standby mode, continuously monitor the operation behavior and power consumption changes of the device. If a user operation or sudden increase in power consumption is detected, immediately wake up the device and exit the standby mode.

[0223] Through continuous training and iterative optimization of machine learning algorithms, continuously improve the intelligence level of the power management strategy, maximize the battery life of the device, and ensure the usage experience at the same time.

[0224] Specifically, as described in the following example, by obtaining the real-time battery life data, including parameters such as the current battery level, voltage, and current, using the support vector machine algorithm, combined with historical data, predict the battery life in different usage scenarios. For example, when the current is 500 mA and the voltage is 3.7 V, the predicted battery life is 3 hours.

[0225] According to the predicted battery life, dynamically adjust the supply voltage and current of the vibration motor and LED lights. When the battery life is less than 2 hours, reduce the supply voltage of the vibration motor from 3.3 V to 2.5 V, and reduce the supply current of the LED lights from 20 mA to 10 mA; when the battery life exceeds 4 hours, increase the supply voltage of the vibration motor to 3.7 V and increase the supply current of the LED lights to 30 mA.

[0226] Analyze the preferences of players for vibration and lighting effects in different game scenarios through the decision tree algorithm. For example, in the battle scenario, 80% of the players prefer strong vibration effects and bright lighting effects. Therefore, in this scenario, increase the supply voltage of the vibration motor by 20% and increase the supply current of the LED lights by 30%.

[0227] Establish a power management strategy knowledge base, and set corresponding vibration and lighting power supply strategies for different battery life states and game scenarios. For example, when the battery level is less than 30% and in the exploration scenario, reduce the supply voltage of the vibration motor by 30% and reduce the supply current of the LED lights by 50%.

[0228] Monitor the change of battery power in real time during the game. When the power is lower than 20%, trigger the adjustment of the power management strategy, reduce the supply voltage of the vibration motor by 50% and the supply current of the LED lamp by 80% to extend the game time;

[0229] Record the battery life data and power management strategy during each game process, and optimize the strategy through the reinforcement learning algorithm. For example, through the Q-learning algorithm, continuously adjust the weights of the vibration and lighting effects and the battery life time to maximize the extension of the battery life while ensuring the game experience;

[0230] Evaluate the effectiveness of the power management strategy and its impact on the vibration and lighting effects through user feedback and game data analysis. Use the neural network algorithm to continuously improve the strategy knowledge base and achieve continuous optimization of power management. If the idle time of the current device exceeds 10 minutes and the power consumption level is lower than 50mW, automatically enter the standby mode to reduce power consumption;

[0231] In the standby mode, monitor the operation behavior and power consumption change of the device every 30 seconds. If it detects user operation or a sudden increase in power consumption exceeding 100mW, immediately wake up the device and exit the standby mode;

[0232] Through continuous training and iterative optimization of the machine learning algorithm, automatically update the power management strategy every 7 days to continuously improve the intelligence level of the strategy, maximize the extension of the device's battery life, and ensure the usage experience at the same time.

[0233] This embodiment has achieved significant technical effects through refined data analysis and intelligent decision-making. First of all, by constructing an accurate power consumption model, the system can accurately predict the power consumption in different game scenarios, providing a scientific basis for subsequent power management strategies. This model is based on the multiple linear regression algorithm and is continuously optimized by combining techniques such as cross-validation to ensure the reliability and accuracy of the prediction results.

[0234] Secondly, this method can deeply analyze the operation habits of players, generate operation habit classification labels, and thus achieve personalized power management. The system dynamically adjusts the time interval threshold of the standby mode according to the operation frequency and habits of different players to ensure that the power consumption is minimized without affecting the game experience. For example, for players with high-frequency operations, the system will appropriately extend the standby time interval to avoid power waste caused by frequent wake-up; while for players with low-frequency operations, the system will shorten the standby time interval and enter the low-power mode in time to save power.

[0235] In addition, by continuously monitoring the vibration frequency, light flashing frequency, and button operation frequency of the gamepad when the game scene switches, the system can quickly determine the power consumption trend of the current scene. Once it detects that the power consumption is lower than the preset threshold, the system will automatically trigger the low-power mode, reducing the power supply for the vibration function and light flashing, and further optimizing power management. At the same time, the system will also determine whether to exit the low-power standby mode based on the environmental information to ensure that the gamepad can maintain the best power management state in different environments.

[0236] In the low-power standby mode, the system continuously monitors the operation signals of the gamepad and intelligently allocates power resources according to the current game scene and the player's operation habits. Through the convolutional neural network algorithm and decision tree algorithm, the system can generate corresponding power resource allocation strategies to ensure the normal operation of the basic functions of the gamepad while minimizing power consumption. In addition, through the continuous training and optimization of the machine learning algorithm, the system will continuously improve the intelligence level of the power management strategy to adapt to the changing usage scenarios and player needs.

[0237] In summary, through a series of innovative technologies and strategies, this embodiment realizes the efficient and intelligent management of the power supply of the gamepad, not only extending the usage duration of the gamepad, improving the power utilization efficiency, but also enhancing the player's gaming experience, with important practical application value.

[0238] Embodiment 2

[0239] As Figure 2 shown, this embodiment provides a gamepad power management system, including a power management module, a data acquisition and processing module, and a data analysis and model construction module;

[0240] The power management module controls the turning on, turning off, and switching of different power consumption states of the gamepad power supply according to the triggering conditions and exiting conditions of the low-power mode;

[0241] According to the preset low-power strategy, reduce the power supply for the vibration function and light flashing, and adjust the working voltage and clock frequency of each hardware module;

[0242] Dynamically adjust the time interval threshold of the standby mode to optimize the power management strategy;

[0243] The data acquisition and processing module acquires the power consumption-related data of the gamepad in different game scenes, as well as the player's operation record data and the environmental information where the gamepad is located, and preprocesses the acquired data;

[0244] Transmit the processed data to the power management module and the data analysis and model construction module to provide data support for the formulation and optimization of the power management strategy;

[0245] The data analysis and model construction module constructs an initial power consumption model based on the preprocessed game controller power consumption data and evaluates the model performance;

[0246] Based on the power consumption model, analyze the impact of players' operation habits on power consumption and count the power consumption levels of each category;

[0247] Analyze the correlation between players' operation habits and changes in game scenarios;

[0248] Construct a Markov model to assist the power management module in making energy-saving decisions;

[0249] Establish a correlation model between environmental data and the power consumption of the controller to support the dynamic adjustment of the standby time threshold for the low-power mode.

[0250] In this embodiment, with precise power consumption modeling and real-time dynamic adjustment strategies, the intelligent level of power management has been significantly improved, the power consumption of the controller has been effectively reduced, and the standby and usage durations have been greatly extended. The system comprehensively considers environmental factors and players' operation habits to provide personalized power management solutions, achieving energy conservation while ensuring the game experience, enhancing the system's adaptability and stability, and continuously optimizing strategies to cope with usage changes, thus significantly improving the user experience and satisfaction, enabling players to enjoy a more stable, lasting, and personalized game control experience without worrying about battery life when playing games.

[0251] The specific embodiments of the invention have been described in detail above, but they are only examples. The invention is not limited to the specific embodiments described above. Those skilled in the art should understand that the above embodiments and the descriptions in the specification only illustrate the principles of the invention. Without departing from the spirit and scope of the invention, the invention will have various changes and improvements, and these changes and improvements fall within the scope of the invention claimed. The scope of the invention claimed is defined by the appended claims and their equivalents.

Claims

1. A game controller power management method, characterized in that: The following steps are involved: S1. Obtain the power consumption data of the game controller in different game scenarios and build a power consumption model; S2. Analyze the impact of players’ operating habits on power consumption based on the power consumption model and generate operating habit classification labels; S3, dynamically adjusting the time interval threshold of the standby mode according to the operation habit classification label and the change trend of the game scene; S4, monitoring the use of the vibration function of the controller, the light flashing frequency, and the key operation frequency when switching game scenes, and determining the power consumption trend of the current scene; S5. If the power consumption trend of the current scene is lower than the preset threshold, a low power consumption mode is triggered to reduce the power supply of the vibration function and the light flashing; S6, judging whether to exit the low-power standby mode according to the time interval threshold of the standby mode and the environment information of the handle; S7: In low-power standby mode, monitor the operation signal of the handle and allocate power resources according to the current game scene and the player's operating habits.

2. A game controller power management method according to claim 1, characterized in that: The process of constructing the power consumption model in step S1 is specifically as follows: Obtaining vibration frequency, light flashing frequency, and key operation frequency data of the game controller in different game scenarios, and preprocessing the data; According to the preprocessed game controller power consumption data, an initial power consumption model is constructed, the vibration frequency, light flashing frequency and key operation frequency are used as independent variables, and the power consumption is used as the dependent variable, and the coefficients of the regression model are fitted by the least square method; Evaluate the initial power consumption model, and judge the model's fit and prediction ability by calculating the mean square error and determination coefficient. If the model performance does not meet the requirements, introduce a regularization term to control the model complexity, and try to use a support vector machine. By comparing the performance of different algorithms, select a modeling method.

3. A game controller power management method according to claim 1, characterized in that: The process of generating the operation habit classification label in step S2 is specifically as follows: According to the power consumption model, the operation record data of the players are obtained, and for each player's operation record data, the frequency of various operations is counted to obtain the operation frequency distribution; According to the operation frequency distribution, determining whether each type of operation is a high-frequency operation or a low-frequency operation, if the operation frequency is higher than a preset threshold, determining it as a high-frequency operation, otherwise determining it as a low-frequency operation; For the high-frequency operation and the low-frequency operation, extract operation features respectively to obtain a high-frequency operation feature set and a low-frequency operation feature set; Classifying the players' operation habits according to the high-frequency operation feature set and the low-frequency operation feature set to obtain an operation habit classification result; For each operation habit classification, the average power consumption of players in the classification is counted to obtain the power consumption level of each operation habit classification.

4. A game controller power management method according to claim 1, characterized in that: The process of dynamically adjusting the time interval threshold of the standby mode in step S3 is specifically as follows: According to the pre-established correspondence between the operation frequency and the time interval threshold, the real-time operation data of the player is obtained, and the current operation frequency of the player is determined through data analysis; Determine whether the operation frequency of the player is a high-frequency operation or a low-frequency operation, if it is a high-frequency operation, set the time interval of the standby mode to a longer interval, if it is a low-frequency operation, set the time interval of the standby mode to a shorter interval; A clustering algorithm is used to classify the operating habits of the players to obtain typical behavior patterns under different operating habits; For each typical behavior pattern, the association rule mining algorithm is used to analyze its correlation with the changes in the game scene, and the scene change trends under different behavior patterns are obtained; The time interval threshold of the standby mode is dynamically adjusted according to the scene change trend.

5. A game controller power management method according to claim 1, characterized in that: The process of determining the power consumption trend of the current scene in step S4 is specifically as follows: Get the trigger signal of game scene switching. If the scene switching signal is detected, start real-time monitoring of the handle vibration frequency, light flashing frequency and key operation frequency; According to the preset vibration frequency threshold, light flashing frequency threshold and key operation frequency threshold, it is determined whether the current handle vibration frequency, light flashing frequency and key operation frequency exceed the threshold; If at least one of the handle vibration frequency, light flashing frequency and key operation frequency exceeds a preset threshold, the current game scene is determined to be a high power consumption scene; Among them, for the game scenes judged to be high power consumption, a support vector machine algorithm is used to train a high power consumption scene prediction model based on the handle vibration frequency, light flashing frequency and button operation frequency data of historical high power consumption scenes.

6. A game controller power management method according to claim 1, characterized in that: The process of triggering the low power consumption mode in step S5 is specifically as follows: Obtain power consumption data in the current scenario, compare it with a preset threshold, and trigger a low power consumption mode if the power consumption is lower than the threshold; According to the preset low power consumption strategy, the reduction degree of the vibration function and the light flashing effect is determined, and the function degradation is achieved by controlling the power supply; Gradually reduce the vibration frequency and light brightness until the lowest power consumption level is reached, while monitoring the power consumption trend; The frequency reduction and dimming schemes are obtained by fitting the functional relationship between power consumption, vibration frequency and light brightness using the least squares method. Using a decision tree algorithm, taking into account current scene features and user usage habits, dynamically adjust the trigger threshold of the low power consumption mode; Build a Markov model to predict the power consumption trend in the future based on historical power consumption data and make energy-saving decisions.

7. A game controller power management method according to claim 1, characterized in that: The process of determining whether to exit the low-power standby mode in step S6 in combination with the environment information of the handle is specifically as follows: Get the current environment information of the controller, including temperature, humidity, light intensity, and noise data; According to the preset threshold range of each dimension of environmental data, determine whether the current environment of the handle meets the normal working conditions; If the current environment does not meet the normal working conditions of the handle, the low-power standby mode of the handle is activated; In the low-power standby mode, the environment data of the handle is collected periodically, and the environment state of the handle in the future is predicted according to the change trend of the environment data; Establishing a correlation model between environmental data and handle power consumption through a machine learning algorithm, and dynamically adjusting the standby time threshold of the low power consumption mode; When the handle continues to be in standby mode for longer than the standby time threshold and the environmental prediction result indicates that the handle will still be in an unavailable environment for a period of time in the future, the handle is controlled to exit the low-power standby mode and enter a dormant state.

8. A game controller power management method according to claim 1, characterized in that: The process of allocating power resources in step S7 is specifically as follows: Get the operation signal of the handle to determine whether the handle is in low-power standby mode. If so, enter the continuous monitoring state, otherwise maintain the normal working state; Under continuous monitoring, the convolutional neural network algorithm is used to extract the features of the handle operation signal to obtain the operation feature vector; According to the pre-established game scene model and player operation habit model, the operation feature vector is classified to determine the current game scene type and player operation habit type; According to different game scene types and player operation habits, the decision tree algorithm is used to generate corresponding power resource allocation strategies; Convert the power resource allocation strategy into control instructions and send them to the power management module of the handle through the communication interface of the handle; The power management module adjusts the operating voltage and clock frequency of each hardware module according to the received control instructions; When the handle ends the low-power standby mode, the default working parameters of each hardware module are restored; Obtain historical operation data and real-time power consumption data of the device to build a training data set for continuous training and optimization of machine learning algorithms; Through continuous training and iterative optimization of machine learning algorithms, the intelligence level of power management strategies is continuously improved.

9. A game controller power management method according to claim 1, characterized in that: Also includes: S8, dynamically adjusting the time interval threshold of the standby mode, and optimizing the power management strategy in combination with historical operation data and current power consumption trends; The process of optimizing the power management strategy in step S8 is specifically as follows: Obtain historical operation data and real-time power consumption data of the device to build a training data set; Adopting a decision tree algorithm, dynamically adjusting the time interval threshold of the standby mode according to the historical operation data and the real-time power consumption trend, to obtain an optimized threshold parameter; Through the support vector machine algorithm, historical data and real-time trends are comprehensively analyzed to determine the current device usage status and power consumption level, and determine whether to trigger the standby mode; If the idle time of the current device exceeds the optimized time interval threshold and the power consumption level is low, it automatically enters the standby mode to reduce power consumption; In the standby mode, the operation behavior and power consumption changes of the device are continuously monitored. If a user operation or a sudden increase in power consumption is detected, the device is awakened and the standby mode is exited.

10. A game controller power management system, characterized in that: It includes power management module, data acquisition and processing module and data analysis and model building module; The power management module controls the turning on and off of the power of the handle and the switching of different power consumption states according to the triggering conditions and exit conditions of the low power consumption mode; According to the preset low-power strategy, the power supply of the vibration function and light flashing is reduced, and the operating voltage and clock frequency of each hardware module are adjusted; Dynamically adjust the time interval threshold of standby mode to optimize power management strategy; The data collection and processing module collects power consumption-related data of the game controller in different game scenarios, as well as the player's operation record data and the environment information of the controller, and pre-processes the collected data; The processed data is transmitted to the power management module and the data analysis and model building module to provide data support for the formulation and optimization of power management strategies; The data analysis and model building module builds an initial power consumption model based on the preprocessed game controller power consumption data, and evaluates the model performance; Based on the power consumption model, analyze the impact of players' operating habits on power consumption and calculate the power consumption level of each category; Analyze the correlation between players’ operating habits and changes in game scenarios; Build a Markov model to assist the power management module in making energy-saving decisions; A correlation model between environmental data and handle power consumption is established to provide support for dynamically adjusting the standby time threshold of the low power mode.

Images