A steering system torque calculation method based on double-loop PID algorithm

Abstract

This invention relates to the field of steering system technology, specifically a steering system torque calculation method based on a dual-loop PID algorithm. The method includes a dual-loop PID algorithm comprising a front loop and a rear loop. The front loop calculates the angle difference by subtracting the requested angle from the actual angle, and the angle control system calculates the target angular velocity based on this angle difference. The rear loop calculates the angular velocity difference by subtracting the target angular velocity from the actual angular velocity, and the angular velocity difference is processed by the PID algorithm to obtain the target motor torque. This target motor torque is then converted into the actual executed motor torque through gradient output to control the steering wheel rotation. The steering system torque calculation method includes driver takeover optimization strategies and ADAS takeover optimization strategies. Compared with existing technologies, this method uses a dual-loop PID algorithm to calculate motor torque and control steering wheel movement to achieve the target angle.

Description

Technical Field

[0001] This invention relates to the field of steering system technology, specifically a steering system torque calculation method based on a dual-loop PID algorithm. Background Technology

[0002] Currently, autonomous driving technology is widely used in the automotive field. Lateral control in autonomous driving primarily relies on steering, and the steering system often uses an electronic power steering (ESP) as its platform to design an angle control system, responding to angle requests from the autonomous driving system to complete steering. From a stability perspective, the autonomous driving system does not want angle control functions to be easily disrupted by external interference. Vehicle systems typically require angle control to disengage only when both the driver's torque and duration reach a set threshold. Furthermore, the ESP should not respond to driver input when the torque is low, ensuring that human intervention does not easily affect the vehicle's path planning. Because the driver's state during takeover cannot be predicted (there may be sudden steering wheel movements or gentle tugging), when both the torque and duration reach the set threshold, the basic power steering torque and angle control torque cannot coordinate effectively, leading to abrupt steering feel issues. When ADAS takes over steering, performance parameters such as response time are prioritized during calibration. Therefore, in the initial stage of ADAS takeover, the steering wheel speed increases rapidly. When the driver's hands are still lightly touching the steering wheel, this can cause the driver to perceive a deviation from their expected steering rotation trend, leading to complaints. Summary of the Invention

[0003] To overcome the shortcomings of the prior art, this invention provides a method for calculating steering system torque based on a dual-loop PID algorithm. The dual-loop PID algorithm is used to calculate the motor torque and control the steering wheel movement to achieve the target angle.

[0004] To achieve the above objectives, a steering system torque calculation method based on a dual-loop PID algorithm is designed. The dual-loop PID algorithm includes a front loop and a rear loop. The front loop calculates the angle difference by subtracting the requested angle from the actual angle, and the angle control system calculates the target angular velocity based on this angle difference. The rear loop calculates the angular velocity difference by subtracting the target angular velocity from the actual angular velocity, and the angular velocity difference is processed by the PID algorithm to obtain the target motor torque. The target motor torque is then converted into the actual executed motor torque through gradient output to control the steering wheel rotation. The method is characterized by including a driver takeover optimization strategy and an ADAS takeover optimization strategy.

[0005] The driver takeover optimization strategy is as follows:

[0006] S11, set the vehicle speed range to 0-5kph, to calculate the different attenuation coefficients for stationary steering and dynamic steering;

[0007] S12, set the hand torque threshold, hand torque time threshold and hand torque integral threshold, and each threshold value can be calibrated according to the vehicle speed;

[0008] S13, Calculate the P-phase attenuation coefficient σ based on the integral value. p The attenuation coefficient σ of the I-phase Delta term iΔ The integral value is the cumulative value of the portion exceeding the hand torque threshold over time.

[0009] S14, based on the locked and stored hand force value TBT when the hand force threshold and hand force time threshold are exceeded. a Calculate the attenuation coefficient σ of the Delay term in the I term. id The aforementioned hand force value TBT a The stored value is the hand torque value that simultaneously exceeds both the hand force threshold and the hand torque time threshold.

[0010] S15, based on the locked and stored hand force value TBT exceeding the hand torque integration threshold. b Calculate the hand torque compensation coefficient σ t The aforementioned hand force value TBT b The hand torque when the integral value in step S13 exceeds the integral threshold;

[0011] S16 calculates the angle control torque and compensated hand torque, and outputs them to the motor.

[0012] The ADAS takeover optimization strategy is as follows:

[0013] S21, Detection indicates that driver takeover mode has been exited;

[0014] S22, calculate the magnitude of angular acceleration Δa by referring to a table based on the target rotational speed;

[0015] S23, gain coefficient φ1 in the vehicle speed lookup table;

[0016] S24, calculate the processed target turning angle.

[0017] The formula for calculating the target motor torque is as follows: Where TarMotTorq is the target motor torque, TarAngSpd is the target steering wheel angular velocity, CurAngSpd is the actual steering wheel angular velocity, Kp is the proportional coefficient, Ki is the integral coefficient, and Kd is the derivative coefficient.

[0018] In step S3, the P-phase attenuation coefficient σ p The calculation formula is σ p (k)-σ p(k-1)≤f PL [v(k)], where I d (k) indicates whether the current driver has taken over; 1 indicates takeover, meaning the driver's hand torque simultaneously meets the hand torque threshold, hand torque time threshold, and integral threshold; 0 indicates no takeover. d (k) represents the current integral value of the hand force; f Pf The integral value of hand force and σ p The correspondence is fitted through testing; f PL For vehicle speed and σ p The correspondence of gradients is fitted through testing.

[0019] In step S3, the attenuation coefficient σ of the Delta term in phase I. iΔ The calculation formula is σ iΔ (k)-σ iΔ (k-1)≤f ΔIL [v(k)], where f ΔIf The integral value of hand force and σ iΔ The correspondence is fitted through testing; f ΔIL For vehicle speed and σ iΔ The correspondence of gradients is fitted through testing.

[0020] 1. The method for calculating steering system torque based on a dual-loop PID algorithm according to claim 1, characterized in that: in step S4, the attenuation coefficient σ of the I-term Delay term... id The calculation formula is σ id (k)-σ id (k-1)≤f DIL [v(k),TBT a ,I Dir (k)], where I Dir The current takeover status is categorized as follows: 0 represents in-situ same direction, 1 represents in-situ opposite direction, 2 represents dynamic median, and 3 represents dynamic end point. Fitting is performed through testing. DIL For vehicle speed, TBT a I Dir With σ iΔ The correspondence of gradients is fitted through testing.

[0021] In step S5, the hand torque compensation coefficient σ t The calculation formula is in, For TBT b I Dir With σt The correspondence of gradients is fitted through testing.

[0022] In step S6, the formula for calculating the angle control torque is as follows: Where TarMotTorq is the target motor torque, σ p Where P is the phase attenuation coefficient, Kp is the proportional coefficient, TarAngSpd is the target angular velocity of the steering wheel, CurAngSpd is the actual angular velocity of the steering wheel, and σ is the phase attenuation coefficient. iΔ The attenuation coefficient for the Delta term of phase I, Ki is the integral coefficient, and σ is the attenuation coefficient for phase I. id This is the attenuation coefficient for the Delay term in section I.

[0023] In step S6, the formula for calculating the compensating hand torque is as follows: Where TarTBT is the compensated hand torque, TBT is the current hand torque, and σ t This is the hand torque compensation coefficient.

[0024] The driver takeover situation is as follows:

[0025] (1) When the driver takes over the direction for the first time in the negative direction, and the ADAS takes over again and returns to 0°, the tire has a certain negative angle due to the ground friction. The torque direction of the angle control is to overcome this negative angle deformation in the positive direction. This torque is the output of the I-term Delay term.

[0026] (2) When the driver takes over the direction for the second time in the negative direction, the torque calculated by the P phase is positive when the angle begins to change, the torque calculated by the Delta term of the I phase is positive, and the basic assist generated by the hand force is negative. The system behaves as the angle control torque prevents the steering wheel angle from changing.

[0027] (3) When the driver takes over the direction for the second time, the torque calculated by the P phase is in the negative direction when the angle begins to change, the torque calculated by the Delta term of the I phase is in the negative direction, and the basic assist generated by the hand force is in the positive direction. The system behaves as the angle control torque prevents the steering wheel angle from changing.

[0028] Since tire deformation will cause the steering wheel to turn in the negative direction, and the angle control torque needs to be completely set to 0 Nm after the driver takes full control, when the second takeover direction is the same as the first takeover direction (both in the negative direction), the Delta terms of the P and I phases need to be set to 0 at the beginning of the takeover so that the torque generated by the basic power assist can drive the steering wheel to turn. At the same time, this avoids the sudden change in feel caused by the increase in angle control torque and setting it to 0 after full takeover. When the second takeover direction is opposite to the first takeover direction, the Delta terms of the P and I phases need to help reduce the angle control torque at the beginning of the takeover. After full takeover, the Delta terms of the P and I phases are set to 0, and the Delay term of the I phase is attenuated to 0 Nm according to a certain attenuation coefficient.

[0029] In step S24, the formula for calculating the target rotation angle is as follows: Where TarStrAng is the target turning angle, and StrAng is the current steering wheel angle. Δa is the gain coefficient, TarStrAngSpd is the target steering wheel rotation speed, Δa is the angular acceleration, and Δt is the sampling time.

[0030] Compared with the prior art, the present invention provides a method for calculating the torque of a steering system based on a dual-loop PID algorithm. The dual-loop PID algorithm is used to calculate the motor torque and control the steering wheel movement to achieve the target angle. Attached Figure Description

[0031] Figure 1 This is a flowchart of the present invention.

[0032] Figure 2 This is a schematic diagram of an embodiment of the present invention. Detailed Implementation

[0033] The present invention will now be further described with reference to the accompanying drawings.

[0034] The angle control system involved in this invention uses a dual-loop PID algorithm to calculate motor torque. The first loop calculates the angle difference by subtracting the requested angle from the actual angle, and the angle control system calculates the target angular velocity based on the angle difference. The second loop calculates the angular velocity difference by subtracting the target angular velocity from the actual angular velocity, and the angular velocity difference is used by the PID algorithm to obtain the target motor torque.

[0035] Taking the gentle pull of the steering wheel to take over as an example, the entire takeover process is divided into the following 4 stages:

[0036] Phase 1: When the driver applies hand force to the steering wheel with a low torque gradient, before the hand force exceeds the preset torque threshold and time threshold, as the hand force increases, the angle control torque also increases to counteract the change in steering wheel angle caused by the increase in hand force. At this time, the steering wheel does not show any change in angle.

[0037] Phase 2: Before the angle control torque decays to 0 Nm, the hand force exceeds the preset torque and time thresholds. The moment the hand force exceeds the threshold, the angle control torque begins to decay, and the base assist torque increases according to the current hand force, causing the steering wheel to begin turning. At this point, the changes in the base assist torque and angle control torque directly affect the current steering feel.

[0038] Phase 3: The angle control torque decays to 0, and before the hand torque reaches the torque and time thresholds for ADAS takeover, it is in a completely basic power assist state;

[0039] Stage 4: When the hand torque exceeds the torque and time thresholds of ADAS takeover, the steering wheel will quickly respond to the target steering wheel speed and the target steering wheel angle output by ADAS.

[0040] The four stages mentioned above are similar when the driver takes over the steering wheel suddenly. Stages 1, 2, and 4 will cause changes in the feel and affect the driving experience. Based on this, this paper proposes a control strategy that allows the driver or ADAS to take over the angle control and adjust the feel to solve the above problems.

[0041] This invention uses a dual-loop PID algorithm to calculate motor torque and control the steering wheel movement to achieve the target angle.

[0042] The specific implementation method is as follows: the front loop part calculates the angle difference by subtracting the requested angle from the actual angle, and the angle control system calculates the target angular velocity based on the angle difference; the rear loop part calculates the angular velocity difference by subtracting the target angular velocity from the actual angular velocity, and the angular velocity difference is used to obtain the target motor torque through a PID algorithm. The target motor torque is then converted into the actual motor torque through gradient output to control the steering wheel rotation.

[0043] The parameters involved in the PID algorithm are calculated based on the vehicle speed, and the P, I, and D terms all have motor torque limits at different vehicle speeds to prevent the risk of vehicle instability caused by excessive motor torque saturation and rapid steering wheel rotation at high speeds.

[0044] The formula for calculating the target motor torque is as follows: Where TarMotTorq is the target motor torque, TarAngSpd is the target steering wheel angular velocity, CurAngSpd is the actual steering wheel angular velocity, Kp is the proportional coefficient, Ki is the integral coefficient, and Kd is the derivative coefficient.

[0045] The existing electronic power steering system calculates the assist torque based on vehicle speed, hand torque, etc., which is called the basic assist mode. The basic assist mode controls the motor to achieve the steering requirements.

[0046] The angle control function involved in this invention can decouple the basic power assist mode from the angle control mode. When the angle control function is activated, if the magnitude and duration of the hand torque do not meet the requirements, the steering will not be changed in response to the hand torque value.

[0047] The adjustable-feel control strategy involved in this invention consists of two parts: a driver takeover optimization strategy and an ADAS takeover optimization strategy. The purpose of the driver takeover optimization strategy is to reduce the angle control torque in advance and to calculate different angle control torque attenuation coefficients and hand torque compensation coefficients based on the magnitude of the hand torque.

[0048] The specific process of the driver takeover optimization strategy is as follows:

[0049] (1) Set the vehicle speed range to 0-5kph to calculate the different attenuation coefficients for stationary steering and dynamic steering;

[0050] (2) Set the hand torque threshold, hand torque time threshold and hand torque integral threshold. Each threshold value can be calibrated according to the vehicle speed.

[0051] (3) Calculate the P-phase attenuation coefficient σ based on the integral value. p The attenuation coefficient σ of the I-phase Delta term iΔ The integral value is the cumulative value of the portion exceeding the hand torque threshold over time.

[0052] (4) Based on the locked and stored hand force value TBT when the hand force threshold and hand force time threshold are exceeded. a Calculate the attenuation coefficient σ of the Delay term in the I term. id The aforementioned hand force value TBT a The stored value is the hand torque value that simultaneously exceeds both the hand force threshold and the hand torque time threshold.

[0053] (5) Based on the locked and stored hand force value TBT exceeding the hand torque integral threshold. b Calculate the hand torque compensation coefficient σ t The aforementioned hand force value TBT b The hand torque when the integral value in step S13 exceeds the integral threshold;

[0054] (6) Calculate the angle control torque and compensation hand torque and output them to the motor end.

[0055] P-phase attenuation coefficient σ p The calculation formula is σ p (k)-σ p (k-1)≤f PL [v(k)], where I d(k) indicates whether the current driver has taken over; 1 indicates takeover, meaning the driver's hand torque simultaneously meets the hand torque threshold, hand torque time threshold, and integral threshold; 0 indicates no takeover. d (k) represents the current integral value of the hand force; f Pf The integral value of hand force and σ p The correspondence is fitted through testing; f PL For vehicle speed and σ p The correspondence of gradients is fitted through testing.

[0056] Phase I Delta term attenuation coefficient σ iΔ The calculation formula is σ iΔ (k)-σ iΔ (k-1)≤f ΔIL [v(k)], where f ΔIf The integral value of hand force and σ iΔ The correspondence is fitted through testing; f ΔIL For vehicle speed and σ iΔ The correspondence of gradients is fitted through testing.

[0057] The attenuation coefficient σ of the delay term in item I id The calculation formula is σ id (k)-σ id (k-1)≤f DIL [v(k),TBT a ,I Dir (k)], where I Dir The current takeover status is categorized as follows: 0 represents in-situ same direction, 1 represents in-situ opposite direction, 2 represents dynamic median, and 3 represents dynamic end point. Fitting is performed through testing. DIL For vehicle speed, TBT a I Dir With σ iΔ The correspondence of gradients is fitted through testing.

[0058] Hand torque compensation coefficient σ t The calculation formula is in, For TBT b I Dir With σ t The correspondence of gradients is fitted through testing.

[0059] The formula for calculating the angle control torque is as follows: Where TarMotTorq is the target motor torque, σ pWhere P is the phase attenuation coefficient, Kp is the proportional coefficient, TarAngSpd is the target angular velocity of the steering wheel, CurAngSpd is the actual angular velocity of the steering wheel, and σ is the phase attenuation coefficient. iΔ The attenuation coefficient for the Delta term of phase I, Ki is the integral coefficient, and σ is the attenuation coefficient for phase I. id This is the attenuation coefficient for the Delay term in section I.

[0060] The formula for calculating the compensating hand torque is as follows: Where TarTBT is the compensated hand torque, TBT is the current hand torque, and σ t This is the hand torque compensation coefficient.

[0061] To further clarify, under stationary operating conditions, the consistency of the driver's two take-over directions significantly impacts the driving feel. Taking an ADAS-requested angle of 0° as an example, when the driver's first take-over direction is negative, and the ADAS reverts to 0°, the tires will have a certain negative angle due to ground friction. The torque for angle control will be directed to overcome this negative angle deformation in the positive direction, and this torque will be output by the I-term Delay.

[0062] If the driver takes over in the negative direction for the second time, the torque calculated by the P phase will be in the positive direction when the angle begins to change, the torque calculated by the Delta term of the I phase will be in the positive direction, and the basic assist generated by the hand force will be in the negative direction. The system will behave as an angle control torque to prevent the steering wheel angle from changing.

[0063] If the driver takes over the steering wheel for the second time in a positive direction, then when the angle begins to change, the torque calculated by the P phase is in a negative direction, the torque calculated by the Delta term of the I phase is in a negative direction, and the basic assist generated by the hand force is in a positive direction. The system behaves as an angle control torque that prevents the steering wheel angle from changing.

[0064] Because tire deformation causes the steering wheel to turn in the negative direction, and the angle control torque needs to be completely set to 0 Nm after the driver takes full control, when the second takeover direction is the same as the first (negative direction), the Delta terms of the P and I phases need to be set to 0 initially to allow the torque generated by the basic power steering to drive the steering wheel, while avoiding a sudden change in feel due to an increase in angle control torque that would be set to 0 after full takeover. When the second takeover direction is opposite to the first, the Delta terms of the P and I phases are needed to reduce the angle control torque initially. After full takeover, the Delta terms of the P and I phases are set to 0, and the Delay term of the I phase is attenuated to 0 Nm according to a certain attenuation coefficient.

[0065] Under driving conditions, similar to stationary conditions, tire deformation becomes the vehicle's restoring torque. Therefore, the connection to the center position is the same as the connection to the stationary position in the same direction, and the connection to the end position is the same as the connection to the stationary position in the opposite direction.

[0066] Therefore, the P-phase attenuation coefficient σ is calculated based on the integral value. p The attenuation coefficient σ of the I-phase Delta term iΔ The value needs to be set to 0 when the hand force exceeds the torque threshold and time threshold, and when the hand force exceeds the torque threshold and time threshold in the stationary working condition, in the mid-position working condition, or when the hand force exceeds the torque threshold and time threshold.

[0067] It should be further explained that, in order to avoid sudden changes in steering wheel angle caused by tire deformation while stationary and dynamic self-centering, the attenuation coefficient σ of the Delay term in section I is... id The required hand force value (TBT) is determined by the hand force threshold and time threshold. a Perform correlation calculations to apply to different driver takeover states.

[0068] The ADAS takeover optimization strategy aims to process the target angle curve and improve the abrupt steering feel caused by the rapid increase in steering wheel starting angular velocity when responding to high-revving target angle requests. Since the target angle rises rapidly during the period of responding to high-revving angle requests, the second derivative of the target angle is processed in the initial stage of the request.

[0069] The specific process for ADAS takeover optimization is as follows:

[0070] (1) The driver takeover mode has been exited.

[0071] (2) Calculate the magnitude of angular acceleration Δa by referring to a table based on the target rotation speed;

[0072] (3) Gain coefficient φ1 of the vehicle speed lookup table;

[0073] (4) Calculate the target turning angle after processing.

[0074] The formula for calculating the target turning angle is: Where TarStrAng is the target turning angle, and StrAng is the current steering wheel angle. Δa is the gain coefficient, TarStrAngSpd is the target steering wheel rotation speed, Δa is the angular acceleration, and Δt is the sampling time.

[0075] like Figure 2 As shown in the actual test, at 55s, TBT increases, indicating that the driver is applying hand torque to take over. When TBT meets the takeover requirements, the takeover state is set from 0 to 1, and the actual steering wheel angle StrAng deviates from the target angle TarStrAng in response to driver intervention. When TBT is detected to weaken at 59s and meets the ADAS takeover requirements, the takeover state is set from 1 to 0, and StrAng is executed to TarStrAng, realizing ADAS takeover.

Claims

1. A method for calculating steering system torque based on a dual-loop PID algorithm, comprising a dual-loop PID algorithm, wherein the dual-loop PID algorithm includes a front loop and a rear loop; the front loop: the angle difference is obtained by subtracting the requested angle from the actual angle, and the angle control system calculates the target angular velocity based on the angle difference; the rear loop: the angular velocity difference is obtained by subtracting the target angular velocity from the actual angular velocity, the angular velocity difference is used by the PID algorithm to obtain the target motor torque, and the target motor torque is converted into the actual executed motor torque through gradient output to control the steering wheel rotation, characterized in that: The steering system torque calculation method includes driver takeover optimization strategy and ADAS takeover optimization strategy. The driver takeover optimization strategy is as follows: S11, set the vehicle speed range to 0-5kph, to calculate the different attenuation coefficients for stationary steering and dynamic steering; S12, set the hand torque threshold, hand torque time threshold and hand torque integral threshold, and each threshold value can be calibrated according to the vehicle speed; S13, Calculate the P-phase attenuation coefficient σ based on the integral value. p The attenuation coefficient σ of the I-phase Delta term iΔ The integral value is the cumulative value of the portion exceeding the hand torque threshold over time. S14, based on the locked and stored hand force value TBT when the hand force threshold and hand force time threshold are exceeded. a Calculate the attenuation coefficient σ of the Delay term in the I term. id The aforementioned hand force value TBT a The stored value is the hand torque value that simultaneously exceeds both the hand force threshold and the hand torque time threshold. S15, based on the locked and stored hand force value TBT exceeding the hand torque integration threshold. b Calculate the hand torque compensation coefficient σ t The aforementioned hand force value TBT b The hand torque when the integral value in step S13 exceeds the integral threshold; S16 calculates the angle control torque and compensated hand torque, and outputs them to the motor. The ADAS takeover optimization strategy is as follows: S21, Detection indicates that driver takeover mode has been exited; S22, calculate the magnitude of angular acceleration Δa by referring to a table based on the target rotational speed; S23, gain coefficient φ1 in the vehicle speed lookup table; S24, calculate the processed target turning angle.

2. The method for calculating steering system torque based on a dual-loop PID algorithm according to claim 1, characterized in that: The formula for calculating the target motor torque is as follows: Where TarMotTorq is the target motor torque, TarAngSpd is the target steering wheel angular velocity, CurAngSpd is the actual steering wheel angular velocity, Kp is the proportional coefficient, Ki is the integral coefficient, and Kd is the derivative coefficient.

3. The method for calculating steering system torque based on a dual-loop PID algorithm according to claim 1, characterized in that: In step S13, the P-phase attenuation coefficient σ p The calculation formula is σ p (k)-σ p (k-1)≤f PL [v(k)], where I d (k) indicates whether the current driver has taken over; 1 indicates takeover, meaning the driver's hand torque simultaneously meets the hand torque threshold, hand torque time threshold, and integral threshold; 0 indicates no takeover. d (k is the current hand strength integral value; f) Pf The integral value of hand force and σ p The correspondence is fitted through testing; f PL For vehicle speed and σ p The correspondence of gradients is fitted through testing.

4. The method for calculating steering system torque based on a dual-loop PID algorithm according to claim 1, characterized in that: In step S13, the attenuation coefficient σ of the I-phase Delta term iΔ The calculation formula is σ iΔ (k)-σ iΔ (k-1)≤f ΔIL [v(k)], where f ΔIf The integral value of hand force and σ iΔ The correspondence is fitted through testing; f ΔIL For vehicle speed and σ iΔ The correspondence of gradients is fitted through testing.

5. The method for calculating steering system torque based on a dual-loop PID algorithm according to claim 1, characterized in that: In step S14, the attenuation coefficient σ of the I-term Delay term id The calculation formula is σ id (k)-σ id (k-1)≤f DIL [v(k),TBT a ,I Dir (k)], where I Dir The current takeover status is categorized as follows: 0 represents in-situ same direction, 1 represents in-situ opposite direction, 2 represents dynamic median, and 3 represents dynamic end point. Fitting is performed through testing. DIL For vehicle speed, TBT a I Dir With σ iΔ The correspondence of gradients is fitted through testing.

6. The method for calculating steering system torque based on a dual-loop PID algorithm according to claim 1, characterized in that: In step S15, the hand torque compensation coefficient σ t The calculation formula is in, For TBT b I Dir With σ t The correspondence of gradients is fitted through testing.

7. The method for calculating steering system torque based on a dual-loop PID algorithm according to claim 1, characterized in that: In step S16, the formula for calculating the angle control torque is as follows: Where TarMotTorq is the target motor torque, σ p Where P is the phase attenuation coefficient, Kp is the proportional coefficient, TarAngSpd is the target angular velocity of the steering wheel, CurAngSpd is the actual angular velocity of the steering wheel, and σ is the phase attenuation coefficient. iΔ The attenuation coefficient for the Delta term of phase I, Ki is the integral coefficient, and σ is the attenuation coefficient for phase I. id This is the attenuation coefficient for the Delay term in section I.

8. The method for calculating steering system torque based on a dual-loop PID algorithm according to claim 1, characterized in that: In step S16, the formula for calculating the compensating hand torque is as follows: Where TarTBT is the compensated hand torque, TBT is the current hand torque, and σ t This is the hand torque compensation coefficient.

9. The method for calculating steering system torque based on a dual-loop PID algorithm according to claim 3, characterized in that: The driver takeover situation is as follows: (1) When the driver takes over the direction for the first time in the negative direction, and the ADAS takes over again and returns to 0°, the tire has a certain negative angle due to the ground friction. The torque direction of the angle control is to overcome this negative angle deformation in the positive direction. This torque is the output of the I-term Delay term. (2) When the driver takes over the direction for the second time in the negative direction, the torque calculated by the P phase is positive when the angle begins to change, the torque calculated by the Delta term of the I phase is positive, and the basic assist generated by the hand force is negative. The system behaves as the angle control torque prevents the steering wheel angle from changing. (3) When the driver takes over the direction for the second time, the torque calculated by the P phase is in the negative direction when the angle begins to change, the torque calculated by the Delta term of the I phase is in the negative direction, and the basic assist generated by the hand force is in the positive direction. The system behaves as the angle control torque prevents the steering wheel angle from changing. Since tire deformation will cause the steering wheel to turn in the negative direction, and the angle control torque needs to be set to 0 Nm after the human hand is fully in control, when the second handover direction is the same as the first handover direction, which is also in the negative direction, the Delta term of the P phase and I phase needs to be set to 0 at the beginning of the handover so that the torque generated by the basic power assist can drive the steering wheel to turn. At the same time, it avoids the sudden change in feel caused by the increase of the angle control torque and setting it to 0 after the handover is fully in control. When the direction of the second takeover is opposite to that of the first takeover, in the initial stage of takeover, the Delta terms of the P phase and I phase are needed to help reduce the angle control torque. After the takeover is complete, the Delta terms of the P phase and I phase are set to 0, and the Delay term of the I phase is attenuated to 0 Nm according to a certain attenuation coefficient.

10. The method for calculating steering system torque based on a dual-loop PID algorithm according to claim 1, characterized in that: In step S24, the formula for calculating the target rotation angle is as follows: Where TarStrAng is the target turning angle, and StrAng is the current steering wheel angle. Δa is the gain coefficient, TarStrAngSpd is the target steering wheel rotation speed, Δa is the angular acceleration, and Δt is the sampling time.

Images