Electric tail gate anti-pinch control method and system

Abstract

The invention relates to the technical field of automobile safety, and provides an electric tail gate anti-pinch control method and system, which introduces a prediction + self-adaption intelligent decision normal form, on one hand, the motion information of a tail gate is obtained, and the time sequence data of an obstacle is combined to be input into an LSTM model to predict the track of the obstacle; a collision prediction conclusion is output to be matched with a graded speed reduction strategy to execute tail gate control, and through prediction in advance, the system response efficiency is effectively improved, and the clamping damage probability is reduced; and on the other hand, when a contact signal of the tail gate is detected, real-time contact force and real-time vehicle working condition information are obtained in real time, a dynamic anti-pinch threshold value is calculated and compared with the real-time contact force for analysis so as to execute tail gate control, the dynamic anti-pinch threshold value is dynamically adjusted according to the vehicle working condition and the tail gate state, scene adaptation is carried out, the anti-pinch recognition accuracy can be improved, and the safety of the vehicle is improved. The false triggering probability is reduced; therefore, the electric tail gate anti-pinch system which is faster in response, more intelligent and more reliable is obtained.

Description

Technical Field

[0001] This invention relates to the field of automotive safety technology, and in particular to an anti-pinch control method and system for electric tailgates. Background Technology

[0002] With the development of the automotive industry, the proportion of electric tailgates is increasing, expanding from high-end models to mid-to-low-end vehicles. While electric tailgates offer convenience, they also pose a risk of accidental contact with people during opening and closing, necessitating the development of anti-pinch functionality. Current electric tailgates trigger the anti-pinch mechanism when the applied pressure reaches a pre-calibrated threshold, which is generally calibrated based on regulations requiring a force of less than 100N for windows and sunroofs. Therefore, the following drawbacks exist: (1) Sensing lag: Mainstream solutions (such as Hall current detection and capacitor strip) require physical contact with obstacles to be triggered, and the system response delay is usually over 150ms, which poses a potential risk of pinching injury.

[0003] (2) Strategy rigidity: Using fixed anti-pinch force or current threshold, it cannot adapt to complex working conditions such as slope changes, hardening of sealing strips due to low temperature, and aging of parts, resulting in a high false trigger rate.

[0004] (3) Power failure: When the vehicle power supply is unstable or the power is cut off, the system fails and the tailgate cannot perform anti-pinch reverse rotation, which poses a safety hazard. Summary of the Invention

[0005] This invention provides a method and system for controlling the anti-pinch function of an electric tailgate, which solves the technical problems of existing electric tailgate anti-pinch technology, such as slow sensing, rigid triggering strategy, resulting in low response sensitivity, high false triggering rate, and potential failure due to power outage.

[0006] To solve the above technical problems, the present invention provides an anti-pinch control method for electric tailgates, comprising: Real-time acquisition of obstacle movement information yields temporal data of the obstacles; The motion information of the tailgate is obtained, and the time series data of the obstacle is combined with the LSTM model to predict the trajectory of the obstacle, and the collision prediction conclusion is output. Tailgate control is executed based on the collision prediction conclusions and a graded deceleration strategy. When a contact signal is detected from the tailgate, real-time contact force and real-time vehicle operating condition information are acquired, and a dynamic anti-pinch threshold is calculated based on the real-time vehicle operating condition information and the motion information. Tailgate control is performed based on the real-time contact force and the dynamic anti-pinch threshold.

[0007] This basic solution introduces a "prediction + adaptation" intelligent decision-making paradigm. On the one hand, it acquires the tailgate's motion information and combines it with the time-series data of obstacles, inputting it into an LSTM model to predict the obstacle's trajectory. It then outputs a collision prediction conclusion to match a graded deceleration strategy for tailgate control. By predicting in advance, it effectively improves system response efficiency and reduces the probability of pinching injuries. On the other hand, when a contact signal is detected from the tailgate, it acquires real-time contact force and real-time vehicle operating condition information, calculates a dynamic anti-pinch threshold, and compares it with the real-time contact force to execute tailgate control. By dynamically adjusting the dynamic anti-pinch threshold based on vehicle operating conditions and tailgate status, it adapts to different scenarios, improving the accuracy of anti-pinch recognition and reducing the probability of false triggering. This results in a faster, smarter, and more reliable electric tailgate anti-pinch system.

[0008] In a further implementation, real-time acquisition of obstacle movement information yields obstacle time-series data, including: A first sensor is installed inside the tailgate to acquire the movement information of obstacles in real time. The movement information is accumulated over a preset time step to obtain the temporal data of the obstacles. The movement information includes one or more of the following: distance, speed, azimuth angle, pitch angle, and signal-to-noise ratio.

[0009] This solution directly detects the movement information of obstacles and accumulates time-series data for trajectory prediction by the LSTM model. By analyzing the movement trajectory and trend, potential collision risks can be predicted in advance, allowing for lead time for tailgate control. This triggers more precise and smooth avoidance actions before clamping occurs, significantly improving the intelligence and reliability of the anti-pinch system.

[0010] In a further implementation, the collision prediction conclusion includes the collision probability, the predicted collision time, and the collision distance between the predicted object and the obstacle. Tailgate control is executed based on the collision prediction conclusion and the graded deceleration strategy. Specifically, the collision probability is extracted from the collision prediction conclusion, the risk level is determined based on the collision probability, and the corresponding risk response strategy in the graded deceleration strategy is matched. Tailgate control is then executed based on the risk response strategy.

[0011] This solution determines the risk level based on the collision probability and matches the corresponding risk response strategy in the graded deceleration strategy. By quantifying the collision probability and adapting the graded response, it decisively applies strong braking to high-risk situations, dynamically calculates the deceleration magnitude to control the deceleration for medium-risk situations based on the predicted collision time, and does not handle low-risk situations. In this way, while ensuring absolute safety, it minimizes unnecessary emergency stops and improves the smoothness of tailgate operation and user experience.

[0012] In a further implementation scheme, when the risk level is high risk, the risk response strategy includes: Enter high-risk alert status and increase the sensitivity of vehicle detection parameters; Calculate the static safety distance based on the motion information and system response performance; The collision distance is determined from the collision prediction conclusion. Based on the static safety distance and the collision distance, it is determined whether to perform an emergency stop. Otherwise, the tailgate deceleration control is performed based on the target speed.

[0013] When entering a high-risk alert state, this solution increases the sensitivity of vehicle detection parameters to enhance the system's ability to capture and respond to potential collisions, thereby further improving risk handling sensitivity. It calculates the static safety distance based on motion information and system response performance, performs adaptive calculations based on actual conditions, and dynamically sets the safety boundary to balance anti-pinch safety with the efficiency of normal tailgate operation.

[0014] In a further implementation scheme, when the risk level is medium risk, the risk response strategy includes: The predicted collision time is determined from the collision prediction conclusion, and the deceleration magnitude is dynamically calculated based on the predicted collision time; then, tailgate deceleration control is executed based on the deceleration magnitude. The deceleration magnitude is dynamically calculated based on the predicted collision time, including: When the predicted collision time is greater than or equal to the maximum time threshold, the minimum deceleration magnitude is used; When the predicted collision time is less than or equal to the minimum time threshold, the maximum deceleration magnitude is used; When the predicted collision time is greater than the minimum time threshold but less than the maximum time threshold, the deceleration magnitude is calculated proportionally.

[0015] This solution sets minimum and maximum deceleration ranges based on maximum and minimum time thresholds, establishing clear deceleration range boundaries. This ensures minimum safety performance while avoiding operational efficiency losses caused by over-response.

[0016] In a further implementation, when the risk level is low risk, the risk response strategy includes maintaining the normal closing speed of the tailgate.

[0017] In a further implementation, the dynamic anti-pinch threshold is calculated based on the real-time vehicle operating condition information and the motion information, including: The vehicle body tilt angle and ambient temperature are obtained from the real-time vehicle operating condition information, and the instantaneous operating speed of the tailgate is obtained from the motion information. A dynamic anti-pinch threshold is calculated based on the vehicle body tilt angle, ambient temperature, and instantaneous operating speed. The calculation formula for the dynamic anti-pinch threshold is as follows:

[0018] Where: m represents the weight of the tailgate, g represents the gravitational acceleration, θ represents the body tilt angle, α represents the angle between the tailgate and the body, T represents the ambient temperature, τ represents the time constant of temperature decay, V represents the instantaneous operating speed of the tailgate, k1 represents the proportionality coefficient of the tailgate's gravity component, k2 represents the proportionality coefficient of the temperature compensation term, and k3 represents the proportionality coefficient of the speed compensation term.

[0019] This solution calculates the dynamic anti-pinch threshold based on the vehicle tilt angle, ambient temperature, and instantaneous operating speed at the current moment, replacing the method of determining the tailgate anti-pinch trigger threshold by using a fixed force or current threshold. It can adapt to changes in slope, temperature, and speed, effectively overcome the resistance changes caused by slopes and low temperatures, improve the system's adaptability and the accuracy of safety judgment under different operating conditions, and fundamentally reduce the false trigger rate.

[0020] In a further implementation, tailgate control is performed based on the real-time contact force and the dynamic anti-pinch threshold, including: Compare the dynamic anti-pinch threshold with the preset upper limit value, and select the smaller value as the target anti-pinch threshold; Determine whether the real-time contact force is greater than the target anti-pinch threshold. If it is, immediately trigger the anti-pinch protection; otherwise, maintain the normal tailgate closing process. The anti-pinch protection includes driving the tailgate to perform a reverse motion until it is completely free from the obstacle or reaches a preset reverse distance.

[0021] To prevent the calculated threshold from being too high under extreme working conditions, this solution sets a preset upper limit for the dynamic anti-pinch threshold in accordance with relevant regulations and safety requirements, ensuring that the clamping force is within a safe range and avoiding harm to the human body.

[0022] Further implementation plans also include: real-time monitoring of the main power supply status, and activation of the backup power supply for emergency power supply when an abnormality or power outage of the main power supply is detected.

[0023] This solution integrates a backup power supply. When the main power supply of the vehicle is detected to be out of power or abnormal, it automatically switches to the backup power supply to ensure that the system can complete a motor reversal action of at least 300ms, meeting the ASIL-B functional safety requirements.

[0024] The present invention also provides an anti-pinch control system for an electric tailgate, for implementing an anti-pinch control method for an electric tailgate as described above, including a main controller and a first to a fifth sensor connected thereto; the first sensor includes a millimeter-wave radar installed on the inside of the tailgate, the second sensor includes a distributed micro-piezoelectric thin film array installed on the lower edge of the tailgate; the third sensor includes a dual-redundant Hall sensor, the fourth sensor includes a tilt sensor, and the fifth sensor includes a temperature sensor. Attached Figure Description

[0025] Figure 1 This is a flowchart of an anti-pinch control method for an electric tailgate provided in an embodiment of the present invention; Figure 2 This is a flowchart of the tailgate anti-pinch system architecture provided in an embodiment of the present invention; Figure 3 This is a flowchart of the tailgate anti-pinch algorithm provided in an embodiment of the present invention. Detailed Implementation

[0026] The embodiments of the present invention are described in detail below with reference to the accompanying drawings. The embodiments are given for illustrative purposes only and should not be construed as limiting the present invention. The accompanying drawings are for reference and illustration only and do not constitute a limitation on the scope of patent protection of the present invention, because many changes can be made to the present invention without departing from the spirit and scope of the present invention.

[0027] Example 1 This invention provides an anti-pinch control method for electric tailgates, such as... Figure 1 , Figure 2 , Figure 3 As shown, in this embodiment, it includes: S01. Monitor the main power supply status in real time. When an abnormality or power failure is detected in the main power supply, start the backup power supply for emergency power supply.

[0028] The backup power supply employs an integrated supercapacitor bank, which is connected to the vehicle's main power supply via a bidirectional DC / DC converter and controlled by a power management IC. This power management IC has multiple wake-up mechanisms, including vehicle ignition wake-up, CAN bus command wake-up, and timer wake-up (e.g., once per hour), ensuring continuous power replenishment of the supercapacitors and maintaining their state of charge at over 95%.

[0029] Furthermore, the system monitors the status of charge and health of the backup power supply in real time and establishes a tiered early warning mechanism. When the status of charge falls below 70% or the health status falls below 80%, the system will restrict the automatic tailgate function and issue a maintenance warning through the human-machine interface. Even in the worst-case scenario of complete backup power failure, the system can still provide basic safety assurance through the mechanical emergency unlocking device.

[0030] This embodiment integrates a backup power supply. When the main power supply of the vehicle is detected to be out of power or abnormal, it automatically switches to the backup power supply to ensure that the system can complete the motor reversal action of at least 300ms and meet the ASIL-B functional safety requirements.

[0031] S02. After the system is powered on, it performs self-tests and calibrations on each sensor (first to fifth sensors); enters standby mode and continuously monitors the tailgate control signal; the millimeter-wave radar module maintains low power consumption and periodically scans the detection area.

[0032] S1. Obtain real-time obstacle movement information to obtain temporal data of the obstacles, including: A first sensor is installed inside the tailgate to acquire the movement information of obstacles in real time. The movement information is accumulated over a preset time step to obtain the temporal data of the obstacles. The movement information includes one or more of the following: distance, speed, azimuth angle, pitch angle, and signal-to-noise ratio.

[0033] Non-contact detection: A 77GHz millimeter-wave radar is used as the first sensor and installed on the inside of the tailgate. The distance accuracy of the millimeter-wave radar is ±0.04m; scanning angle: horizontal ±75°, vertical: ±15°; distance resolution: ±0.1m; it can detect approaching obstacles within a range of 5-30cm and provide early warning.

[0034] Specifically, when a tailgate closing command is detected, the system enters radar pre-scanning mode, and the millimeter-wave radar continuously collects obstacle movement information at 50ms intervals, including: (1) Real-time distance between the obstacle and the tailgate; (2) Radial velocity of the obstacle; (3) The azimuth and pitch angles of the obstacle; (4) Radar echo signal quality (signal-to-noise ratio); The time-series data collected by millimeter-wave radar over 10 time steps are input into the LSTM prediction model for obstacle trajectory analysis.

[0035] This embodiment directly detects the movement information of obstacles and accumulates time-series data for the LSTM model to predict the trajectory. By analyzing the movement trajectory and trend, potential collision risks can be predicted in advance, allowing for lead time for tailgate control. This triggers a more precise and smooth avoidance action before clamping occurs, significantly improving the intelligence and reliability of the anti-pinch system.

[0036] S2. Obtain the motion information of the tailgate, combine it with the time series data of the obstacle, input it into the LSTM model to predict the trajectory of the obstacle, and output the collision prediction conclusion. Among them, the LSTM model is a predictive decision engine based on LSTM.

[0037] Model Function: The LSTM (Long Short-Term Memory) model is used to analyze the time series data of obstacle point cloud provided by millimeter-wave radar to predict its movement trajectory and intrusion risk in the next few milliseconds.

[0038] The LSTM model is trained on time-series millimeter-wave radar point cloud data collected from large-scale real-vehicle testing and simulated environments, with a data scale exceeding 1 million samples, covering various scenarios and conditions. For example, the model input consists of radar point cloud data for 10 consecutive time steps (100ms), with each point containing 5-dimensional features such as distance, velocity, azimuth, pitch, and signal-to-noise ratio, and undergoing coordinate transformation, alignment, and normalization preprocessing. The model output includes collision probability, normalized collision distance, and normalized collision time. The model structure is optimized, with approximately 100,000 parameters and a model size of approximately 400KB, enabling inference to be completed within 5ms on an in-vehicle embedded platform, meeting real-time requirements.

[0039] Workflow: When the collision probability predicted by the LSTM model is greater than 90%, the system reduces the tailgate running speed to 30% in advance to achieve proactive protection of "early warning before contact".

[0040] For example, based on radar data (mobility information) from 10 consecutive time steps, the following predictions are calculated using an LSTM model: (1) Collision probability The probability of an obstacle colliding with the tailgate (0-1); (2) Collision distance : The estimated relative distance (in millimeters) between the obstacle and the tailgate at the time of the collision; (3) Predicting collision time : The estimated remaining time (in milliseconds) required for a collision to occur.

[0041] In this embodiment, the collision prediction conclusion includes the collision probability, the predicted collision time, and the collision distance between the predicted collision and the obstacle.

[0042] S3. Perform tailgate control by matching the graded deceleration strategy according to the collision prediction conclusion. Specifically, extract the collision probability from the collision prediction conclusion, determine the risk level according to the collision probability and match the corresponding risk response strategy in the graded deceleration strategy, and perform tailgate control according to the risk response strategy.

[0043] This embodiment determines the risk level based on the collision probability and matches the corresponding risk response strategy in the graded deceleration strategy. By quantifying the collision probability and adapting the graded response, it decisively applies strong braking to high-risk situations, dynamically calculates the deceleration magnitude based on the predicted collision time to control the deceleration for medium-risk situations, and does not handle low-risk situations. In this way, while ensuring absolute safety, it minimizes unnecessary emergency stops and improves the smoothness of tailgate operation and user experience.

[0044] In this embodiment, when the risk level is high risk, the risk response strategy includes: Enter high-risk alert status and increase the sensitivity of vehicle detection parameters; Calculate the static safety distance based on the motion information and system response performance; The collision distance is determined from the collision prediction conclusion. Based on the static safety distance and the collision distance, it is determined whether to perform an emergency stop. Otherwise, the tailgate deceleration control is performed based on the target speed.

[0045] For example, high-risk collision response ( >0.9): Immediately reduce the tailgate operating speed to 30% of the rated speed; Key action: Predicting collision distance With respect to the preset static safety distance Comparison: If < This executes an emergency stop, which addresses the inadequacy of response that might result from relying solely on probability.

[0046] The system has entered a state of high alert.

[0047] Static safety distance It is an empirical value based on the worst-case response time and tailgate kinematics of the system, and the calculation formula is as follows: =( × )+

[0048] in, It is the current instantaneous speed of the tailgate. It is the total response time of the system under the worst-case scenario (including system processing time, communication delay time, and actuator response delay time, etc.). It is an additional cache safety margin.

[0049] The specific logic for emergency stop is as follows: (1) Upon first trigger, immediately reverse and pause, and give a warning (audio-visual reminder); (2) After a short wait, retry very slowly to avoid impact; (3) If the emergency stop is triggered 5 times, it is judged as a persistent obstacle, the tailgate system will stop completely and report an error; (4) After the system stops, the driver needs to manually reset the tailgate opening and closing system for it to work normally.

[0050] When the system detects a collision probability greater than 90%, it is defined as a high alert state. In this state, the entire control loop is configured with the highest priority and highest sensitivity, taking the following actions for highly probable physical contact and anti-pinch triggering: 1) Increase the refresh rate of millimeter-wave radar to 100Hz; piezoelectric thin film array: increase the sampling rate from 1kHz to 2kHz, doubling the timeliness; dual redundant Hall current sensor: increase the sampling rate from 10kHz to 20kHz, capturing the smallest change in motor torque. 2) Shorten the data fusion cycle: The calculation cycle for multi-sensor data fusion and decision-making is reduced from 10ms to 5ms.

[0051] In this embodiment, when entering a high-risk alert state, the sensitivity of vehicle detection parameters is increased to improve the system's ability to capture and respond to potential collisions, thereby further enhancing risk handling sensitivity. The static safety distance is calculated based on motion information and system response performance, and adaptive calculation is performed according to the actual situation to dynamically set the safety boundary, balancing anti-pinch safety with the efficiency of normal tailgate operation.

[0052] In this embodiment, when the risk level is medium risk, the risk response strategy includes: The predicted collision time is determined from the collision prediction conclusion, and the deceleration magnitude is dynamically calculated based on the predicted collision time; then, tailgate deceleration control is executed based on the deceleration magnitude. The deceleration magnitude is dynamically calculated based on the predicted collision time, including: When the predicted collision time is greater than or equal to the maximum time threshold, the minimum deceleration magnitude is used; When the predicted collision time is less than or equal to the minimum time threshold, the maximum deceleration magnitude is used; When the predicted collision time is greater than the minimum time threshold but less than the maximum time threshold, the deceleration magnitude is calculated proportionally.

[0053] For example, the response to a moderate collision risk (0.7≤ ≤0.9) The collision probability thresholds are determined through optimization using extensive empirical data, based on system functional safety objectives and user experience requirements. Specifically, the first threshold of 0.7 ensures the system has high sensitivity to potential risks, enabling early warning and gradual intervention; the second threshold of 0.9 ensures the system only triggers a strong response in highly certain emergency situations. This dual-threshold structure together constitutes a tiered decision-making system that balances safety and false alarm tolerance.

[0054] Furthermore, an adaptive threshold adjustment mechanism is set up to dynamically fine-tune the first and second thresholds within a range of ±0.1 based on environmental conditions, obstacle types, and system operating status, in order to adapt to complex and ever-changing real-world application scenarios.

[0055] Key action: Utilize Dynamically calculate deceleration rate; predict collision time The shorter the length, the greater the deceleration.

[0056] use The specific method for dynamically calculating the deceleration range is as follows: Set time threshold and : when ≤ At that time, the maximum deceleration range is adopted. ; when ≥ At that time, the minimum deceleration amplitude is adopted. ; when < < The deceleration rate is calculated using the following formula: A= +( - )*( - ) / ( - ) in, and Based on the medium-risk level, for example =10%, =50%.

[0057] Thus, the deceleration rate A decreases as follows As the percentage increases, it decreases linearly from 50% to 10%.

[0058] The system continues to monitor radar data and updates the risk assessment in real time.

[0059] This embodiment sets the minimum and maximum deceleration ranges based on the maximum and minimum time thresholds, establishing clear deceleration range boundaries. This ensures both minimum safety performance and avoids operational efficiency loss due to over-response.

[0060] In this embodiment, when When the risk level is determined to be low risk, the risk response strategy includes maintaining the normal closing speed of the tailgate.

[0061] S4. When a contact signal of the tailgate is detected, real-time contact force and real-time vehicle operating condition information are acquired, and a dynamic anti-pinch threshold is calculated based on the real-time vehicle operating condition information and the motion information. The tailgate contact signal is acquired using a distributed micro-piezoelectric thin-film array, which is installed along the lower edge of the tailgate, covering the entire tailgate area. The pressure detected by the sensor is converted into an electrical signal through energy conversion, as shown in the following conversion relationship: (t)=d (t)+ (t) in: It is the effective area of ​​the (i,j)th unit. (t) is the average pressure at that unit location. (t) represents the noise of the unit, and d is the piezoelectric constant. This is well-known technology in the field, and will not be described further in this embodiment.

[0062] The final safety anti-pinch reversal relies on physical contact signals (i.e., the tailgate contact signals). The high-collision-risk state predicted by the LSTM model is used for warning and deceleration, but cannot directly trigger the anti-pinch reversal.

[0063] In this embodiment, calculating the dynamic anti-pinch threshold based on the real-time vehicle operating condition information and the motion information includes: The vehicle body tilt angle and ambient temperature are obtained from the real-time vehicle operating condition information, and the instantaneous operating speed of the tailgate is obtained from the motion information. A dynamic anti-pinch threshold is calculated based on the vehicle body tilt angle, ambient temperature, and instantaneous operating speed. The calculation formula for the dynamic anti-pinch threshold is as follows:

[0064] In the formula, m represents the weight of the tailgate, g represents the gravitational acceleration, θ represents the vehicle tilt angle (the angle of inclination of the vehicle relative to the horizontal plane), α represents the angle between the tailgate and the vehicle body (i.e., the tailgate opening angle, which is obtained in real time by an angle sensor installed on the tailgate hinge), T represents the ambient temperature, τ represents the time constant of temperature decay, V represents the instantaneous operating speed of the tailgate (in m / s), k1 represents the proportionality coefficient of the tailgate gravity component, k2 represents the proportionality coefficient of the temperature compensation term, and k3 represents the proportionality coefficient of the speed compensation term. The coefficients k1, k2, and k3 are obtained through experimental calibration, and their typical value ranges are as follows: k1∈[0.7,0.9], k2∈[1.0,1.5], k3∈[0.03,0.07].

[0065] The vehicle body tilt angle is obtained through a tilt sensor; the ambient temperature is obtained through a temperature sensor and is used to compensate for the stiffness of the sealing strip.

[0066] This embodiment calculates the dynamic anti-pinch threshold based on the vehicle body tilt angle, ambient temperature, and instantaneous operating speed at the current moment, replacing the method of determining the tailgate anti-pinch trigger threshold by using a fixed force or current threshold. It can adapt to changes in slope, temperature, and speed, effectively overcome the resistance changes caused by slopes and low temperatures, improve the system's adaptability and the accuracy of safety judgment under different operating conditions, and fundamentally reduce the false trigger rate.

[0067] S5. Perform tailgate control based on the real-time contact force and the dynamic anti-pinch threshold, including: Compare the dynamic anti-pinch threshold with the preset upper limit value, and select the smaller value as the target anti-pinch threshold; Determine whether the real-time contact force is greater than the target anti-pinch threshold. If it is, immediately trigger the anti-pinch protection; otherwise, maintain the normal tailgate closing process. The anti-pinch protection includes driving the tailgate to perform a reverse motion until it is completely free from the obstacle or reaches a preset reverse distance.

[0068] For example, dynamic anti-pinch threshold Set a preset upper limit value (As an absolute upper limit value) It is usually set to 100N, that is Take the calculated value and The smaller one.

[0069] Specifically, when an obstacle enters the contact area of ​​the tailgate sealing strip, the piezoelectric film array detects the contact signal, the system collects contact force data (i.e., contact signal) in real time, and compares it with the dynamically calculated anti-pinch threshold. Compare them.

[0070] Perform the corresponding action based on the comparison result: If real-time contact force ≤ Continue with the normal shutdown process; If the real-time contact force > This immediately triggers the anti-pinch protection.

[0071] In this embodiment, the anti-pinch protection is implemented and the system recovers as follows: (1) When the anti-pinch protection is triggered, the control unit sends a reverse command to the motor driver within 80ms; (2) The supercapacitor energy storage module ensures that it can still reliably perform the reverse operation when the main power supply is abnormal; (3) The tailgate reverses until it is completely away from the obstacle or reaches the preset reversal distance; (4) The system records relevant data of this anti-pinch event for subsequent analysis and model optimization; (5) After the anti-pinch action is completed, the system returns to standby mode and waits for the next operation command.

[0072] To prevent the calculated threshold from being too high under extreme working conditions, this embodiment sets a preset upper limit value for the dynamic anti-pinch threshold according to relevant regulations and safety requirements, ensuring that the clamping force is within a safe range and avoiding harm to the human body.

[0073] S6. Continuously monitor the working status of each device. When a fault or communication abnormality is detected, automatically switch to a fault handling mode that matches the severity of the fault.

[0074] The equipment operating status includes the operating status of sensors, actuators, and power supply.

[0075] The fault handling modes include mild degradation mode, moderate degradation mode, severe degradation mode, emergency safety mode, and power degradation mode, specifically: L1: Slight downgrade (warning function limited) Triggering condition: Failure of a single non-contact sensor (e.g., millimeter-wave radar failure, communication interruption).

[0076] System Actions and Strategies: 1. Disable prediction-based graded velocity control (i.e., skip steps S1-S3).

[0077] 2. Relying entirely on contact-type anti-pinch (steps S4~S5) as the sole safety measure.

[0078] 3. Dynamic threshold The algorithm remains enabled to handle changes in slope and temperature.

[0079] 4. The tailgate closing speed should be maintained at a normal speed or a conservative fixed speed should be used.

[0080] L2: Moderate downgrading (sensitivity improvement) Triggering condition: Failure of a critical contact sensor (e.g., failure of a portion of a distributed piezoelectric thin film array, but the system can still detect contact).

[0081] System Actions and Strategies: 1. If the radar is functioning normally, continue to implement the non-contact prediction and graded velocity control strategy.

[0082] 2. Significantly reduces the dynamic threshold of contact anti-pinch mechanism. (For example, multiply it by a degradation factor less than 1) (e.g., 0.5-0.7), or directly use a fixed, lower security threshold.

[0083] 3. Increase the anti-pinch judgment weight of the current sensor as a supplement to the piezoelectric signal.

[0084] L3: Severe downgrade (minimum security mode) Triggering conditions: Complete failure of the piezoelectric thin film array or failure of a single one of the dual redundant current sensors.

[0085] System Actions and Strategies: 1. Disable the automatic shutdown function and only allow manual shutdown (if this function is available).

[0086] 2. Alternatively, switch the system to pure current detection anti-pinch mode and use a fixed, very conservative current threshold. This threshold is set extremely low, resulting in a high false trigger rate, but it ensures safety.

[0087] 3. If the radar is functioning normally, it can be retained for triggering "obstacle alerts" but not for automatic control.

[0088] L4: Emergency Safety Mode Triggering conditions: Both redundant current sensors fail or the main control unit fails to communicate with the motor driver.

[0089] System Actions and Strategies: 1. Immediately and safely stop (brake) the tailgate movement.

[0090] 2. Disable all automatic operation; only respond to "start" commands or manual operations.

[0091] 3. Guide users to use the mechanical emergency unlocking device.

[0092] Power degradation mode: Triggering condition: Abnormal or power failure of the vehicle's main power supply (unrelated to sensor status).

[0093] System Actions and Strategies: 1. The supercapacitor bank immediately takes over the power supply.

[0094] 2. The system's primary task is to ensure the execution of the "anti-pinch reverse" function.

[0095] 3. If a shutdown action is being performed during power switching and the anti-pinch function is triggered, ensure that the reversal is completed.

[0096] 4. The capacitor's charge is only used to ensure safe operation and may disable unnecessary automatic functions to save power.

[0097] In this embodiment, the tailgate anti-pinch control process is presented as a specific example as follows: Step 1: Start the tailgate.

[0098] When the user presses the close button, the tailgate begins to close at the rated speed (assuming 0.2 m / s).

[0099] Step 2: Radar detection.

[0100] The millimeter-wave radar scanned at a frequency of 100Hz and detected an obstacle (arm) on the tailgate path; the radar data (point cloud) was preprocessed to extract features such as the distance, radial velocity, and angle of the obstacle.

[0101] Assume that at time t=0, the radar detects the obstacle for the first time. At this time, the obstacle is 0.3m away from the edge of the tailgate and its radial velocity (relative to the tailgate) is -0.05m / s (indicating that it is close).

[0102] Step 3: Collision prediction using the LSTM model.

[0103] The system inputs radar data from the most recent 10 time steps (100ms) (each time step includes several features such as distance, velocity, and angle) into the LSTM model.

[0104] The LSTM model outputs three values: Collision probability =0.75 (collision probability); Collision distance =0.15m (predicted collision distance); Predicted collision time =1.2s (predicted collision time); Step 4: Speed ​​adjustment (based on prediction).

[0105] The system makes decisions based on collision probability and prediction time: because =0.75>0.7, which is considered a medium risk.

[0106] set up =0.5s, =3.0s, =10%, =60%.

[0107] because (0.5s)< (1.2s)< (3.0s), based on the response strategy for medium-risk situations, the basic speed reduction is calculated using the following formula: A= +( - )*( - ) / ( - ) That is: A = 10% + (60% - 10%) * (3.0 - 1.2) / (3.0 - 0.5) = 46% The system reduces the tailgate speed by 46%, and the target speed is adjusted to: 0.2m / s*(1-46%)=0.108m / s.

[0108] Step 5: Continuous monitoring and updating In the next 100ms (the next time step), the radar detected the obstacle again, at a distance of 0.28m and a radial velocity of -0.1m / s (a faster approach velocity).

[0109] The LSTM model re-predicts based on the new 10 frames of data (including the latest frame): =0.92; =0.12m; =0.8s; System Decision-Making: =0.92>0.9, triggering a high-risk alert.

[0110] The system enters a high alert state and reduces the tailgate speed to 30% of the rated speed, i.e., 0.06 m / s.

[0111] Step 6: Contact detection.

[0112] Under high alert conditions, the system samples the piezoelectric film and current sensor at 200Hz.

[0113] When the tailgate comes into contact with an obstacle, the piezoelectric film detects the contact force, assuming the real-time contact force is 8N.

[0114] At the same time, the system calculates dynamic thresholds. : Assuming parameters: m = 22.5 kg, θ = 0°, α = 60°, T = 25℃, V = 0.060 m / s =0.8*22.5*9.8*cos(0°)*sin(60°)++1.2*e^(-25 / 50)+0.05*(60)^1.5 =0.8*220.5*0.87+1.2*e^(-0.5)+0.05*465 ≈152.77 + 1.2 * 0.6065 + 23.25 ≈176.75N because Greater than (100N), then =100N.

[0115] Comparison: Real-time contact force 8N < =100N, therefore the anti-pinch reversal is not triggered.

[0116] Step 7: Continuous Contact and Anti-Pinch Trigger As the obstacle (arm) remains in place, the tailgate continues to close, and the contact force gradually increases.

[0117] When the contact force reaches 15N, the system compares again: 15N < 100N, so it does not trigger.

[0118] Assuming that the contact force continues to increase to 105N as the tailgate continues to move, at which point 105N>100N, the system will immediately trigger the anti-pinch reverse rotation.

[0119] Step 8: Anti-pinch reverse The system issues a reverse command within 80ms, the motor rotates in the opposite direction, the tailgate opens to a certain angle (e.g., a reverse time of 300ms), and then stops.

[0120] The system records anti-pinch events and alerts the user via warning lights or sounds.

[0121] Step 9: User Intervention After the user removes their arm, they press the close button again, and the tailgate closes successfully. At this point, there are no obstructions.

[0122] Example 2 This invention also provides an anti-pinch control system for an electric tailgate, used to implement the anti-pinch control method for an electric tailgate as described above, including a main controller and a first to a sixth sensor connected thereto; the first sensor includes a millimeter-wave radar installed on the inside of the tailgate, the second sensor includes a distributed micro-piezoelectric thin film array installed on the lower edge of the tailgate; the third sensor includes a dual-redundant Hall sensor, the fourth sensor includes a tilt sensor, the fifth sensor includes a temperature sensor, and the sixth sensor includes an angle sensor.

[0123] This embodiment uses millimeter-wave radar and distributed micro-piezoelectric thin film arrays as representatives of non-contact detection and contact sensing, combined with dual redundant Hall sensors, tilt sensors, temperature sensors and angle sensors for system status monitoring. Through multimodal sensing fusion, real-time monitoring of tailgate closure is achieved.

[0124] This invention introduces a "prediction + adaptation" intelligent decision-making paradigm. On one hand, it acquires the tailgate's motion information and combines it with the time-series data of obstacles, inputting it into an LSTM model to predict the obstacle's trajectory. The resulting collision prediction conclusion is then matched with a graded deceleration strategy to execute tailgate control. This advance prediction effectively improves system response efficiency and reduces the probability of pinching injuries. On the other hand, when a tailgate contact signal is detected, it acquires real-time contact force and real-time vehicle operating condition information, calculates a dynamic anti-pinch threshold, and compares it with the real-time contact force to execute tailgate control. By dynamically adjusting the dynamic anti-pinch threshold based on vehicle operating conditions and tailgate status, and adapting to different scenarios, the accuracy of anti-pinch recognition can be improved, and the probability of false triggering can be reduced. This results in a faster, smarter, and more reliable electric tailgate anti-pinch system.

[0125] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. A method for preventing pinching of an electric tailgate, characterized in that, include: Real-time acquisition of obstacle movement information yields temporal data of the obstacles; The motion information of the tailgate is obtained, and the time series data of the obstacle is combined with the LSTM model to predict the trajectory of the obstacle, and the collision prediction conclusion is output. Tailgate control is executed based on the collision prediction conclusions and a graded deceleration strategy. When a contact signal is detected from the tailgate, real-time contact force and real-time vehicle operating condition information are acquired, and a dynamic anti-pinch threshold is calculated based on the real-time vehicle operating condition information and the motion information. Tailgate control is performed based on the real-time contact force and the dynamic anti-pinch threshold.

2. The electric tailgate anti-pinch control method as described in claim 1, characterized in that, Real-time acquisition of obstacle movement information yields obstacle time-series data, including: A first sensor is installed inside the tailgate to acquire the movement information of obstacles in real time. The movement information is accumulated over a preset time step to obtain the temporal data of the obstacles. The movement information includes one or more of the following: distance, speed, azimuth angle, pitch angle, and signal-to-noise ratio.

3. The electric tailgate anti-pinch control method as described in claim 2, characterized in that: The collision prediction results include the collision probability, the predicted collision time, and the collision distance between the predicted object and the obstacle. Tailgate control is executed based on the collision prediction conclusion and the graded deceleration strategy. Specifically, the collision probability is extracted from the collision prediction conclusion, the risk level is determined based on the collision probability, and the corresponding risk response strategy in the graded deceleration strategy is matched. Tailgate control is then executed based on the risk response strategy.

4. The electric tailgate anti-pinch control method as described in claim 3, characterized in that, When the risk level is high risk, the risk response strategy includes: Enter high-risk alert status and increase the sensitivity of vehicle detection parameters; Calculate the static safety distance based on the motion information and system response performance; The collision distance is determined from the collision prediction conclusion. Based on the static safety distance and the collision distance, it is determined whether to perform an emergency stop. Otherwise, the tailgate deceleration control is performed based on the target speed.

5. The electric tailgate anti-pinch control method as described in claim 3, characterized in that, When the risk level is medium risk, the risk response strategy includes: The predicted collision time is determined from the collision prediction conclusion, and the deceleration magnitude is dynamically calculated based on the predicted collision time; then, tailgate deceleration control is executed based on the deceleration magnitude. The deceleration magnitude is dynamically calculated based on the predicted collision time, including: When the predicted collision time is greater than or equal to the maximum time threshold, the minimum deceleration magnitude is used; When the predicted collision time is less than or equal to the minimum time threshold, the maximum deceleration magnitude is used; When the predicted collision time is greater than the minimum time threshold but less than the maximum time threshold, the deceleration magnitude is calculated proportionally.

6. The electric tailgate anti-pinch control method as described in claim 4, characterized in that, When the risk level is low risk, the risk response strategy includes maintaining the normal closing speed of the tailgate.

7. The electric tailgate anti-pinch control method as described in claim 1, characterized in that, Calculating the dynamic anti-pinch threshold based on the real-time vehicle operating condition information and the motion information includes: The vehicle body tilt angle and ambient temperature are obtained from the real-time vehicle operating condition information, and the instantaneous operating speed of the tailgate is obtained from the motion information. A dynamic anti-pinch threshold is calculated based on the vehicle body tilt angle, ambient temperature, and instantaneous operating speed. The calculation formula for the dynamic anti-pinch threshold is as follows: Where: m represents the weight of the tailgate, g represents the gravitational acceleration, θ represents the body tilt angle, α represents the angle between the tailgate and the body, T represents the ambient temperature, τ represents the time constant of temperature decay, V represents the instantaneous operating speed of the tailgate, k1 represents the proportionality coefficient of the tailgate's gravity component, k2 represents the proportionality coefficient of the temperature compensation term, and k3 represents the proportionality coefficient of the speed compensation term.

8. The electric tailgate anti-pinch control method as described in claim 4, characterized in that, Tailgate control is performed based on the real-time contact force and the dynamic anti-pinch threshold, including: Compare the dynamic anti-pinch threshold with the preset upper limit value, and select the smaller value as the target anti-pinch threshold; Determine whether the real-time contact force is greater than the target anti-pinch threshold. If it is, immediately trigger the anti-pinch protection; otherwise, maintain the normal tailgate closing process. The anti-pinch protection includes driving the tailgate to perform a reverse motion until it is completely free from the obstacle or reaches a preset reverse distance.

9. The electric tailgate anti-pinch control method as described in claim 1, characterized in that, Also includes: The system monitors the main power supply status in real time. When an abnormality or power failure is detected in the main power supply, the backup power supply is activated for emergency power supply.

10. An anti-pinch control system for an electric tailgate, used to implement the anti-pinch control method for an electric tailgate as described in any one of claims 1 to 9, characterized in that: It includes a main controller and a first to a fifth sensor connected to it; the first sensor includes a millimeter-wave radar installed inside the tailgate, the second sensor includes a distributed micro-piezoelectric thin film array installed on the lower edge of the tailgate, the third sensor includes a dual-redundant Hall sensor, the fourth sensor includes a tilt sensor, and the fifth sensor includes a temperature sensor.

Images