A method for adaptive calibration of a vehicle acceleration brake map
By employing a recursive least squares algorithm to update the acceleration and braking mapping table in real time in autonomous vehicles, the problems of low calibration efficiency and insufficient accuracy in existing technologies are solved. This achieves efficient and real-time mapping table calibration, adapts to individual vehicle differences and environmental changes, and improves control accuracy.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- YIXIAN INTELLIGENCE
- Filing Date
- 2026-03-04
- Publication Date
- 2026-06-02
AI Technical Summary
Existing technologies for calibrating vehicle acceleration and braking maps in autonomous vehicles suffer from low efficiency, poor real-time performance, insufficient accuracy, and poor adaptability, failing to adapt to individual vehicle differences and changes in environmental factors.
The Recursive Least Squares (RLS) algorithm is used to collect vehicle status data in real time. Through data verification, pitch angle compensation and delay compensation, the mapping table is updated adaptively. Combined with a multi-dimensional data verification mechanism and accuracy evaluation, the real-time performance and accuracy of the calibration are ensured.
It enables automatic real-time calibration of vehicle acceleration and braking mapping tables, improving calibration efficiency and accuracy, ensuring the quality of calibration data, adapting to different application scenarios, and eliminating the impact of road slope on calibration accuracy.
Smart Images
Figure CN122132656A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of autonomous vehicle control technology, and in particular to an adaptive calibration method for a vehicle acceleration and braking mapping table. Background Technology
[0002] In autonomous driving systems, vehicle control precision directly impacts driving safety and comfort. The acceleration-braking mapping table is a core component of the vehicle control system, used to establish the mapping relationship between pedal position, vehicle speed, and acceleration; it is typically stored in a two-dimensional table format. Due to individual vehicle differences, age, and environmental conditions, each vehicle's actual acceleration and braking characteristics vary. Directly using a universal mapping table would lead to a decrease in control precision.
[0003] Currently, common calibration methods mainly include: static calibration methods, which require manual data collection and adjustment under specific test conditions, which are time-consuming and cannot adapt to changes in vehicle characteristics; offline calibration methods, which involve collecting a large amount of historical data and then analyzing and updating it offline, resulting in long calibration cycles and the inability to respond in real time; and calibration methods based on fixed parameters, which do not consider individual vehicle differences and environmental factors (such as road slope), resulting in limited calibration accuracy.
[0004] These existing technologies generally suffer from drawbacks such as low calibration efficiency, poor real-time performance, insufficient accuracy, poor adaptability, and inability to guarantee data quality. Therefore, there is an urgent need for a method that can automatically, in real-time, and with high precision complete calibration during normal vehicle operation. Summary of the Invention
[0005] The purpose of this invention is to provide an automatic real-time calibration of a vehicle acceleration and braking mapping table to improve calibration efficiency and accuracy, and ensure the quality of calibration data.
[0006] To achieve the above objectives, the present invention includes the following steps: Step S1: Collect vehicle status data, including vehicle speed, pedal position, steering angle, and pitch angle; Step S2: Perform a validity check on the collected vehicle status data. If the data is valid, perform data verification and filter out abnormal data. If the data is invalid, record the reason for failure and return. Step S3: Process the verified vehicle status data, perform pitch angle compensation and delay compensation, and use the processed data to look up the mapping table index; Step S4: Based on the speed and pedal value, find the corresponding index position in the mapping table, and update the mapping table using the recursive least squares (RLS) algorithm; Step S5: Calculate the root mean square error between the predicted acceleration and the actual measured acceleration before and after calibration, and evaluate the calibration effect based on the error comparison results; Step S6: Output the calibrated mapping table.
[0007] Preferably, the data verification in step S2 includes: Speed verification: Verify whether the current speed is greater than the minimum speed threshold; Speed index matching verification: Verify whether the difference between the current vehicle speed and the speed index value in the mapping table is less than the speed index matching threshold; Pedal index matching verification: Verify whether the difference between the current pedal position and the pedal index value in the mapping table is less than the pedal index matching threshold; Steering angle verification: Verify whether the absolute value of the steering angle is less than the steering angle threshold; Pitch angle verification: Verify whether the absolute value of the pitch angle is less than the pitch angle threshold; Jerkness verification: Verify whether the absolute value of the rate of change of acceleration is less than the jerkness threshold; Pedal speed verification: Verify whether the pedal movement speed is less than the pedal speed threshold; Pedal Mutual Exclusion Verification: Verify whether the accelerator pedal and brake pedal are not both greater than zero at the same time.
[0008] Preferably, pitch angle compensation is achieved in the following ways: The formula for calculating acceleration is: ,in It is the acceleration due to gravity. Let be the pitch angle. When the vehicle is going uphill, the pitch angle is positive, and the acceleration decreases after compensation. When the vehicle is going downhill, the pitch angle is negative, and the acceleration increases after compensation.
[0009] Preferably, the mapping table is updated by finding pedal data in the historical pedal data queue that differs from the current time by a preset delay time, and using this data.
[0010] Preferably, when updating the mapping table using the Recursive Least Squares (RLS) algorithm in step S4, at least one of the following update modes is supported: Update cell by cell: Perform RLS update independently on each cell in the mapping table; Global offset update: Update the entire mapping table by applying a uniform offset; Four-cell surrounding update: Performs RLS update on the four adjacent cells of the current data point simultaneously.
[0011] Preferably, in the four-cell update mode, the normalized distance is calculated based on the position of the current data point among the four cells, and the update is performed based on the weights assigned to the distance.
[0012] Preferably, step S4 further includes maintaining the consistency of the mapping table after updating it, ensuring that the mapping table satisfies at least one of the following physical constraints: For the acceleration mapping table, at the same speed, the larger the pedal value, the greater the acceleration; at the same pedal value, the larger the speed, the smaller the acceleration. For the brake mapping table, at the same speed, the larger the brake pedal value, the greater the deceleration; at the same pedal value, the greater the speed, the smaller the deceleration.
[0013] Preferably, step S5 includes: calculating the root mean square error (RMSE) between the acceleration predicted using the original mapping table and the actual measured acceleration using the calibrated mapping table, respectively; Calculate the ratio of the RMSE of the calibrated mapping table to the RMSE of the original mapping table; If the ratio is less than a preset threshold, the calibration is deemed valid.
[0014] Preferably, step S1 further includes preprocessing the collected vehicle status data, including at least one of the following: The acceleration is limited and low-pass filtered; Apply low-pass filtering to the pitch angle; Calculate the jerkiness and apply amplitude limiting and low-pass filtering to it; Calculate the pedal speed and perform low-pass filtering.
[0015] The beneficial effects of this invention are as follows: The adaptive mapping table update method based on the RLS algorithm adopts the recursive least squares algorithm to adaptively update the corresponding cells in the mapping table according to the real-time collected vehicle status data, thereby realizing the automatic calibration of the mapping table. The system boasts high accuracy. By acquiring the vehicle's pitch angle, it calculates the longitudinal component of gravitational acceleration and compensates for the measured acceleration, eliminating the impact of road slope on calibration accuracy. Considering the delay between pedal command and actual acceleration, it uses historical pedal data for delay compensation, thereby improving calibration accuracy. A multi-dimensional data verification mechanism is adopted, which includes speed, pedal position, steering angle, pitch angle, jerkiness, and pedal speed, to effectively filter out abnormal data. It offers three modes: update cell by cell, overall offset update, and update around four cells, to suit different application scenarios; After updating the mapping table, check and maintain the physical consistency of the mapping table to ensure that the acceleration mapping table and the braking mapping table satisfy the monotonicity constraint. The real-time accuracy assessment method evaluates the calibration effect in real time by calculating the RMSE of the original mapping table and the calibrated mapping table, and determines whether the mapping table needs to be updated based on the error ratio. Attached Figure Description
[0016] Figure 1This is a system structure block diagram of one embodiment of the present invention.
[0017] Figure 2 This is a system main flowchart of one embodiment of the present invention.
[0018] Figure 3 This is a flowchart of data acquisition and preprocessing according to an embodiment of the present invention.
[0019] Figure 4 This is a flowchart of the RLS update algorithm according to an embodiment of the present invention (taking mode 1 as an example).
[0020] Figure 5 This is a diagram showing the four-cell update calibration result according to an embodiment of the present invention. Detailed Implementation
[0021] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings. The method described in the embodiments of the present invention can be executed by an on-board computing platform or similar device including a processor, memory, and necessary sensors.
[0022] This invention proposes an adaptive calibration method for vehicle acceleration and braking mapping tables based on the Recursive Least Squares (RLS) algorithm. This method collects vehicle state data in real time during normal vehicle operation, processes the data through data verification, pitch angle compensation, and delay compensation, adaptively updates the corresponding cells in the mapping table using the RLS algorithm, evaluates the calibration accuracy in real time, and finally outputs the calibrated mapping table.
[0023] like Figure 1 As shown, one embodiment of the present invention includes the following modules: a data acquisition module, which includes vehicle-mounted sensors, etc., for acquiring state data such as vehicle speed, pedal position, steering angle, and pitch angle; a data verification module for validating the acquired data and filtering abnormal data; a data processing module for pitch angle compensation and delay compensation; a mapping table update module for updating the mapping table using the RLS algorithm; an accuracy evaluation module for calculating the root mean square error (RMSE) to evaluate the calibration accuracy; and a mapping table output module for outputting the calibrated mapping table.
[0024] like Figure 2As shown, the specific process of one embodiment of the present invention begins with system initialization. First, a default mapping table is loaded, a verification threshold is set, and system parameters such as RLS parameters are initialized. After a timer is started, the system begins the mapping table update process. When the timer is triggered, the system collects vehicle status data, including speed, pedal position, steering angle, and pitch angle. After data collection, the system performs a data validity check, determining whether the data is empty or has timed out. If the data is valid, it enters the data verification stage, performing eight checks. If verification fails, the system records the reason for the failure and returns; if verification passes, data processing is performed, including pitch angle compensation and delay compensation. The processed data is used to look up mapping table indices, including pedal and speed indices. According to the update mode, the system executes the RLS update algorithm. After the update is completed, the system performs consistency maintenance, checking the monotonicity of the mapping table and correcting outliers. Then, an accuracy evaluation is performed, calculating RMSE and error ratio. After the evaluation is completed, the system publishes the results, including mapping table visualization, error information, and update suggestion flags. Finally, the system periodically outputs a CSV file, completing one update cycle.
[0025] The initialization process during system startup is as follows: Parameter loading: Load all configuration parameters from the ROS parameter server, including verification thresholds, RLS parameters, update modes, etc.; Default mapping table loading: Load the default acceleration mapping table and braking mapping table (CSV format) from the specified directory; Data structure initialization: Initialize the calibrated mapping table (copy the values of the default mapping table), initialize the covariance matrix (mode 1 and mode 3), initialize the offset vector, and initialize the data queue; ROS interface initialization: Subscribe to vehicle status topics (speed, steering, actuator status), publish mapping table visualization topics, publish error information topics, and register service interfaces; Timer startup: Includes the main update timer (executes mapping table updates) and the CSV output timer (outputs the calibrated mapping table); Diagnostic system initialization: Initializes the diagnostic updater to publish calibration status information.
[0026] The data acquisition module collects vehicle status data in real time and performs preprocessing. The system obtains longitudinal speed through the vehicle speed sensor. The accelerator pedal position is obtained through the actuator state. and brake pedal position The steering angle (steer) is obtained through the steering sensor, and then transformed from the map coordinate system to the map coordinate system via coordinate transformation (TF). Obtain the vehicle's pitch angle using a coordinate system.
[0027] The detailed process of the data acquisition module is as follows: Figure 3As shown, the system first performs a timeout detection to determine if the data is within the valid time range. If the data times out, the system skips the update; if the data is valid, data acquisition and preprocessing are performed. The system acquires speed data from the speed sensor, stores it in the speed data queue, and then searches for the data with a time difference in the historical queue. The velocity data is used to calculate the raw acceleration through differential processing, followed by amplitude limiting and low-pass filtering to obtain the final acceleration.
[0028] The system calculates jerk from acceleration data, maintains the previous acceleration value and timestamp, and if the time interval is valid, calculates the original jerk through differential calculation, followed by amplitude limiting and low-pass filtering. The system acquires pedal data from the actuator status and stores it in the pedal data queue, then searches the historical queue for time differences... The system calculates pedal speed using differential calculations based on pedal data and then processes it using a low-pass filter. The pitch angle data is also low-pass filtered. All processed data is stored in corresponding data queues, managed using a FIFO (First-In, First-Out) strategy; the oldest data is automatically deleted when a queue is full.
[0029] The data validation module verifies the validity of the collected data and filters out abnormal data. Specifically, it filters out the following abnormal situations: speed verification requires the current speed to be greater than a minimum speed threshold. To avoid the impact of stationary or reversing data on calibration; speed index matching verification requires that the difference between the current speed and the speed index value in the mapping table must be less than a threshold. Ensure that the data points fall within the valid speed range; the pedal index matching verification requires that the difference between the current pedal position and the pedal index value in the mapping table must be less than a threshold. Ensure that the data points fall within the valid pedal range; steering angle verification requires that the absolute value of the steering angle must be less than the threshold. To avoid the effects of lateral acceleration during cornering; pitch angle verification requires that the absolute value of the pitch angle must be less than a threshold. To avoid the influence of steep slopes on acceleration measurements; jerkiness verification requires that the absolute value of jerkiness (the rate of change of acceleration) must be less than a threshold. To avoid the effects of sudden acceleration or braking; pedal speed verification requires that the pedal movement speed must be less than a threshold. To avoid data instability when the pedal changes rapidly; the pedal mutual exclusion verification requires that the accelerator pedal and brake pedal cannot be greater than zero at the same time to ensure the correctness of the data logic.
[0030] The detailed process of the data validation module involves the system performing eight validation checks sequentially. First, it checks if the speed is below a minimum threshold; if so, validation fails. Next, it checks the speed index matching, determining if the difference between the current speed and the speed index value in the mapping table is within an acceptable range. Then, it checks the pedal index matching, determining if the difference between the current pedal position and the pedal index value in the mapping table is within an acceptable range. Following this, it checks if the steering angle exceeds a maximum threshold; if so, validation fails. Next, it checks if the pitch angle exceeds a maximum threshold; if so, validation fails. Then, it checks if the jerk exceeds a maximum threshold; if so, validation fails. Next, it checks if the pedal speed exceeds a threshold; if so, validation fails. Finally, it checks pedal mutual exclusion; if both the accelerator and brake pedals are greater than zero, validation fails. Only after passing all eight checks can the data be used for mapping table updates.
[0031] The data processing module performs pitch angle compensation and delay compensation on the verified data. Since road slope affects acceleration measurements, pitch angle compensation is necessary. The formula for calculating the compensated acceleration is as follows: ,in It is the acceleration due to gravity. The pitch angle is the angle of acceleration. When the vehicle is going uphill, the pitch angle is positive, and the compensated acceleration will decrease; when the vehicle is going downhill, the pitch angle is negative, and the compensated acceleration will increase.
[0032] Because there is a delay between the pedal command and the actual acceleration, historical pedal data is needed to compensate for this delay. The system maintains a historical pedal data queue. When the mapping table needs to be updated, the system retrieves the pedal data from the queue before the delay time and uses this data to update the mapping table.
[0033] The mapping table update module uses a recursive least squares algorithm to update the mapping table. The system searches for the corresponding index position in the mapping table based on the current speed and pedal value. Search methods include exact match, nearest neighbor, and lower bound search, each used for different update modes.
[0034] The mapping table update employs a recursive least squares algorithm, supporting three update modes. The algorithm uses a forgetting factor, which causes the weight of historical data to decay over time, giving greater weight to new data. A larger forgetting factor value results in a slower response to new data, but better stability. In the four-cell update mode, an adaptive forgetting factor is used to prevent covariance matrix saturation.
[0035] The detailed process of the RLS update algorithm is as follows: Figure 5 As shown, taking Mode 1 (updating cell by cell) as an example, the system first reads the current cell state, including offset, covariance, and prediction acceleration. Then, it calculates the design matrix, updates the covariance matrix, calculates the gain coefficient, and calculates the error. Update offset Finally, update the mapping table values. .
[0036] in It is the compensated acceleration. It is the acceleration predicted by the mapping table; It is the offset at the current moment. It is the offset from the previous moment. It is the gain coefficient; This is the updated mapping table. It is the original mapping table.
[0037] Mode 1 updates cell by cell, updating each cell in the mapping table independently. For each cell, the system calculates the covariance update and gain coefficient, then updates the offset based on the prediction error, and finally adds the offset to the original mapping table value to obtain the updated mapping table value.
[0038] Mode 2 is a global offset update, applying a uniform offset to the entire mapping table. The system calculates the global offset and applies the same offset to all cells; the update formula is as follows. .
[0039] in It is the global offset. It is the updated mapping table value. It is a pedal index. It is a speed index.
[0040] Mode 3 involves updating the four cells surrounding the current data point simultaneously. The system calculates the normalized distance based on the current data point's position among these four cells, then uses the design vector and state vector to calculate the predicted value, performs RLS updates, and finally applies the updated offsets to the four cells respectively. Figure 5 As shown: The updated mapping table must satisfy physical constraints to ensure its monotonicity. For the acceleration mapping table, at the same speed, a larger pedal value results in greater acceleration; conversely, at the same pedal value, greater speed results in smaller acceleration. For the braking mapping table, at the same speed, a larger brake pedal value results in greater deceleration; conversely, at the same pedal value, greater speed results in smaller deceleration. If the updated mapping table does not satisfy these constraints, the system performs consistency correction, fine-tuning the mapping table values to ensure monotonicity.
[0041] The accuracy assessment module evaluates calibration accuracy by calculating the root mean square error (RMSE). The system calculates the prediction errors of the original mapping table and the calibrated mapping table in real time, and judges the calibration effect by comparing the magnitude of the two errors. Specifically, the system uses the original mapping table to predict acceleration, calculates the RMS error between the predicted and actual measured values to obtain the original mapping table error; it then uses the calibrated mapping table to predict acceleration, calculates the RMS error between the predicted and actual measured values to obtain the calibrated mapping table error; finally, it calculates the ratio of the calibrated mapping table error to the original mapping table error. When the error ratio is less than a set threshold, it indicates a significant improvement in the accuracy of the calibrated mapping table, and the system recommends updating the mapping table.
[0042] The system uses multiple error data queues to store historical error data, used for long-term and short-term evaluations respectively. For each data point, the system looks up the predicted acceleration from the corresponding mapping table based on the current pedal value and velocity value, then calculates the difference between the predicted acceleration and the actual measured acceleration to obtain the error value. The system performs statistical calculations on all error values to obtain the root mean square error (RMSE). By comparing the RMSSE of the original mapping table and the calibrated mapping table, the system determines whether the calibration is effective. If the error of the calibrated mapping table is significantly smaller than that of the original mapping table, the system recommends updating the mapping table.
[0043] The system also provides a data quality assessment function, which counts the number of data points collected in each mapping table cell, evaluates the distribution and coverage of the data, and helps users understand the calibration progress and data quality. The system can display the mapping table status, data collection status, and error information in real time, facilitating user monitoring of the calibration process.
[0044] The mapping table output module outputs the calibrated mapping tables. The system periodically outputs the calibrated mapping tables as comma-separated value (CSV) files, including acceleration and braking mapping tables, each containing a speed index, pedal index, and corresponding acceleration value. The file format is a two-dimensional table; the first row contains the speed index values, the first column contains the pedal index values, and the remaining cells contain the corresponding acceleration values. The specific format is: the first row contains identifiers, v0, v1, v2, ..., v n Subsequent actions pᵢ, aᵢ0, aᵢ1, aᵢ2, ..., aᵢ n Where vⱼ (j = 0, 1, ..., n) is the velocity index value, pᵢ (i = 0, 1, ..., m) is the pedal index value, and aᵢⱼ is the acceleration value of the corresponding cell (i, j). The system also provides a service interface that allows users to save the calibrated mapping table at any time. The service request includes the target directory path, and the system saves the calibrated mapping table to the specified directory. The output CSV format is shown in the table below:
[0045] Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.
Claims
1. An adaptive calibration method for a vehicle acceleration and braking mapping table, characterized in that, Includes the following steps: Step S1: Collect vehicle status data, including vehicle speed, pedal position, steering angle, and pitch angle; Step S2: Perform a validity check on the collected vehicle status data. If it is valid, perform data verification and filter out abnormal data. If it is invalid, record the reason for failure and return. Step S3: Process the verified vehicle status data, perform pitch angle compensation and delay compensation, and use the processed data to look up the mapping table index; Step S4: Based on the speed and pedal value, find the corresponding index position in the mapping table, and update the mapping table using the recursive least squares (RLS) algorithm; Step S5: Calculate the root mean square error between the predicted acceleration and the actual measured acceleration before and after calibration, and evaluate the calibration effect based on the error comparison results; Step S6: Output the calibrated mapping table.
2. The adaptive calibration method for a vehicle acceleration and braking mapping table according to claim 1, characterized in that, The data verification in step S2 includes: Speed verification: Verify whether the current speed is greater than the minimum speed threshold; Speed index matching verification: Verify whether the difference between the current vehicle speed and the speed index value in the mapping table is less than the speed index matching threshold; Pedal index matching verification: Verify whether the difference between the current pedal position and the pedal index value in the mapping table is less than the pedal index matching threshold; Steering angle verification: Verify whether the absolute value of the steering angle is less than the steering angle threshold; Pitch angle verification: Verify whether the absolute value of the pitch angle is less than the pitch angle threshold; Jerkness verification: Verify whether the absolute value of the rate of change of acceleration is less than the jerkness threshold; Pedal speed verification: Verify whether the pedal movement speed is less than the pedal speed threshold; Pedal Mutual Exclusion Verification: Verify whether the accelerator pedal and brake pedal are not both greater than zero at the same time.
3. The adaptive calibration method for a vehicle acceleration and braking mapping table according to claim 2, characterized in that, The pitch angle compensation is achieved in the following way: The formula for calculating acceleration is: ,in It is the acceleration due to gravity. Let be the pitch angle. When the vehicle is going uphill, the pitch angle is positive, and the acceleration decreases after compensation. When the vehicle is going downhill, the pitch angle is negative, and the acceleration increases after compensation.
4. The adaptive calibration method for a vehicle acceleration and braking mapping table according to claim 2, characterized in that, The delay compensation is achieved by searching for pedal data in the historical pedal data queue that differs from the current time by a preset delay time, and then using this data to update the mapping table.
5. The adaptive calibration method for a vehicle acceleration and braking mapping table according to any one of claims 1-4, characterized in that, When updating the mapping table using the Recursive Least Squares (RLS) algorithm in step S4, at least one of the following update modes is supported: Update cell by cell: Perform RLS update independently on each cell in the mapping table; Global offset update: Update the entire mapping table by applying a uniform offset; Four-cell surrounding update: Performs RLS update on the four adjacent cells of the current data point simultaneously.
6. The adaptive calibration method for a vehicle acceleration and braking mapping table according to claim 5, characterized in that, In the four-cell update mode, the normalized distance is calculated based on the position of the current data point among the four cells, and the update is performed by assigning weights based on the distance.
7. The adaptive calibration method for a vehicle acceleration and braking mapping table according to claim 6, characterized in that, Step S4 further includes maintaining the consistency of the mapping table after updating it, ensuring that the mapping table satisfies at least one of the following physical constraints: For the acceleration mapping table, at the same speed, the larger the pedal value, the greater the acceleration; at the same pedal value, the larger the speed, the smaller the acceleration. For the brake mapping table, at the same speed, the larger the brake pedal value, the greater the deceleration; at the same pedal value, the greater the speed, the smaller the deceleration.
8. The adaptive calibration method for a vehicle acceleration and braking mapping table according to claim 7, characterized in that, Step S5 includes: Calculate the root mean square error (RMSE) between the acceleration predicted using the original mapping table and the actual measured acceleration using the calibrated mapping table, respectively. Calculate the ratio of the RMSE of the calibrated mapping table to the RMSE of the original mapping table; If the ratio is less than a preset threshold, the calibration is deemed valid.
9. The adaptive calibration method for a vehicle acceleration and braking mapping table according to claim 8, characterized in that, Step S1 further includes preprocessing the collected vehicle status data, including at least one of the following: The acceleration is limited and low-pass filtered; Low-pass filtering is applied to the pitch angle; Calculate the jerkiness and apply amplitude limiting and low-pass filtering to it; Calculate the pedal speed and perform low-pass filtering.