Traveling method of electric kick scooter and electric kick scooter

Abstract

The application relates to the technical field of artificial intelligence, and discloses a driving method of an electric scooter and the electric scooter.The method comprises the following steps: based on sensor data of the electric scooter, analyzing a user riding behavior mode, identifying a deceleration event, generating behavior mode data and deceleration event data, calculating a theoretical power output value and an energy recovery potential value, adjusting parameters of a motor controller of the electric scooter, then controlling an energy recovery process of the electric scooter, dynamically adjusting recovery intensity, generating recovery energy data, optimizing an analysis process of the riding behavior mode, obtaining optimized behavior mode data, finally outputting adjusted driving state and energy recovery information to the user, and providing a recovery efficiency report; and the application can improve the accuracy of driving process control of the electric scooter.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a method for driving an electric scooter and the electric scooter itself. Background Technology

[0002] Currently, the riding control and energy recovery systems of electric scooters mostly operate independently, lacking precise analysis and dynamic adaptation to user riding behavior patterns. Existing technologies typically rely solely on fixed parameters to perform power output adjustment and energy recovery operations, failing to link riding behavior pattern recognition with energy recovery strategies. This makes it impossible to adjust the recovery plan based on the user's riding habits and real-time deceleration behavior, resulting in low energy recovery efficiency and a significant waste of kinetic energy during deceleration. Furthermore, it is difficult to adapt to the riding needs of different users. In addition, the control logic of traditional systems suffers from unreasonable step-by-step transitions, often performing energy recovery before behavior analysis, causing the recovery strategy to lag behind changes in user behavior and further reducing the accuracy of energy utilization.

[0003] Traditional electric scooters use a static setting for energy recovery intensity, ignoring the impact of varying riding styles on recovery efficiency and riding experience. Using the same recovery intensity parameter for both aggressive and conservative riders not only wastes energy but can also lead to discomfort. For example, insufficient recovery intensity during aggressive riding may result in missed recovery opportunities, while excessive intensity during conservative riding can negatively impact ride smoothness. Furthermore, current technology lacks a real-time optimization mechanism for the recovery process. It cannot dynamically optimize riding behavior analysis based on motor power adjustment data and recovered energy data, leading to low behavior recognition accuracy. Consequently, power regulation and energy recovery control remain in a suboptimal state, severely limiting the electric scooter's range and user experience. Therefore, improving the real-time optimization mechanism is a pressing issue that needs to be addressed. Summary of the Invention

[0004] The present invention provides a method for driving an electric scooter and an electric scooter to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a method for driving an electric scooter, comprising: S1 analyzes user riding behavior patterns based on sensor data from the electric scooter, identifies deceleration events, and generates behavior pattern data and deceleration event data. S2, based on the behavior pattern data, the deceleration event data, and the external road condition data, calculate the theoretical power output value and energy recovery potential value of the electric scooter; S3, Based on the theoretical power output value, adjust the parameters of the motor controller of the electric scooter to dynamically adjust the motor power output and obtain the adjusted motor power data; S4. Based on the behavior pattern data, deceleration event data, and energy recovery potential value, control the energy recovery process of the electric scooter, dynamically adjust the recovery intensity, and generate recovered energy data. S5. Based on the adjusted motor power data and the recovered energy data, monitor the battery status, driving data and energy recovery status of the electric scooter in real time, and optimize the analysis process of the riding behavior pattern to obtain optimized behavior pattern data. S6. Based on the adjusted motor power data and recovered energy data, output the adjusted driving status and energy recovery information to the user and provide a recovery efficiency report.

[0006] In a preferred embodiment, the analysis of user riding behavior patterns based on sensor data from the electric scooter, and the identification of deceleration events to generate behavior pattern data and deceleration event data, includes: Extract speed data, acceleration data, steering angle data, and brake signal data from the sensor data of the electric scooter; Based on the speed data, acceleration data, steering angle data, and braking signal data, time-domain features are extracted, and statistical indicators such as mean, variance, peak value, and rate of change are calculated to generate a cycling feature vector that quantifies the user's cycling habits. The cycling feature vectors are mapped to predefined behavior pattern categories to obtain behavior pattern data, wherein the behavior pattern categories include aggressive, moderate, and conservative. Based on the velocity data and the acceleration data, the instantaneous deceleration is calculated and compared with a preset deceleration event threshold to identify deceleration event data.

[0007] In a preferred embodiment, calculating the theoretical power output and energy recovery potential of the electric scooter based on the behavioral pattern data, the deceleration event data, and the external road condition data includes: The power demand factor of the electric scooter is analyzed based on the behavioral pattern data and external road condition data. The power demand factor is multiplied by the preset reference power value to obtain the theoretical power output value, wherein the preset reference power value is the rated power value of the electric scooter under standard conditions. The deceleration intensity in the deceleration event data is extracted as the main input variable, and the road surface type coefficient and traffic density coefficient in the external road condition data are used as auxiliary variables to perform recovery potential analysis to obtain the basic recovery value. The behavior pattern data is mapped to a preset recycling efficiency coefficient to obtain the correspondence between the behavior pattern data and the recycling efficiency coefficient. The basic recycling value is multiplied by the recycling efficiency coefficient corresponding to the behavior pattern data to obtain the energy recycling potential value.

[0008] In a preferred embodiment, the step of analyzing the power demand factor of the electric scooter based on the behavioral pattern data and external road condition data includes: First, the behavioral pattern data is mapped to numerical behavioral scores, where the aggressive mode corresponds to a score of 1.2, the stable mode corresponds to a score of 1.0, and the conservative mode corresponds to a score of 0.8. The external road condition data is then converted into a road condition score through normalization processing. The external road condition data includes slope value, road surface type coefficient and traffic density coefficient. Finally, the behavior score and road condition score are linearly combined according to preset weights to generate the power demand factor, where the preset weight for the behavior score is 0.6 and the weight coefficient for the road condition score is 0.4.

[0009] In a preferred embodiment, adjusting the parameters of the electric scooter's motor controller based on the theoretical power output value to dynamically adjust the motor power output and obtain adjusted motor power data includes: The power deviation value is obtained by subtracting the current motor power data from the theoretical power output value. Based on the power deviation value, a proportional-integral-derivative (PI-DE) control calculation is performed to obtain the controller adjustment parameters. The mathematical expression for the PI-DE control calculation is as follows: In the formula, Adjust the parameters for the controller. It is a function of the power deviation value changing with time. This is the proportionality coefficient. The integral coefficient is... These are the differential coefficients. It is a time variable; Based on the controller adjustment parameters, the parameter settings of the motor controller are updated to obtain the adjusted motor power data.

[0010] In a preferred embodiment, controlling the energy recovery process of the electric scooter based on the behavioral pattern data, deceleration event data, and energy recovery potential value, dynamically adjusting the recovery intensity, and generating recovered energy data includes: Based on the behavioral pattern data, the deceleration event data, and the energy recovery potential value, the behavioral pattern data is mapped to a behavioral influence coefficient through a recovery intensity calculation function. Then, the deceleration intensity in the deceleration event data is extracted as the main adjustment factor, and the energy recovery potential value is used as the benchmark value. Finally, the recovery intensity parameter is generated by weighted combination calculation. Based on the aforementioned recovery intensity parameter, the energy recovery process is dynamically adjusted using a recovery control algorithm to obtain adjusted recovery process data. Based on the adjusted recovery process data, an integral operation is performed using an energy calculation function to generate recovered energy data.

[0011] In a preferred embodiment, the process of real-time monitoring of the electric scooter's battery status, riding data, and energy recovery status based on the adjusted motor power data and the recovered energy data, and optimizing the riding behavior pattern analysis process, yields optimized behavior pattern data, including: Based on the adjusted motor power data and the recovered energy data, a comprehensive status index is calculated using a data fusion algorithm. Based on the comprehensive state index, the behavior pattern analysis parameters are adjusted through an optimization function to obtain optimized analysis parameters; Based on the optimized analysis parameters, the cycling behavior pattern is re-analyzed to obtain optimized behavior pattern data.

[0012] In a preferred embodiment, the mathematical expression of the optimization function is as follows: In the formula, To optimize the analysis parameters, Parameters for analyzing the current behavioral pattern. This is the learning rate coefficient. This represents the gradient of the loss function with respect to the parameters.

[0013] In a preferred embodiment, based on the adjusted motor power data and recovered energy data, the adjusted driving status and energy recovery information are output to the user, providing a recovery efficiency report, including: Based on the adjusted motor power data and recovered energy data, the adjusted motor power data is divided by the preset rated power value through the state efficiency calculation function to generate a power normalization value as a driving state index. Then, the recovered energy data is divided by the total energy consumption estimate to generate an energy normalization value as a recovery efficiency index. Based on the driving status index and the energy recovery efficiency index, the driving status index is classified into low, medium and high power states according to the numerical range of the driving status index by the information generation algorithm, and corresponding descriptive text is generated to obtain user-readable driving status information and energy recovery information. Based on the user-readable driving status information and energy recovery information, a recovery efficiency report is generated.

[0014] To address the above problems, the present invention also provides an electric scooter, the electric scooter comprising: The cycling behavior pattern analysis module is used to analyze user cycling behavior patterns based on sensor data from electric scooters, identify deceleration events, and generate behavior pattern data and deceleration event data. The road condition power demand prediction module is used to calculate the theoretical power output value and energy recovery potential value of the electric scooter based on the behavior pattern data, the deceleration event data and the external road condition data. An adaptive power adjustment module is used to adjust the parameters of the motor controller of the electric scooter based on the theoretical power output value, so as to dynamically adjust the motor power output and obtain the adjusted motor power data. The energy recovery execution module is used to control the energy recovery process of the electric scooter based on the behavior pattern data, deceleration event data and energy recovery potential value, dynamically adjust the recovery intensity, and generate recovered energy data; The real-time optimization monitoring module is used to monitor the battery status, driving data and energy recovery status of the electric scooter in real time based on the adjusted motor power data and the recovered energy data, and optimize the analysis process of the riding behavior pattern to obtain optimized behavior pattern data. The user interaction and output module is used to output the adjusted driving status and energy recovery information to the user based on the adjusted motor power data and recovered energy data, and to provide a recovery efficiency report.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention significantly improves the energy recovery efficiency and energy management accuracy of electric scooters by pioneering a method that dynamically combines riding behavior pattern recognition with energy recovery strategies. The method first analyzes the user's riding behavior patterns (including deceleration events) and then adjusts the sequence of energy recovery control, breaking the limitations of traditional technologies where these two processes operate independently. This avoids the lag in recovery caused by relying solely on power parameters or road condition data. Specifically, it non-obviously combines behavioral pattern data (such as aggressive, steady, and conservative riding classifications) with energy recovery potential values, enabling the system to accurately match user riding habits and recovery needs. For example, when a steady riding mode is identified, the basic recovery value can be optimized based on its corresponding recovery efficiency coefficient, forming a collaborative energy management mechanism of "behavior analysis - potential assessment - precise recovery." This not only maximizes the capture of kinetic energy during deceleration but also reduces ineffective energy consumption, effectively extending the electric scooter's range.

[0016] 2. This invention introduces adaptive energy recovery intensity adjustment based on behavioral patterns, completely solving the drawbacks of traditional static recovery settings and balancing energy recovery effectiveness with user riding experience. For the first time, the system uses the user's riding style (e.g., aggressive or steady) as the core basis for recovery control. By mapping the relationship between behavioral patterns and recovery intensity parameters, it achieves dynamic adaptation of recovery intensity: for aggressive riders, a lower recovery intensity coefficient is matched to avoid excessive recovery affecting riding control; for conservative riders, the recovery intensity coefficient is increased to enhance energy capture and reduce energy waste. This personalized recovery control method avoids the problems of "aggressive riders missing recovery opportunities and conservative riders experiencing discomfort due to excessive recovery" under traditional fixed recovery intensity. Furthermore, through real-time optimized behavioral pattern analysis (combined with adjusted motor power and recovered energy data), it continuously calibrates the recovery strategy, ensuring efficient operation that aligns with user habits throughout, significantly improving product comfort and energy utilization efficiency. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating a method for driving an electric scooter according to an embodiment of the present invention. Figure 2 This is a functional block diagram of an electric scooter provided in an embodiment of the present invention; The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0019] This application provides a method for driving an electric scooter. The entity executing this method includes, but is not limited to, at least one of the following electronic devices: a server, a terminal, or any other electronic device configured to execute the method provided in this application. In other words, the method for driving the electric scooter can be executed by software or hardware installed on a terminal device or a server. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, or a Content Delivery Network (CDN). Cloud servers that provide basic cloud computing services such as big data and artificial intelligence platforms.

[0020] Reference Figure 1 The diagram shown is a flowchart illustrating a method for driving an electric scooter according to an embodiment of the present invention. In this embodiment, the method for driving the electric scooter includes: S1 analyzes user riding behavior patterns based on sensor data from the electric scooter, identifies deceleration events, and generates behavior pattern data and deceleration event data. In this embodiment of the invention, the step of analyzing user riding behavior patterns based on sensor data from the electric scooter, identifying deceleration events, and generating behavior pattern data and deceleration event data includes: Extract speed data, acceleration data, steering angle data, and brake signal data from the sensor data of the electric scooter; Based on the speed data, acceleration data, steering angle data, and braking signal data, time-domain features are extracted, and statistical indicators such as mean, variance, peak value, and rate of change are calculated to generate a cycling feature vector that quantifies the user's cycling habits. The cycling feature vectors are mapped to predefined behavior pattern categories to obtain behavior pattern data, wherein the behavior pattern categories include aggressive, moderate, and conservative. Based on the velocity data and the acceleration data, the instantaneous deceleration is calculated and compared with a preset deceleration event threshold to identify deceleration event data.

[0021] It should be noted that the cycling feature vector is a set of multi-dimensional numerical features, including average speed, standard deviation of acceleration, rate of change of steering angle, and brake signal frequency. It is used to comprehensively characterize the user's cycling dynamics and to compress continuous sensor data into a fixed-dimensional feature space, which is convenient for subsequent pattern classification algorithms.

[0022] It should be noted that the specific mapping process of the predefined behavior pattern categories is as follows: First, the cycling feature vector (including average speed, acceleration standard deviation, steering angle change rate and brake signal frequency) is used as input features and input into the pre-trained support vector machine classification model; Subsequently, the classification model calculates the decision function value between the cycling feature vector and each behavior pattern category (aggressive, steady, conservative), where the decision function is constructed based on the radial basis kernel function; Finally, the category with the largest decision function value is selected as the behavioral pattern data output.

[0023] It should be noted that the behavioral pattern data are category labels output by the classification operation, representing the current user's riding style and quantifying the user's behavioral tendencies. This data is used for the dynamic adjustment of subsequent adaptive power regulation and energy recovery strategies.

[0024] It should be noted that the deceleration event identification process first monitors the acceleration value in real time. When the acceleration is lower than the negative deceleration event threshold and the speed continues to decrease, it is determined to be a deceleration event. The acceleration threshold is calibrated based on typical cycling data and is set to 0.5 m / s².

[0025] It should be noted that the deceleration event data includes the event timestamp, deceleration intensity, and duration, recording deceleration segments during the user's ride of the electric scooter, which is used to assess energy recovery potential and optimize recovery intensity control.

[0026] S2, based on the behavior pattern data, the deceleration event data, and the external road condition data, calculate the theoretical power output value and energy recovery potential value of the electric scooter; In this embodiment of the invention, calculating the theoretical power output value and energy recovery potential value of the electric scooter based on the behavior pattern data, the deceleration event data, and the external road condition data includes: The power demand factor of the electric scooter is analyzed based on the behavioral pattern data and external road condition data. The power demand factor is multiplied by the preset reference power value to obtain the theoretical power output value, wherein the preset reference power value is the rated power value of the electric scooter under standard conditions. The deceleration intensity in the deceleration event data is extracted as the main input variable, and the road surface type coefficient and traffic density coefficient in the external road condition data are used as auxiliary variables to perform recovery potential analysis to obtain the basic recovery value. The behavior pattern data is mapped to a preset recycling efficiency coefficient to obtain the correspondence between the behavior pattern data and the recycling efficiency coefficient. The basic recycling value is multiplied by the recycling efficiency coefficient corresponding to the behavior pattern data to obtain the energy recycling potential value.

[0027] It should be noted that the external road condition data includes slope, road surface type, and traffic density.

[0028] It should be noted that the power demand factor is multiplied by the preset reference power value through multiplication, and the theoretical power output value is directly output, ensuring that the dimensions are consistent and the unit is watt.

[0029] It should be noted that the theoretical power output value is an estimate of the motor output power required by the electric scooter under the current user behavior and road conditions. It reflects the dynamic power demand and is used to optimize motor controller parameters and energy management strategies.

[0030] It should be noted that the recovery potential analysis is based on deceleration event data and external road condition data. A linear regression calculation is performed to obtain the basic recovery value. The deceleration event data includes deceleration intensity and event duration, while the external road condition data includes road surface type and traffic density. The specific process of the linear regression calculation is as follows: First, the deceleration intensity in the deceleration event data is extracted as the main input variable, and the road surface type coefficient and traffic density coefficient in the external road condition data are used as auxiliary variables. Subsequently, the input variables and auxiliary variables are weighted and summed using preset regression coefficients to generate the baseline recovery value. The linear regression calculation expression used for the recovery potential analysis is given below: In the formula, Based on the recovery value, For deceleration strength, For road surface type coefficient, Traffic density coefficient, For the intercept term, The deceleration intensity regression coefficient is... For road surface type regression coefficients, This is the traffic density regression coefficient.

[0031] Furthermore, the preset regression coefficients are obtained through linear regression analysis of historical cycling data, including deceleration intensity, road surface type, traffic density, and actual recovered energy value. The regression coefficient values ​​are determined based on least squares fitting, with typical values ​​being [value missing]. , , , These coefficients are used to ensure that the prediction error is minimized through cross-validation.

[0032] It should be noted that the base recovery value is a numerical value representing the potential recoverable energy, used to assess the base amount of energy that can be recovered during a deceleration event, taking into account the impact of road conditions on recovery efficiency.

[0033] It should be noted that the behavioral pattern data is used to determine the recycling efficiency coefficient. The mapping relationship between the behavioral pattern data and the preset recycling efficiency coefficient is as follows: the recycling efficiency coefficient corresponding to the aggressive mode is 0.9, the recycling efficiency coefficient corresponding to the steady mode is 1.0, and the recycling efficiency coefficient corresponding to the conservative mode is 1.1. The recycling efficiency coefficient is used to adjust the recycling potential based on the user's riding style. The basic recycling value is multiplied by the recycling efficiency coefficient corresponding to the behavioral mode through a multiplication operation to obtain the capability recycling potential value.

[0034] It should be noted that the energy recovery potential value is the energy recovery power or energy estimate that an electric scooter can achieve after taking into account user behavior and road conditions. It is used to dynamically control the energy recovery process and optimize the recovery intensity, and its value reflects the actual recovery possibility.

[0035] In this embodiment of the invention, the step of analyzing the power demand factor of the electric scooter based on the behavioral pattern data and external road condition data includes: First, the behavioral pattern data is mapped to numerical behavioral scores, where the aggressive mode corresponds to a score of 1.2, the stable mode corresponds to a score of 1.0, and the conservative mode corresponds to a score of 0.8. The external road condition data is then converted into a road condition score through normalization processing. The external road condition data includes slope value, road surface type coefficient and traffic density coefficient. Finally, the behavior score and road condition score are linearly combined according to preset weights to generate the power demand factor, where the preset weight for the behavior score is 0.6 and the weight coefficient for the road condition score is 0.4.

[0036] It should be noted that converting external road condition data into a road condition score through normalization involves first performing minimum-maximum normalization on the slope value, which maps the actual slope value to a value between zero and one. The minimum slope value is set to -10 degrees, and the maximum slope value is set to +10 degrees. The normalized slope value is obtained by calculating the difference between the actual slope value and the minimum slope value, and then dividing by the difference between the maximum slope value and the minimum slope value. Secondly, predefined discrete values ​​are used directly for the road surface type coefficient. For example, a coefficient of 1.0 corresponds to a smooth road surface and a coefficient of 0.5 corresponds to a rough road surface, without the need for additional conversion. Then, the traffic density coefficient is subjected to minimum-maximum normalization, which maps the actual traffic density value to a value between zero and one. The minimum traffic density value is set to zero to indicate no traffic, and the maximum traffic density value is set to one hundred to indicate high-density traffic. The normalized traffic density value is obtained by calculating the difference between the actual traffic density value and the minimum traffic density value and then dividing it by the difference between the maximum traffic density value and the minimum traffic density value. Finally, the normalized slope value, road surface type coefficient, and normalized traffic density value are weighted and summed according to preset weights to generate a road condition score. The slope value has a weight of 0.5, the road surface type coefficient has a weight of 0.3, and the traffic density value has a weight of 0.2. The total weight is 1, ensuring that the road condition score comprehensively reflects the impact of external road conditions.

[0037] It should be noted that the power demand factor is a dimensionless scaling factor used to quantify the combined impact of user riding behavior and external road conditions on the power demand of electric scooters. A value greater than 0 indicates the adjustment range relative to the baseline power, which is used for subsequent calculations of theoretical power output values.

[0038] S3, Based on the theoretical power output value, adjust the parameters of the motor controller of the electric scooter to dynamically adjust the motor power output and obtain the adjusted motor power data; In this embodiment of the invention, adjusting the parameters of the motor controller of the electric scooter based on the theoretical power output value to dynamically adjust the motor power output and obtain adjusted motor power data includes: The power deviation value is obtained by subtracting the current motor power data from the theoretical power output value. Based on the power deviation value, a proportional-integral-derivative (PI-DE) control calculation is performed to obtain the controller adjustment parameters. The mathematical expression for the PI-DE control calculation is as follows: In the formula, Adjust the parameters for the controller. It is a function of the power deviation value changing with time. This is the proportionality coefficient. The integral coefficient is... These are the differential coefficients. It is a time variable; Based on the controller adjustment parameters, the parameter settings of the motor controller are updated to obtain the adjusted motor power data.

[0039] It should be noted that the theoretical power output value is an estimate of the power required by the electric scooter under current conditions, while the current motor power data is the measured value of the actual power output of the motor.

[0040] It should be noted that the power deviation value is the difference between the theoretical power demand and the actual power output. It represents the magnitude and direction of motor power adjustment. A positive value indicates that the power output needs to be increased, and a negative value indicates that the power output needs to be reduced. It is used for subsequent controller parameter adjustments.

[0041] It should be noted that the essence of PID control calculation is to dynamically adjust the controller parameters by calculating the proportional, integral, and derivative components of the power deviation in real time, so as to minimize the power deviation and improve the system response speed. The proportional term responds quickly based on the current deviation, the integral term eliminates steady-state error, and the derivative term predicts the trend of deviation change.

[0042] It should be noted that the controller adjustment parameters are dynamic adjustment values ​​of the motor controller parameters, used to modify the controller's gain or settings to achieve precise adjustment of the motor's power output. Their values ​​reflect the compensation requirements for power deviation.

[0043] It should be noted that updating the motor controller's parameter settings is based on adjusting the controller parameters and performing a parameter superposition operation to obtain the adjusted motor power data. The parameter superposition operation adds the adjusted controller parameters to the current parameter values ​​of the motor controller to generate the updated parameters. The specific process of the parameter superposition operation is as follows: First, read the current parameter values ​​of the motor controller, such as proportional gain, integral time, or derivative time; then, superimpose the adjusted controller parameters onto the current parameter values ​​according to the corresponding items; finally, write the updated parameters into the motor controller to drive the motor to output the adjusted power. It should be noted that the adjusted motor power data is the actual power output value of the motor after updating the controller parameters, reflecting the dynamically adjusted power state. This data is used for subsequent energy recovery monitoring and user interaction to ensure that the electric scooter's power output matches the theoretical requirements.

[0044] Furthermore, the proportional coefficient is set to 0.8 and the integral coefficient to 0.2s. -1 The value of the differential coefficient is 0.1s.

[0045] S4. Based on the behavior pattern data, deceleration event data, and energy recovery potential value, control the energy recovery process of the electric scooter, dynamically adjust the recovery intensity, and generate recovered energy data. In this embodiment of the invention, controlling the energy recovery process of the electric scooter based on the behavioral pattern data, deceleration event data, and energy recovery potential value, dynamically adjusting the recovery intensity, and generating recovered energy data includes: Based on the behavioral pattern data, the deceleration event data, and the energy recovery potential value, the behavioral pattern data is mapped to a behavioral influence coefficient through a recovery intensity calculation function. Then, the deceleration intensity in the deceleration event data is extracted as the main adjustment factor, and the energy recovery potential value is used as the benchmark value. Finally, the recovery intensity parameter is generated by weighted combination calculation. Based on the aforementioned recovery intensity parameter, the energy recovery process is dynamically adjusted using a recovery control algorithm to obtain adjusted recovery process data. Based on the adjusted recovery process data, an integral operation is performed using an energy calculation function to generate recovered energy data.

[0046] It should be noted that the recovery intensity calculation function is based on behavioral pattern data, deceleration event data, and energy recovery potential value. It performs a multi-factor fusion calculation operation to obtain the recovery intensity parameter. The behavioral pattern data includes user riding style classification (such as aggressive, smooth, or conservative), the deceleration event data includes deceleration intensity and event duration, and the energy recovery potential value is the estimated value of energy recovery that the electric scooter can achieve.

[0047] Furthermore, the specific process of multi-factor fusion calculation is as follows: First, the behavioral pattern data is mapped to behavioral influence coefficients, for example, the coefficient for the aggressive mode is 0.8, the coefficient for the stable mode is 1.0, and the coefficient for the conservative mode is 1.2. Subsequently, the deceleration intensity was extracted from the deceleration event data as the main adjustment factor, and the energy recovery potential value was used as the benchmark value. Finally, the recovery intensity parameter was generated by weighted combination calculation, where the weights were evenly distributed.

[0048] It should be noted that the recovery intensity parameter is a dimensionless adjustment coefficient used to quantify the intensity level of the energy recovery process. A value greater than 0 indicates the adjustment range of the recovery intensity, which dynamically optimizes the recovery efficiency based on user riding behavior and deceleration events.

[0049] It should be noted that the recovery control algorithm is based on the recovery intensity parameter and performs a proportional adjustment operation to dynamically adjust the energy recovery process and obtain the adjusted recovery process data. The proportional adjustment operation is to multiply the recovery intensity parameter with the current recovery process parameter to generate the updated recovery process parameter.

[0050] Furthermore, the mathematical expression for the proportional adjustment operation is as follows: In the formula, For the adjusted recycling process data, These are the parameters for the current recycling process. For recovery strength parameters; Furthermore, the essence of the proportional adjustment operation is to smoothly adjust the energy recovery intensity by linearly scaling the current recovery process parameters, ensuring that the recovery process matches the user's behavior and road conditions, and avoiding drastic changes that could cause riding discomfort.

[0051] It should be noted that the adjusted recovery process data are the operating parameters of the energy recovery system after dynamic adjustment, reflecting the real-time status of the recovery intensity, and are used for subsequent energy calculation and monitoring to ensure that the recovery process is efficient and stable.

[0052] It should be noted that the energy calculation function performs an integral operation based on the adjusted recovery process data to generate recovered energy data. The integral operation involves accumulating the adjusted recovery process data over time. Furthermore, the mathematical expression for the integral operation is as follows: In the formula, To recover energy data, For the adjusted recycling process data, It is a time variable.

[0053] Furthermore, the essence of integral operation is to calculate the total recovered energy by accumulating the power or intensity parameters in the time domain, ensuring dimensional consistency (unit: joule) and reflecting the actual recovery effect.

[0054] It should be noted that the recovered energy data is the total amount of energy actually recovered by the electric scooter during the energy recovery process. It is used to evaluate recovery efficiency, optimize riding behavior analysis, and provide user feedback. Its value quantifies the actual benefits of energy recovery.

[0055] S5. Based on the adjusted motor power data and the recovered energy data, monitor the battery status, driving data and energy recovery status of the electric scooter in real time, and optimize the analysis process of the riding behavior pattern to obtain optimized behavior pattern data. In this embodiment of the invention, the process of real-time monitoring of the battery status, driving data, and energy recovery status of the electric scooter based on the adjusted motor power data and the recovered energy data, and optimizing the riding behavior pattern analysis process to obtain optimized behavior pattern data, includes: Based on the adjusted motor power data and the recovered energy data, a comprehensive status index is calculated using a data fusion algorithm. Based on the comprehensive state index, the behavior pattern analysis parameters are adjusted through an optimization function to obtain optimized analysis parameters; Based on the optimized analysis parameters, the cycling behavior pattern is re-analyzed to obtain optimized behavior pattern data.

[0056] It should be noted that the data fusion algorithm performs a weighted average calculation based on the adjusted motor power data and the recovered energy data to obtain a comprehensive status index. The adjusted motor power data is the actual output power value of the motor, and the recovered energy data is the total energy actually recovered by the electric scooter during the energy recovery process.

[0057] Furthermore, the specific process of the weighted average calculation operation is as follows: First, the adjusted motor power data is normalized into a power index, and the recovered energy data is normalized into an energy index. Then, the power index and energy index are linearly weighted and combined according to the preset weight coefficients to generate a comprehensive status index.

[0058] It should be noted that the comprehensive status index is a dimensionless value used to quantify the overall operating status of the electric scooter, including the combined impact of battery status, driving data, and energy recovery status. Its value reflects the system efficiency and stability and is used for subsequent behavior pattern analysis and optimization.

[0059] It should be noted that the optimization function is based on the comprehensive state index and performs parameter adjustment calculations to obtain the optimization analysis parameters. The parameter adjustment calculations are achieved by minimizing the deviation between the comprehensive state index and the target state.

[0060] It should be noted that the optimized analysis parameters are the adjusted settings in the behavior pattern analysis process. They are used to modify the sensitivity or weight of the analysis algorithm to improve the accuracy and adaptability of behavior pattern recognition. Their values ​​are dynamically optimized based on real-time monitoring data.

[0061] It should be noted that the reanalysis of cycling behavior patterns is based on optimizing the analysis parameters and performing a pattern classification operation to obtain optimized behavior pattern data. The pattern classification operation maps cycling feature vectors to predefined behavior pattern categories.

[0062] Furthermore, the specific process of the pattern classification operation is as follows: First, extract the cycling feature vector from the sensor data, including average speed, acceleration standard deviation, etc. Then, use optimization analysis parameters to adjust the decision boundary of the classification algorithm, recalculate the matching degree between the feature vector and the behavior pattern category, and finally output the updated behavior pattern category label.

[0063] It should be noted that the optimized behavioral pattern data is the user riding style classification result generated after reanalysis. It reflects the optimization effect based on real-time monitoring data and is used for further improvement of subsequent power regulation and energy recovery strategies. Its value ensures the dynamic adaptability of behavioral pattern analysis.

[0064] In this embodiment of the invention, the mathematical expression of the optimization function is as follows: In the formula, To optimize the analysis parameters, Parameters for analyzing the current behavioral pattern. This is the learning rate coefficient. This represents the gradient of the loss function with respect to the parameters.

[0065] It should be noted that the essence of parameter adjustment calculation is to iteratively adjust the behavioral pattern analysis parameters through the gradient descent method in order to minimize the difference between the comprehensive state index and the expected target, and ensure that the analysis process adapts to the real-time running state. The gradient descent process first calculates the loss gradient under the current parameters, and then updates the parameters according to the learning rate.

[0066] Furthermore, the learning rate coefficient is set to 0.01.

[0067] S6. Based on the adjusted motor power data and recovered energy data, output the adjusted driving status and energy recovery information to the user and provide a recovery efficiency report.

[0068] In this embodiment of the invention, the step of outputting adjusted driving status and energy recovery information to the user based on the adjusted motor power data and recovered energy data, and providing a recovery efficiency report, includes: Based on the adjusted motor power data and recovered energy data, the adjusted motor power data is divided by the preset rated power value through the state efficiency calculation function to generate a power normalization value as a driving state index. Then, the recovered energy data is divided by the total energy consumption estimate to generate an energy normalization value as a recovery efficiency index. Based on the driving status index and the energy recovery efficiency index, the driving status index is classified into low, medium and high power states according to the numerical range of the driving status index by the information generation algorithm, and corresponding descriptive text is generated to obtain user-readable driving status information and energy recovery information. Based on the user-readable driving status information and energy recovery information, a recovery efficiency report is generated.

[0069] It should be noted that the state efficiency calculation function is based on the adjusted motor power data and recovered energy data, and performs a normalization calculation operation to obtain the driving state index and recovery efficiency index. The adjusted motor power data is the actual output power value of the motor, and the recovered energy data is the total energy actually recovered by the electric scooter during the energy recovery process.

[0070] Furthermore, the specific process of the normalization calculation operation is as follows: First, the adjusted motor power data is divided by the preset rated power value to generate a power normalization value as a driving status indicator. Then, the recovered energy data is divided by the total energy consumption estimate to generate an energy normalization value as a recovery efficiency indicator. The total energy consumption estimate is calculated based on historical driving data.

[0071] It should be noted that the driving status index is a dimensionless value that represents the ratio of the current motor power output to the rated power. It is used to quantify the efficiency of the driving status. A value greater than 0 and less than 1 indicates an energy-saving state, while a value greater than 1 indicates a high-power state. This is used by users to understand the current driving performance.

[0072] It should be noted that the energy recovery efficiency index is a dimensionless value that represents the ratio of energy recovered to total energy consumed. It is used to quantify the efficiency of the recovery process, and its value is between 0 and 1. The higher the value, the better the recovery efficiency. It is used to evaluate the energy recovery effect.

[0073] It should be noted that the information generation algorithm is based on driving status indicators and energy recovery efficiency indicators. It performs text conversion operations to generate user-readable driving status information and energy recovery information. The text conversion operation maps numerical indicators into descriptive language. The specific process of the text conversion operation is as follows: First, based on the numerical range of the driving status indicators, they are classified into low, medium, and high power states, and corresponding descriptive text is generated; then, based on the numerical range of the energy recovery efficiency indicators, they are classified into poor, medium, and excellent recovery levels, and corresponding descriptive text is generated; finally, these texts are combined to generate user-readable information.

[0074] It should be noted that the user-readable riding status information is text information describing the current power output status of the electric scooter, such as "Current power output: Medium", which is used to help users intuitively understand the riding status. Its physical meaning is a common representation of power usage.

[0075] It should be noted that the user-readable energy recovery information is text information describing the energy recovery effect, such as "recovery efficiency: excellent", which is used to help users evaluate energy recovery performance and provides feedback to optimize cycling behavior.

[0076] It should be noted that the report is generated based on user-readable driving status information and energy recovery information, and a formatting operation is performed to generate a recovery efficiency report. The formatting operation organizes the text information into a structured document.

[0077] Furthermore, the mathematical expression for the formatting operation is as follows: In the formula, For the recycling efficiency report, Provides user-readable driving status information. Provides user-readable energy recovery information. and , where is the weighting coefficient, which is used to adjust the proportion of information in the report, and is always one-half.

[0078] Furthermore, the essence of the formatting operation is to generate a comprehensive report by linearly combining user-readable driving status information and energy recovery information, ensuring that the report content is balanced and easy to understand. The linear combination process is as follows: first, the driving status information is multiplied by a weighting coefficient α, then the energy recovery information is multiplied by a weighting coefficient β, and finally the two products are added together to obtain the recovery efficiency report.

[0079] It should be noted that the energy recovery efficiency report is a structured document that includes riding status and energy recovery information. It is used to provide users with comprehensive performance feedback. Its physical meaning is a summary of the efficiency of the riding process, and its purpose is to help users optimize riding habits and improve energy utilization efficiency.

[0080] like Figure 2 The diagram shown is a functional block diagram of an electric scooter provided in an embodiment of the present invention.

[0081] The electric scooter 100 described in this invention can be installed in an electronic device. Depending on the functions implemented, the electric scooter 100 may include a riding behavior pattern analysis module 101, a road condition power demand prediction module 102, an adaptive power adjustment module 103, an energy recovery execution module 104, a real-time optimization monitoring module 105, and a user interaction and output module 106. The module described in this invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and are stored in the memory of the electronic device.

[0082] In this embodiment, the functions of each module / unit are as follows: The cycling behavior pattern analysis module is used to analyze the user's cycling behavior pattern based on the sensor data of the electric scooter, identify deceleration events, and generate behavior pattern data and deceleration event data. The road condition power demand prediction module is used to calculate the theoretical power output value and energy recovery potential value of the electric scooter based on the behavior pattern data, the deceleration event data and the external road condition data. The adaptive power adjustment module is used to adjust the parameters of the motor controller of the electric scooter based on the theoretical power output value, so as to dynamically adjust the motor power output and obtain the adjusted motor power data. The energy recovery execution module is used to control the energy recovery process of the electric scooter based on the behavior pattern data, deceleration event data and energy recovery potential value, dynamically adjust the recovery intensity, and generate recovered energy data. The real-time optimization monitoring module is used to monitor the battery status, driving data and energy recovery status of the electric scooter in real time based on the adjusted motor power data and the recovered energy data, and optimize the analysis process of the riding behavior pattern to obtain optimized behavior pattern data. The user interaction and output module is used to output the adjusted driving status and energy recovery information to the user based on the adjusted motor power data and recovered energy data, and to provide a recovery efficiency report.

[0083] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0084] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0085] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0086] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0087] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0088] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for driving an electric scooter, characterized in that, The method includes: S1 analyzes user riding behavior patterns based on sensor data from the electric scooter, identifies deceleration events, and generates behavior pattern data and deceleration event data. S2, based on the behavior pattern data, the deceleration event data, and the external road condition data, calculate the theoretical power output value and energy recovery potential value of the electric scooter; S3, Based on the theoretical power output value, adjust the parameters of the motor controller of the electric scooter to dynamically adjust the motor power output and obtain the adjusted motor power data; S4. Based on the behavior pattern data, deceleration event data, and energy recovery potential value, control the energy recovery process of the electric scooter, dynamically adjust the recovery intensity, and generate recovered energy data. S5. Based on the adjusted motor power data and the recovered energy data, monitor the battery status, driving data and energy recovery status of the electric scooter in real time, and optimize the analysis process of the riding behavior pattern to obtain optimized behavior pattern data. S6. Based on the adjusted motor power data and recovered energy data, output the adjusted driving status and energy recovery information to the user and provide a recovery efficiency report.

2. The method for driving an electric scooter as described in claim 1, characterized in that, The sensor data based on the electric scooter is used to analyze the user's riding behavior patterns and identify deceleration events, generating behavior pattern data and deceleration event data, including: Extract speed data, acceleration data, steering angle data, and brake signal data from the sensor data of the electric scooter; Based on the speed data, acceleration data, steering angle data, and braking signal data, time-domain features are extracted, and statistical indicators such as mean, variance, peak value, and rate of change are calculated to generate a cycling feature vector that quantifies the user's cycling habits. The cycling feature vectors are mapped to predefined behavior pattern categories to obtain behavior pattern data, wherein the behavior pattern categories include aggressive, moderate, and conservative. Based on the velocity data and the acceleration data, the instantaneous deceleration is calculated and compared with a preset deceleration event threshold to identify deceleration event data.

3. The method for driving an electric scooter as described in claim 1, characterized in that, The calculation of the theoretical power output and energy recovery potential of the electric scooter based on the behavioral pattern data, the deceleration event data, and the external road condition data includes: The power demand factor of the electric scooter is analyzed based on the behavioral pattern data and external road condition data. The power demand factor is multiplied by the preset reference power value to obtain the theoretical power output value, wherein the preset reference power value is the rated power value of the electric scooter under standard conditions. The deceleration intensity in the deceleration event data is extracted as the main input variable, and the road surface type coefficient and traffic density coefficient in the external road condition data are used as auxiliary variables to perform recovery potential analysis to obtain the basic recovery value. The behavior pattern data is mapped to a preset recycling efficiency coefficient to obtain the correspondence between the behavior pattern data and the recycling efficiency coefficient. The basic recycling value is multiplied by the recycling efficiency coefficient corresponding to the behavior pattern data to obtain the energy recycling potential value.

4. The method for driving an electric scooter as described in claim 3, characterized in that, The process of analyzing the power demand factor of the electric scooter based on the behavioral pattern data and external road condition data includes: First, the behavioral pattern data is mapped to numerical behavioral scores, where the aggressive mode corresponds to a score of 1.2, the stable mode corresponds to a score of 1.0, and the conservative mode corresponds to a score of 0.

8. The external road condition data is then converted into a road condition score through normalization processing. The external road condition data includes slope value, road surface type coefficient and traffic density coefficient. Finally, the behavior score and road condition score are linearly combined according to preset weights to generate the power demand factor, where the preset weight for the behavior score is 0.6 and the weight coefficient for the road condition score is 0.

4.

5. The method for driving an electric scooter as described in claim 1, characterized in that, The process of adjusting the parameters of the electric scooter's motor controller based on the theoretical power output value to dynamically adjust the motor power output and obtain adjusted motor power data includes: The power deviation value is obtained by subtracting the current motor power data from the theoretical power output value. Based on the power deviation value, a proportional-integral-derivative (PI-DE) control calculation is performed to obtain the controller adjustment parameters. The mathematical expression for the PI-DE control calculation is as follows: In the formula, Adjust the parameters for the controller. It is a function of the power deviation value changing with time. This is the proportionality coefficient. The integral coefficient is... The differential coefficients are... It is a time variable; Based on the controller adjustment parameters, the parameter settings of the motor controller are updated to obtain the adjusted motor power data.

6. The method for driving an electric scooter as described in claim 1, characterized in that, The process of controlling the energy recovery process of the electric scooter based on the behavioral pattern data, deceleration event data, and energy recovery potential value, dynamically adjusting the recovery intensity, and generating recovered energy data includes: Based on the behavioral pattern data, the deceleration event data, and the energy recovery potential value, the behavioral pattern data is mapped to a behavioral influence coefficient through a recovery intensity calculation function. Then, the deceleration intensity in the deceleration event data is extracted as the main adjustment factor, and the energy recovery potential value is used as the benchmark value. Finally, the recovery intensity parameter is generated by weighted combination calculation. Based on the aforementioned recovery intensity parameter, the energy recovery process is dynamically adjusted using a recovery control algorithm to obtain adjusted recovery process data. Based on the adjusted recovery process data, an integral operation is performed using an energy calculation function to generate recovered energy data.

7. The method for driving an electric scooter as described in claim 1, characterized in that, Based on the adjusted motor power data and the recovered energy data, the system monitors the battery status, riding data, and energy recovery status of the electric scooter in real time, and optimizes the analysis process of the riding behavior pattern to obtain optimized behavior pattern data, including: Based on the adjusted motor power data and the recovered energy data, a comprehensive status index is calculated using a data fusion algorithm. Based on the comprehensive state index, the behavior pattern analysis parameters are adjusted through an optimization function to obtain optimized analysis parameters; Based on the optimized analysis parameters, the cycling behavior pattern is re-analyzed to obtain optimized behavior pattern data.

8. The method for driving an electric scooter as described in claim 7, characterized in that, The mathematical expression of the optimization function is as follows: In the formula, To optimize the analysis parameters, Parameters for analyzing the current behavioral pattern. This is the learning rate coefficient. This represents the gradient of the loss function with respect to the parameters.

9. The method for driving an electric scooter as described in claim 1, characterized in that, Based on the adjusted motor power data and recovered energy data, the system outputs adjusted driving status and energy recovery information to the user, providing a recovery efficiency report, including: Based on the adjusted motor power data and recovered energy data, the adjusted motor power data is divided by the preset rated power value through the state efficiency calculation function to generate a power normalization value as a driving state index. Then, the recovered energy data is divided by the total energy consumption estimate to generate an energy normalization value as a recovery efficiency index. Based on the driving status index and the energy recovery efficiency index, the driving status index is classified into low, medium and high power states according to the numerical range of the driving status index by the information generation algorithm, and corresponding descriptive text is generated to obtain user-readable driving status information and energy recovery information. Based on the user-readable driving status information and energy recovery information, a recovery efficiency report is generated.

10. An electric scooter, used to implement the method of driving the electric scooter according to claims 1-9, characterized in that, The electric scooter includes: The cycling behavior pattern analysis module is used to analyze user cycling behavior patterns based on sensor data from electric scooters, identify deceleration events, and generate behavior pattern data and deceleration event data. The road condition power demand prediction module is used to calculate the theoretical power output value and energy recovery potential value of the electric scooter based on the behavior pattern data, the deceleration event data and the external road condition data. An adaptive power adjustment module is used to adjust the parameters of the motor controller of the electric scooter based on the theoretical power output value, so as to dynamically adjust the motor power output and obtain the adjusted motor power data. The energy recovery execution module is used to control the energy recovery process of the electric scooter based on the behavior pattern data, deceleration event data and energy recovery potential value, dynamically adjust the recovery intensity, and generate recovered energy data; The real-time optimization monitoring module is used to monitor the battery status, driving data and energy recovery status of the electric scooter in real time based on the adjusted motor power data and the recovered energy data, and optimize the analysis process of the riding behavior pattern to obtain optimized behavior pattern data. The user interaction and output module is used to output the adjusted driving status and energy recovery information to the user based on the adjusted motor power data and recovered energy data, and to provide a recovery efficiency report.

Images