A method for controlling a permanent magnet synchronous motor to achieve driving mode switching

Abstract

A method for controlling a permanent magnet synchronous motor to achieve driving mode switching relates to a new energy vehicle motor control method, comprising the following steps: S1 sensor data acquisition: a microcontroller regularly collects camera image data, acceleration and angle data from an inertial measurement unit, and data from a seat pressure sensor; S2 data processing and analysis: an S32K144 microcontroller processes and analyzes the collected data; S3 driving mode decision-making: based on the processed data, a fuzzy rule-based driving mode switching decision-making mechanism is used to determine one of three driving modes: sport, economy, and comfort; and S4 motor control strategy adjustment: the microcontroller adjusts the motor control strategy based on the selected driving mode. The present invention can automatically adjust the driving mode by comprehensively considering different road condition information and driving behavior information. It features excellent flexibility, high intelligence, high reliability, and low power consumption, and can comprehensively improve driving comfort and power response.

Description

Technical Field

[0001] The present invention relates to a method for controlling a motor of a new energy vehicle, and in particular to a method for controlling a permanent magnet synchronous motor to achieve driving mode switching. Background Art

[0002] In the field of new energy vehicles, diverse driving modes are crucial for meeting diverse driving needs and enhancing the driving experience. However, current methods for switching driving modes have numerous drawbacks. Traditional electric vehicle driving mode switching is typically achieved through mechanical switches or simple electronic signal switching. For example, the driver uses an in-vehicle mode selector button to send a discrete signal to a controller, which then changes the motor's torque and speed control strategy based on preset parameters. While this method is relatively straightforward, it lacks flexibility and intelligence.

[0003] Currently, there are some control methods on the market that can achieve automatic driving mode switching. However, these methods often rely on a single or a few parameters to make decisions, failing to comprehensively consider complex factors such as road conditions and driving behavior. For example, judging traffic conditions based solely on vehicle speed ignores key information such as road type and vehicle distribution. This leads to inaccurate and untimely driving mode switching, making it difficult to adapt to the diversity of real-world driving scenarios.

[0004] Furthermore, existing methods for controlling permanent magnet synchronous motors (PMSMs) to achieve driving mode switching suffer from a crude and simplistic motor control strategy. They cannot precisely adjust the motor's speed and torque output between driving modes, failing to balance power performance, energy efficiency, and driving comfort. In economy mode, motor efficiency cannot be effectively optimized, resulting in energy waste. In sport mode, the motor's power advantage is not fully utilized. In comfort mode, motor vibration and noise cannot be effectively suppressed. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method for controlling a permanent magnet synchronous motor to achieve driving mode switching, so as to solve the shortcomings of the prior art such as lack of intelligence and flexibility and inaccurate driving mode switching.

[0006] The technical solution to the above technical problem is: a method for controlling a permanent magnet synchronous motor to achieve driving mode switching, comprising the following steps:

[0007] S1: Sensor Data Acquisition

[0008] After the vehicle is started, various sensors begin to work, and the microcontroller regularly collects image data from the camera, acceleration and angle data from the inertial measurement unit, and data from the seat pressure sensor;

[0009] S2: Data processing and analysis

[0010] The microcontroller processes and analyzes the collected data;

[0011] S3: Driving mode decision

[0012] According to the processed data, a driving mode switching decision mechanism based on fuzzy rules is used to make decisions on the three driving modes: sport, economy, and comfort.

[0013] S4: Motor control strategy adjustment

[0014] The microcontroller adapts the motor control strategy according to the selected driving mode.

[0015] A further technical solution of the present invention is: in step S1, before starting the vehicle, the following preparatory work is also performed:

[0016] The microcontroller is connected to the permanent magnet synchronous motor through a motor driver, and then a camera is installed on the car to identify road conditions. The inertial measurement unit is installed on the vehicle chassis to sense the dynamic characteristics of the vehicle, and the pressure sensor is installed on the seat to sense changes in the driver's sitting posture.

[0017] A further technical solution of the present invention is: the specific contents of step S2 include:

[0018] S21. Data transmission and preprocessing: Data from each sensor is transmitted to the microcontroller. The ADC and related signal processing units within the microcontroller preprocess the data, converting the analog signal into a digital signal, and performing filtering and amplification to obtain accurate sensor data.

[0019] S22. Data Processing and Analysis: Camera image data is processed using a CNN algorithm to identify road type. Acceleration and angle data from the inertial measurement unit are used to calculate the vehicle's acceleration rate dynamic parameters. Seat pressure sensor data is analyzed for pressure variations. Driving behavior analysis is performed using acceleration rate dynamic parameters and seat pressure variations.

[0020] A further technical solution of the present invention is: the data processing in step S22 is achieved by constructing a layered sensor fusion architecture; at the bottom layer, the data of the camera, inertial measurement unit and seat pressure sensor are time synchronized and spatially calibrated; timestamp technology is used to ensure that the data from different sensors are in the same time reference system, and the spatial data of different sensors are unified into the vehicle coordinate system through coordinate transformation; at the middle layer, Kalman filtering or particle filtering data fusion algorithm is used to fuse the sensor data processed at the bottom layer; at the high layer, a comprehensive judgment of driving intention and road conditions is made based on the fused data.

[0021] A further technical solution of the present invention is: in step S22, data analysis includes the following specific steps:

[0022] S221. Road type identification

[0023] S2211. Data processing: First, the road image captured by the camera is processed, including image cropping, normalization, and resizing;

[0024] S2212. Build a network architecture: Use the VGG-Net architecture, which consists of multiple consecutive convolutional and pooling layers, and finally a fully connected layer to map the extracted features to different road type categories.

[0025] S2213. Data training: Training is performed using labeled road image data, which includes highway image data, urban road image data, and mountain road image data. During training, the network weights are adjusted using a backpropagation algorithm to minimize the error between the network's output road type predictions and the true labels.

[0026] S222. Traffic Condition Identification

[0027] S2221. Use optical flow to estimate vehicle speed and indirectly determine traffic conditions. Specifically, by calculating the optical flow between adjacent frames, the speed of the vehicle relative to the surrounding environment is estimated. If the speed of most vehicles is low and relatively stable, it indicates traffic congestion; if the speed of the vehicles is high and varies greatly, it indicates smooth traffic.

[0028] S2222. Use an object detection algorithm to identify vehicles and traffic signs to assist in determining traffic conditions. Specifically, the image is divided into multiple grid cells, with each cell responsible for predicting a certain number of objects. The prediction results for each cell are post-processed to accurately determine the location and category of vehicles and traffic signs. If a large number of vehicles are detected waiting in line or stop signs are frequently seen, traffic congestion is indicated. If vehicles are sparsely distributed and traffic signs primarily indicate high speed limits, traffic is smooth.

[0029] S223. Driving behavior analysis

[0030] S2231. Driving Behavior Analysis Based on Inertial Measurement Unit (IMU) Data

[0031] S22311. Acceleration Data Processing: Acceleration data collected by the Inertial Measurement Unit (IMU) includes acceleration information about the vehicle's X, Y, and Z axes. By integrating this acceleration data, we obtain vehicle velocity and displacement information. This acceleration data is filtered to remove noise interference and obtain more accurate vehicle dynamic information.

[0032] S22312. Angular velocity data processing: The inertial measurement unit (IMU) collects the vehicle's angular velocity information to determine the vehicle's steering status. A sustained non-zero angular velocity indicates the vehicle is turning. The magnitude and duration of the angular velocity are analyzed to determine the severity of the turn. A high angular velocity that persists for a period of time indicates a sharp turn, while a low angular velocity indicates a gentle turn or a slight directional adjustment.

[0033] S2232. Driving behavior analysis based on seat pressure sensor data

[0034] S22321. Pressure Change Monitoring: The pressure data collected by the seat pressure sensor reflects changes in the driver's sitting posture. By performing differential calculations on the pressure data, the magnitude and frequency of pressure changes are monitored. If the pressure fluctuates significantly within a short period of time, it indicates that the driver is leaning forward or moving left or right, which can be related to stressful driving.

[0035] S22322. Pressure distribution analysis: By analyzing the pressure distribution in different areas of the seat, if the pressure in the front of the seat increases and the pressure in the back decreases, it means that the driver's body is leaning forward, and the analysis shows that he is accelerating or focusing on the road conditions. If the pressure on both sides of the seat is uneven, it means that the driver's body is leaning to one side, and the analysis shows that it is related to the steering operation.

[0036] A further technical solution of the present invention is that step S3 includes the following specific steps:

[0037] S31. Parameter fuzzification: Fuzzify the parameters related to road conditions and driving behavior; determine their membership in various fuzzy sets based on actual data;

[0038] S32. Establish a fuzzy rule base: Establish a fuzzy rule base based on experience and understanding of the characteristics of different driving modes;

[0039] S33. Fuzzy reasoning: Reasoning based on fuzzy rules to obtain fuzzy output of driving mode;

[0040] S34. Defuzzification: Use defuzzification methods to convert fuzzy outputs into definite driving modes.

[0041] A further technical solution of the present invention is: the fuzzy rule base in step S32 is: if the traffic condition is smooth and the driving behavior is aggressive, the driving mode is the sports driving mode; if the traffic condition is congested and the driving behavior is conservative, the driving mode is the economic mode or the comfortable driving mode.

[0042] A further technical solution of the present invention is: step S4 includes the following contents:

[0043] The microcontroller changes the motor control method by adjusting the PWM signal sent to the permanent magnet synchronous motor driver; in economy mode, the motor speed and torque output are reduced to optimize efficiency; in sports mode, the motor speed and torque output are increased to provide stronger power; in comfort mode, the motor speed and torque are maintained at a moderate level to reduce vibration and noise.

[0044] A further technical solution of the present invention is: the method further comprises step S5. Adaptive motor control parameter adjustment: adaptively adjusting the motor control parameters based on real-time feedback of the motor operating state.

[0045] A further technical solution of the present invention is that step S5 includes the following contents:

[0046] During motor operation, the motor's temperature, current, and voltage parameters are monitored in real time. When the motor temperature rises, the motor's power output is reduced and the current reference value in the magnetic field-oriented control algorithm is adjusted to prevent the motor from overheating. At the same time, the motor's control parameters are adaptively adjusted according to the vehicle's starting, acceleration, constant speed, and deceleration stages to improve the motor's efficiency and performance and better adapt to driving mode switching.

[0047] Due to the above structure, the method of controlling a permanent magnet synchronous motor to achieve driving mode switching according to the present invention has the following beneficial effects compared with the prior art:

[0048] 1. High intelligence and good flexibility

[0049] This invention regularly collects camera image data, acceleration and angle data from an inertial measurement unit (IMU), and data from seat pressure sensors. After analyzing and processing this data, it determines one of three driving modes: sport, economy, and comfort. A microcontroller controls the motor based on the selected driving mode. Consequently, this invention can automatically adjust the driving mode based on varying road conditions and driving behavior, eliminating the need for mechanical switches and demonstrating a high level of intelligence. Furthermore, this ability to comprehensively determine and adjust driving modes based on multiple information makes driving mode switching more flexible and adaptable to various practical driving scenarios.

[0050] In addition, since the present invention can obtain dynamic changes in road conditions and driving behavior in real time, the system can continuously and adaptively adjust the driving mode without the need for manual intervention by the driver, thereby improving driving convenience and flexibility.

[0051] 2. Improve the accuracy of driving mode switching decisions

[0052] This invention achieves data processing and analysis by constructing a layered sensor fusion architecture. At the bottom layer, data from cameras, inertial measurement units (IMUs), and seat pressure sensors are time-synchronized and spatially aligned. Timestamp technology ensures that data from different sensors are in the same time reference system, and coordinate transformation unifies the spatial data from these sensors into the vehicle coordinate system. At the middle layer, a Kalman filter or particle filter data fusion algorithm is used to fuse the processed sensor data from the bottom layer. For example, a Kalman filter can fuse road curvature information detected by the camera with vehicle angular velocity information measured by the IMU to more accurately predict the vehicle's trajectory. At the top layer, this fused data is used to comprehensively determine driving intent and road conditions. This layered architecture enables more efficient use of sensor data and improves the accuracy of driving mode switching decisions.

[0053] 3. Achieve a balance between power performance, energy efficiency and driving comfort

[0054] This invention fine-tunes the control method of the permanent magnet synchronous motor according to different driving modes. In Economy mode, the motor speed and torque output are reduced, effectively optimizing energy efficiency and reducing energy consumption. In Sport mode, the motor speed and torque are increased to fully unleash the motor's power, satisfying the driver's pursuit of passionate driving. In Comfort mode, the motor speed and torque are maintained at moderate levels, significantly reducing vibration and noise, improving driving comfort, and providing a more pleasant driving environment for passengers.

[0055] Furthermore, the present invention monitors the motor's operating parameters, such as temperature, current, and voltage, in real time. When the motor temperature rises, it automatically reduces power output and adjusts the current reference value in the field-oriented control algorithm to prevent motor overheating, ensure safe and stable operation, and extend the motor's service life. Furthermore, during the vehicle's various driving phases, such as starting, acceleration, constant speed, and deceleration, the system adaptively adjusts motor control parameters to maintain optimal motor operation, improving motor efficiency and performance, further enhancing the smoothness and stability of driving mode switching, and ultimately enhancing overall vehicle performance and the driving experience.

[0056] Below, the technical features of a method for controlling a permanent magnet synchronous motor to achieve driving mode switching according to the present invention are further described in conjunction with embodiments. DETAILED DESCRIPTION Example

[0057] A method for controlling a permanent magnet synchronous motor to switch driving modes comprises the following steps:

[0058] S1: Sensor Data Acquisition

[0059] After the vehicle is started, various sensors begin to work, and the microcontroller regularly collects image data from the camera, acceleration and angle data from the IMU, and data from the seat pressure sensor.

[0060] S2: Data processing and analysis

[0061] The S32K144 microcontroller processes and analyzes the collected data;

[0062] S3: Driving mode decision

[0063] According to the processed data, a driving mode switching decision mechanism based on fuzzy rules is used to make decisions on the three driving modes: sport, economy, and comfort.

[0064] S4: Motor control strategy adjustment

[0065] The microcontroller adjusts the motor control strategy according to the selected driving mode;

[0066] S5. Adaptive motor control parameter adjustment

[0067] Based on real-time feedback of the motor's operating status, the motor control parameters are adaptively adjusted.

[0068] In step S1, before starting the vehicle, the following preparations are performed:

[0069] The microcontroller is connected to the permanent magnet synchronous motor through a motor driver, and then a camera is installed on the car to identify road conditions. The inertial measurement unit is installed on the vehicle chassis to sense the dynamic characteristics of the vehicle, and the pressure sensor is installed on the seat to sense changes in the driver's sitting posture.

[0070] The specific contents of step S2 include:

[0071] S21. Data transmission and preprocessing: Data from each sensor is transmitted to the microcontroller. The ADC and related signal processing units within the microcontroller preprocess the data, converting the analog signal into a digital signal, and performing filtering and amplification to obtain accurate sensor data.

[0072] S22. Data Processing and Analysis: Camera image data is processed using a CNN algorithm to identify road types. Acceleration and angle data from the inertial measurement unit (IMU) are used to calculate the vehicle's acceleration rate dynamic parameters. Seat pressure sensor data is analyzed for pressure variations. Driving behavior analysis is performed using acceleration rate dynamic parameters and seat pressure variations.

[0073] The data processing in step S22 is achieved by constructing a layered sensor fusion architecture. At the bottom layer, data from the camera, inertial measurement unit (IMU), and seat pressure sensor are time-synchronized and spatially aligned. Timestamp technology is used to ensure that data from different sensors are in the same time reference system, and coordinate transformation is used to unify the spatial data from different sensors into the vehicle coordinate system. At the middle layer, a Kalman filter or particle filter data fusion algorithm is used to fuse the processed sensor data from the bottom layer. For example, a Kalman filter can fuse road curvature information detected by the camera with vehicle angular velocity information measured by the IMU to more accurately predict the vehicle's driving trajectory. At the top layer, a comprehensive assessment of driving intent and road conditions is made based on this fused data. This layered architecture enables more efficient use of sensor data and improves the accuracy of driving mode switching decisions.

[0074] In step S22, the specific steps of data analysis are as follows:

[0075] S221. Road type identification

[0076] S2211. Data processing: First, the road image captured by the camera is processed, including image cropping, normalization, and resizing;

[0077] S2212. Build a network architecture: Use the VGG-Net architecture, which consists of multiple consecutive convolutional and pooling layers, and finally a fully connected layer to map the extracted features to different road type categories.

[0078] S2213. Data training: Training is performed using labeled road image data, which includes highway image data, urban road image data, and mountain road image data. During training, the network weights are adjusted using a backpropagation algorithm to minimize the error between the network's output road type predictions and the true labels.

[0079] S222. Traffic Condition Identification

[0080] S2221. Use optical flow to estimate vehicle speed and indirectly determine traffic conditions. Optical flow is the velocity vector of objects moving in an image. By calculating the optical flow between adjacent frames, the vehicle's speed relative to the surrounding environment is estimated. A specific algorithm, such as the Lucas-Kanade optical flow algorithm, assumes that object motion is linear within a small local area. For each pixel in the image, its displacement between adjacent frames is calculated. This pixel displacement is then converted into actual speed based on camera parameters (such as focal length). If the speeds of most vehicles are low and relatively stable, this may indicate traffic congestion. If the speeds of vehicles are high and highly variable, this may indicate smooth traffic.

[0081] S2222. Use an object detection algorithm to identify vehicles and traffic signs and assist in determining traffic conditions: The image is divided into multiple grid cells, with each cell responsible for predicting a certain number of objects. The prediction results for each cell are post-processed to accurately determine the location and category of vehicles and traffic signs. If a large number of vehicles are detected waiting in line or if stop signs are frequently seen, this indicates traffic congestion. If vehicles are sparsely distributed and traffic signs primarily indicate high speed limits, traffic is flowing smoothly.

[0082] S223. Driving behavior analysis

[0083] S2231. Driving Behavior Analysis Based on Inertial Measurement Unit (IMU) Data

[0084] S22311. Acceleration Data Processing: Acceleration data collected by the inertial measurement unit (IMU) includes acceleration information along the vehicle's X, Y, and Z axes. By integrating this acceleration data, vehicle velocity and displacement information is obtained. For example, during linear acceleration, the X-axis acceleration will have a significant positive value; during cornering, the Y-axis acceleration will vary. Acceleration data is filtered to remove noise interference and obtain more accurate vehicle dynamic information.

[0085] S22312. Angular velocity data processing: The inertial measurement unit (IMU) collects the vehicle's angular velocity information to determine the vehicle's steering status. A sustained non-zero angular velocity indicates the vehicle is turning. The magnitude and duration of the angular velocity are analyzed to determine the severity of the turn. A high angular velocity that persists for a period of time indicates a sharp turn, while a low angular velocity indicates a gentle turn or a slight directional adjustment.

[0086] S2232. Driving behavior analysis based on seat pressure sensor data

[0087] S22321. Pressure Change Monitoring: The pressure data collected by the seat pressure sensor reflects changes in the driver's sitting posture. By performing differential calculations on the pressure data, the magnitude and frequency of pressure changes are monitored. If the pressure fluctuates significantly within a short period of time, it may indicate that the driver is leaning forward or moving left or right, which is related to stressful driving.

[0088] S22322. Pressure distribution analysis: By analyzing the pressure distribution in different areas of the seat, if the pressure in the front of the seat increases and the pressure in the back decreases, it means that the driver's body is leaning forward, and the analysis shows that he is accelerating or focusing on the road conditions. If the pressure on both sides of the seat is uneven, it means that the driver's body is leaning to one side, and the analysis shows that it is related to the steering operation.

[0089] The step S3 includes the following specific steps:

[0090] S31. Parameter Fuzzification: Fuzzify parameters related to road conditions and driving behavior. For example, for traffic conditions, define three fuzzy sets: "congested," "normal," and "smooth." For driving aggressiveness, define three fuzzy sets: "conservative," "moderate," and "aggressive." Determine the degree of membership in each fuzzy set based on actual data such as average vehicle speed, acceleration rate, and seat pressure change. For example, if the average vehicle speed is 20 km / h, its membership in the traffic condition fuzzy set might be 0.8 for "congested," 0.2 for "normal," and 0 for "smooth."

[0091] S32. Establish a fuzzy rule base: Based on experience and understanding of the characteristics of different driving modes, establish a fuzzy rule base, such as "If traffic conditions are smooth and driving behavior is aggressive, then the driving mode is sport driving mode"; "If traffic conditions are congested and driving behavior is conservative, then the driving mode is economy mode or comfort driving mode."

[0092] S33. Fuzzy reasoning: Fuzzy outputs of driving modes are derived based on reasoning based on fuzzy rules. For example, by matching and reasoning fuzzy inputs of traffic conditions and driving behavior, the membership of sport mode is 0.6, that of economy mode is 0.3, and that of comfort mode is 0.1.

[0093] S34. Defuzzification: Use a defuzzification method such as the centroid method to convert the fuzzy output into a specific driving mode. For example, based on the aforementioned membership degree, the driving mode is ultimately determined to be the sport mode.

[0094] The step S4 includes the following contents:

[0095] The microcontroller changes the motor control method by adjusting the PWM signal sent to the permanent magnet synchronous motor driver; in economy mode, the motor speed and torque output are reduced to optimize efficiency; in sports mode, the motor speed and torque output are increased to provide stronger power; in comfort mode, the motor speed and torque are maintained at a moderate level to reduce vibration and noise.

[0096] The step S5 includes the following contents:

[0097] During the operation of the permanent magnet synchronous motor, the motor's temperature, current, and voltage parameters are monitored in real time. When the motor temperature rises, the motor's power output is reduced and the current reference value in the field-oriented control (FOC) algorithm is adjusted to prevent the motor from overheating. At the same time, the motor's control parameters are adaptively adjusted according to the vehicle's starting, acceleration, constant speed, and deceleration stages to improve the motor's efficiency and performance and better adapt to driving mode switching.

Claims

1. A method for controlling a permanent magnet synchronous motor to achieve driving mode switching, characterized in that: The following steps are involved: S1: Sensor Data Acquisition After the vehicle is started, various sensors begin to work, and the microcontroller regularly collects image data from the camera, acceleration and angle data from the inertial measurement unit, and data from the seat pressure sensor; S2: Data processing and analysis The microcontroller processes and analyzes the collected data; S3: Driving mode decision According to the processed data, a driving mode switching decision mechanism based on fuzzy rules is used to make decisions on the three driving modes: sport, economy, and comfort. S4: Motor control strategy adjustment The microcontroller adjusts the motor control strategy according to the selected driving mode; The specific contents of step S2 include: S21. Data transmission and preprocessing: Data from each sensor is transmitted to the microcontroller. The ADC and related signal processing units within the microcontroller preprocess the data, converting the analog signal into a digital signal, and performing filtering and amplification to obtain accurate sensor data. S22. Data Processing and Analysis: Camera image data is processed using a CNN algorithm to identify road type. Acceleration and angle data from the inertial measurement unit are used to calculate the vehicle's acceleration rate dynamic parameters. Seat pressure sensor data is analyzed for pressure variation. Driving behavior analysis is performed using acceleration rate dynamic parameters and seat pressure variation. The data processing in step S22 is achieved by constructing a layered sensor fusion architecture; at the bottom layer, the data from the camera, inertial measurement unit and seat pressure sensor are time synchronized and spatially calibrated; timestamp technology is used to ensure that the data from different sensors are in the same time reference system, and the spatial data of different sensors are unified into the vehicle coordinate system through coordinate transformation; at the middle layer, Kalman filtering or particle filtering data fusion algorithm is used to fuse the sensor data processed at the bottom layer; at the top layer, a comprehensive judgment of driving intention and road conditions is made based on the fused data.

2. The method for controlling a permanent magnet synchronous motor to switch driving modes according to claim 1, characterized in that: In step S1, before starting the vehicle, the following preparations are performed: The microcontroller is connected to the permanent magnet synchronous motor through a motor driver, and then a camera is installed on the car to identify road conditions. The inertial measurement unit is installed on the vehicle chassis to sense the dynamic characteristics of the vehicle, and the pressure sensor is installed on the seat to sense changes in the driver's sitting posture.

3. The method for controlling a permanent magnet synchronous motor to switch driving modes according to claim 1, characterized in that: In step S22, data analysis includes the following specific steps: S221. Road type identification S2211. Data processing: First, the road image captured by the camera is processed, including image cropping, normalization, and resizing; S2212. Build the network architecture: Use the VGG-Net architecture, which consists of multiple consecutive convolutional and pooling layers, and finally a fully connected layer to map the extracted features to different road type categories. S2213. Data training: Training is performed using labeled road image data, which includes highway image data, urban road image data, and mountain road image data. During training, the network weights are adjusted using a backpropagation algorithm to minimize the error between the network's output road type predictions and the true labels. S222. Traffic Condition Identification S2221. Use optical flow to estimate vehicle speed and indirectly determine traffic conditions. Specifically, by calculating the optical flow between adjacent frames, the speed of the vehicle relative to the surrounding environment is estimated. If the speed of most vehicles is low and relatively stable, it indicates traffic congestion; if the speed of the vehicles is high and varies greatly, it indicates smooth traffic. S2222. Use an object detection algorithm to identify vehicles and traffic signs to assist in determining traffic conditions. Specifically, the image is divided into multiple grid cells, with each cell responsible for predicting a certain number of objects. The prediction results for each cell are post-processed to accurately determine the location and category of vehicles and traffic signs. If a large number of vehicles are detected waiting in line or stop signs are frequently seen, traffic congestion is indicated. If vehicles are sparsely distributed and traffic signs primarily indicate high speed limits, traffic is smooth. S223. Driving behavior analysis S2231. Driving Behavior Analysis Based on Inertial Measurement Unit (IMU) Data S22311. Acceleration Data Processing: Acceleration data collected by the Inertial Measurement Unit (IMU) includes acceleration information about the vehicle's X, Y, and Z axes. By integrating this acceleration data, we obtain vehicle velocity and displacement information. This acceleration data is filtered to remove noise interference and obtain more accurate vehicle dynamic information. S22312. Angular velocity data processing: The inertial measurement unit (IMU) collects the vehicle's angular velocity information to determine the vehicle's steering status. A sustained non-zero angular velocity indicates the vehicle is turning. The magnitude and duration of the angular velocity are analyzed to determine the severity of the turn. A high angular velocity that persists for a period of time indicates a sharp turn, while a low angular velocity indicates a gentle turn or a slight directional adjustment. S2232. Driving behavior analysis based on seat pressure sensor data S22321. Pressure Change Monitoring: The pressure data collected by the seat pressure sensor reflects changes in the driver's sitting posture. By performing differential calculations on the pressure data, the magnitude and frequency of pressure changes are monitored. If the pressure fluctuates significantly within a short period of time, it indicates that the driver is leaning forward or moving left or right, which can be related to stressful driving. S22322. Pressure distribution analysis: By analyzing the pressure distribution in different areas of the seat, if the pressure in the front of the seat increases and the pressure in the back decreases, it means that the driver's body is leaning forward, and the analysis shows that he is accelerating or focusing on the road conditions. If the pressure on both sides of the seat is uneven, it means that the driver's body is leaning to one side, and the analysis shows that it is related to the steering operation.

4. The method for controlling a permanent magnet synchronous motor to switch driving modes according to claim 3, characterized in that: The step S3 includes the following specific steps: S31. Parameter fuzzification: Fuzzify the parameters related to road conditions and driving behavior; determine their membership in various fuzzy sets based on actual data; S32. Establish a fuzzy rule base: Establish a fuzzy rule base based on experience and understanding of the characteristics of different driving modes; S33. Fuzzy reasoning: Reasoning based on fuzzy rules to obtain fuzzy output of driving mode; S34. Defuzzification: Use defuzzification methods to convert fuzzy outputs into definite driving modes.

5. The method for controlling a permanent magnet synchronous motor to switch driving modes according to claim 4, characterized in that: The fuzzy rule base in step S32 is: if the traffic condition is smooth and the driving behavior is aggressive, the driving mode is the sports driving mode; if the traffic condition is congested and the driving behavior is conservative, the driving mode is the economic mode or the comfortable driving mode.

6. The method for controlling a permanent magnet synchronous motor to switch driving modes according to claim 1, characterized in that: The step S4 includes the following contents: The microcontroller changes the motor control method by adjusting the PWM signal sent to the permanent magnet synchronous motor driver. In the economy mode, the motor speed and torque output are reduced to optimize efficiency. In sport mode, the motor's speed and torque output are increased to provide stronger power; In comfort mode, the motor speed and torque are maintained at moderate levels to reduce vibration and noise.

7. The method for controlling a permanent magnet synchronous motor to switch driving modes according to claim 1, characterized in that: The method further includes step S5. Adaptive motor control parameter adjustment: adaptively adjusting the motor control parameters based on real-time feedback of the motor operating state.

8. The method for controlling a permanent magnet synchronous motor to switch driving modes according to claim 7, characterized in that: The step S5 includes the following contents: During motor operation, the motor's temperature, current, and voltage parameters are monitored in real time. When the motor temperature rises, the motor's power output is reduced and the current reference value in the magnetic field-oriented control algorithm is adjusted to prevent the motor from overheating. At the same time, the motor's control parameters are adaptively adjusted according to the vehicle's starting, acceleration, constant speed, and deceleration stages to improve the motor's efficiency and performance and better adapt to driving mode switching.