Self-adaptive anti-pinch safety protection system of electric seat
The electric seat anti-pinch system, which integrates multi-sensor information fusion and dual judgment logic, solves the problem of interference in existing electric seat anti-pinch systems, achieves early prediction and precise triggering, and improves the safety and reliability of seat adjustment.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-21
- Publication Date
- 2026-03-31
AI Technical Summary
Existing anti-pinch systems for electric seats are susceptible to interference, resulting in false or missed triggers. They are unable to achieve a precise and rapid response before the clamping force reaches the injury threshold, leading to insufficient safety and reliability.
It adopts multi-sensor information fusion and dual judgment logic, including a sensor signal acquisition module, a seat control module, a motor drive module and an alarm module. The motor drive module collects data on the seat operation status and the surrounding environment, monitors and predicts the risk of pinching in real time, and dynamically adjusts the anti-pinch strategy by combining the adaptive anti-pinch control algorithm and the anti-pinch judgment logic of the Hall sensor. When the Hall sensor fails, it switches to the current detection backup strategy.
It enables early prediction and precise triggering of anti-pinch protection, reduces the probability of false alarms and missed alarms, and improves the safety and reliability of seat adjustment.
Smart Images
Figure CN121756985A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of automotive seat safety technology, specifically relating to an adaptive anti-pinch safety protection system for electric seats. Background Technology
[0002] With the increasing demand for intelligent and comfortable vehicles, electric seat adjustment systems have become a core feature of modern vehicles. However, during complex movements such as backrest folding, fore-and-aft sliding, and height adjustment, there is a potential risk of pinching occupants or interfering with obstacles inside the vehicle. Therefore, developing and integrating highly reliable anti-pinch protection functions has become a key challenge in improving driving safety and system robustness. Currently, most mainstream anti-pinch solutions are based on monitoring mechanisms of a single signal (such as motor current signal). Current signals are easily affected by factors such as power supply voltage fluctuations, system friction, temperature changes, and motor aging, which may lead to false triggering or even more dangerous missed triggering. Relying solely on a hysteretic parameter like current makes it difficult to achieve a precise and rapid response before the pinching force reaches the injury threshold.
[0003] Therefore, the industry urgently needs an anti-pinch protection strategy with stronger anti-interference capabilities, earlier judgment, and higher reliability to make up for the shortcomings of traditional single signal detection schemes and effectively improve the safety performance of automotive seat systems. Summary of the Invention
[0004] In view of this, the purpose of this invention is to provide an adaptive anti-pinch safety protection system for electric seats. This invention aims to achieve early prediction and precise triggering of anti-pinch protection through multi-sensor information fusion and dual judgment logic, reducing the probability of false alarms and missed alarms, and comprehensively improving the safety and reliability of seat adjustment.
[0005] This invention provides an adaptive anti-pinch safety protection system for an electric seat, including a sensor signal acquisition module, a seat control module, a motor drive module, and an alarm module;
[0006] Sensor signal acquisition module: used to collect seat operating status and surrounding environment data in real time and in multiple dimensions; used to filter and preprocess the collected data, convert the data into digital information suitable for analysis, and then transmit it to the seat control module;
[0007] The sensor signal acquisition module collects data including motor drive current, system power supply voltage, ambient temperature, pressure signals distributed on the seat contact surface, and Hall sensor signals used to detect motor speed and position.
[0008] Seat control module: Used to ensure the control accuracy and operational safety of the seat;
[0009] Motor drive module: Used to control each seat motor to operate normally according to instructions. If the anti-pinch condition is activated, the seat motor will stop immediately and move in the opposite direction a preset distance to quickly eliminate the danger.
[0010] Alarm module: Used to issue an audible alarm when anti-pinch behavior is detected or a potential anti-pinch trend is identified, so as to promptly alert the user.
[0011] Furthermore, the system also has location learning and location memory functions;
[0012] The position learning function means that by driving each seat motor to complete the full stroke, the number of Hall pulse strokes and the operating current of each seat motor are collected and recorded to locate the current position of the seat.
[0013] The position memory function means that the real-time position data of the seat can be permanently saved;
[0014] Each seat is equipped with a pressure sensor to detect the occupant's presence in real time and dynamically divide the anti-pinch zone and the stall zone accordingly.
[0015] The system will update the position learning value when both of the following conditions are met: first, the seat is at the end of its mechanical stroke; second, the motor stalls.
[0016] Furthermore, the seat control module incorporates an adaptive anti-pinch control algorithm and an anti-pinch judgment logic based on a Hall sensor;
[0017] The adaptive anti-pinch control algorithm is used for real-time safety monitoring in the anti-pinch area, and dynamically predicts and updates the anti-pinch Hall pulse change rate K value.
[0018] The anti-pinch judgment logic based on Hall sensor is used to provide stall protection in the stall area at the end of the stroke.
[0019] Furthermore, the adaptive anti-pinch control algorithm obtains the anti-pinch Hall pulse change rate K value as follows: I. First, the data collected by the sensor signal acquisition module is used as input. The sensor signal acquisition module samples with a fixed time window T and converts each parameter into a time series vector, which together constitute the input tensor.
[0020] II. Subsequently, each sensor sequence is passed through an independent branch of a one-dimensional convolutional neural network to extract its unique temporal features; for any sensor sequence X_i, the convolutional feature extraction formula is:
[0021] F_i = ReLU(W_i * X_i + b_i)
[0022] Where * denotes a one-dimensional convolution operation; W_i is the convolution kernel weight; and b_i is the bias term.
[0023] III. Next, the feature vectors output from all branches are concatenated to obtain the comprehensive feature F_combined. Then, the comprehensive feature is fed into an attention mechanism layer to calculate the normalized attention weight α_i.
[0024] IV. The weighted fusion output anti-pinch Hall pulse change rate K value is calculated using the following formula:
[0025] K = Σ(α_i·C_i)
[0026] Here, C_i is obtained by mapping the feature vector F_i through a fully connected layer.
[0027] Furthermore, the anti-pinch judgment logic based on the Hall sensor is as follows:
[0028] I. First, set a maximum Hall pulse cycle anti-pinch force threshold when the seat is cold-started to ensure a rapid response to potential clamping at the moment of start-up;
[0029] II. Subsequently, the baseline learning phase begins, using the average Hall pulse period during normal seat operation as the baseline value T_avg; during the real-time monitoring phase, the current average Hall pulse period T_now is continuously monitored.
[0030] The anti-pinch detection is divided into two levels: if T_now exceeds K times T_avg, it is determined that a pinch has occurred, the seat will immediately reverse to a safe distance and trigger an alarm; if T_now shows a continuous increasing trend but does not reach the anti-pinch threshold, a warning will be issued and an audible alarm will be activated first.
[0031] Furthermore, the seat control module has a built-in backup anti-pinch strategy based on current detection. That is, when the Hall sensor fails, the system will automatically activate the backup anti-pinch strategy based on current detection.
[0032] Furthermore, the backup anti-pinch strategy based on current detection: by monitoring the current change during the entire movement of the seat, the stall and anti-pinch conditions are determined, thereby ensuring the safe operation of the seat in the sensor failure mode.
[0033] Beneficial effects:
[0034] This invention discloses an adaptive anti-pinch safety protection system for electric seats. This system integrates multi-source sensor data from the target seat to identify anti-pinch risks in real time and adopts corresponding anti-pinch strategies based on the seat's location within different areas. Simultaneously, the system automatically activates an audible alarm, thus achieving intelligent adaptation and comprehensive protection for various complex seat adjustment conditions. This anti-pinch system can precisely adjust the anti-pinch force based on the seat's real-time movement status and dynamic information of the surrounding environment, improving the accuracy and real-time performance of anti-pinch protection, reducing the probability of false or missed pinches, and further enhancing the safety and reliability of the seat adjustment process.
[0035] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description
[0036] Figure 1 This is a flowchart illustrating the learning process of an adaptive anti-pinch safety protection system for an electric seat according to the present invention.
[0037] Figure 2 This is a general schematic diagram of an adaptive anti-pinch safety protection system for an electric seat according to the present invention;
[0038] Figure 3 This is a flowchart of the adaptive algorithm for an adaptive anti-pinch safety protection system for an electric seat according to the present invention.
[0039] Figure 4 This is a flowchart of the anti-pinch algorithm for an adaptive anti-pinch safety protection system for an electric seat according to the present invention. Detailed Implementation
[0040] To make the technical solutions, advantages, and objectives of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention without creative effort are within the protection scope of this application.
[0041] like Figure 1 As shown, the present invention provides an adaptive anti-pinch safety protection system for an electric seat, including a sensor signal acquisition module, a seat control module, a motor drive module, and an alarm module;
[0042] Sensor signal acquisition module: Used to collect real-time, multi-dimensional data on the seat's operating status and surrounding environment; used to filter and preprocess the collected data (aiming to convert the raw signals into digital information suitable for analysis before transmitting them to the seat control module. These are all conventional processing techniques, such as RC filtering, data normalization, and data conversion), converting the data into digital information suitable for analysis before transmitting it to the seat control module;
[0043] The sensor signal acquisition module collects data including motor drive current, system power supply voltage, ambient temperature, pressure signals distributed on the seat contact surface, and Hall sensor signals used to detect motor speed and position.
[0044] Seat control module: Used to ensure the control accuracy and operational safety of the seat;
[0045] Motor drive module: Used to control each seat motor to operate normally according to instructions. If the anti-pinch condition is activated, the seat motor will stop immediately and move in the opposite direction a preset distance to quickly eliminate the danger.
[0046] Alarm module: It connects with the vehicle's infotainment system and can issue an audible alarm through the vehicle's large screen when it detects anti-pinch behavior or identifies a potential anti-pinch trend, thus promptly alerting the user.
[0047] As a preferred embodiment, the system also has a location learning function and a location memory function;
[0048] The position learning function means that by driving each seat motor to complete the full stroke, the number of Hall pulse strokes and the operating current of each seat motor are collected and recorded to locate the current position of the seat.
[0049] The position memory function can persistently save the real-time position data of the seat, effectively preventing the loss of position information due to power failure or system restart;
[0050] Each seat is equipped with a pressure sensor to detect the occupant's presence in real time, dynamically dividing the area into anti-pinch and stall protection zones. Different anti-pinch strategies are applied to different zones. Specifically, the system monitors seat pressure signals in real time to determine the occupant's status. When no one is detected, the anti-pinch zone remains at its default setting (5%-95%); when someone is detected, the system dynamically sets the anti-pinch and stall protection zones based on the direction of travel and the occupant's position, and specifically expands the anti-pinch range on the occupant side by 3% for precise protection.
[0051] As a preferred embodiment, the seat control module incorporates an adaptive anti-pinch control algorithm and anti-pinch judgment logic based on Hall sensors;
[0052] The adaptive anti-pinch control algorithm is used for real-time safety monitoring in the anti-pinch area, dynamically predicting and updating the anti-pinch Hall pulse change rate K value. The steps of the adaptive anti-pinch control algorithm to obtain the anti-pinch Hall pulse change rate K value are as follows:
[0053] I. First, the data collected by the sensor signal acquisition module is used as input. The sensor signal acquisition module samples the data in a fixed time window T and converts each parameter into a time series vector, which together constitute the input tensor.
[0054] II. Subsequently, each sensor sequence is passed through an independent branch of a one-dimensional convolutional neural network to extract its unique temporal features; for any sensor sequence X_i, the convolutional feature extraction formula is:
[0055] F_i = ReLU(W_i * X_i + b_i)
[0056] Where * denotes a one-dimensional convolution operation; W_i is the convolution kernel weight; and b_i is the bias term.
[0057] III. Next, the feature vectors output from all branches are concatenated to obtain the comprehensive feature F_combined. Then, the comprehensive feature is fed into an attention mechanism layer to calculate the normalized attention weight α_i.
[0058] IV. The weighted fusion output anti-pinch Hall pulse change rate K value is calculated using the following formula:
[0059] K = Σ(α_i·C_i)
[0060] Here, C_i is obtained by mapping the feature vector F_i through a fully connected layer.
[0061] The Hall sensor-based anti-pinch judgment logic is used to provide stall protection in the stall area at the end of the stroke, while simultaneously disabling the anti-pinch function, thus balancing safety and operational reliability. The Hall sensor-based anti-pinch judgment logic is as follows:
[0062] I. First, set a maximum Hall pulse cycle anti-pinch force threshold when the seat is cold-started, and use the sampling cycle of the initialized stable Hall sensor as a reference to ensure that the anti-pinch function is triggered in a timely and accurate manner;
[0063] II. Subsequently, the baseline learning phase begins, using the average Hall pulse period during normal seat operation as the baseline value T_avg; during the real-time monitoring phase, the current average Hall pulse period T_now is continuously monitored.
[0064] The anti-pinch detection is divided into two levels: if T_now exceeds K times T_avg, it is determined that a pinch has occurred, the seat will immediately reverse to a safe distance and trigger an alarm; if T_now shows a continuous increasing trend but does not reach the anti-pinch threshold, a warning will be issued and an audible alarm will be activated first.
[0065] As a preferred embodiment, the seat control module has a built-in backup anti-pinch strategy based on current detection. That is, when the Hall sensor fails, the system will automatically activate the backup anti-pinch strategy based on current detection.
[0066] Backup anti-pinch strategy based on current detection: By monitoring the current change during the entire movement of the seat, the stall and anti-pinch conditions are determined, thereby ensuring the safe operation of the seat in the event of a sensor failure.
[0067] The operation process of the adaptive anti-pinch safety protection system for an electric seat according to the present invention is as follows:
[0068] like Figure 1 As shown, before implementing anti-pinch control of the seat, a self-learning process must be executed to obtain the seat's real-time position information. This primarily involves acquiring the total number of Hall pulses from the seat motor across its entire travel range, thus establishing an accurate correspondence between displacement and pulse count. Based on this, by monitoring the Hall pulse count during movement in real time, the current position of the seat can be accurately calculated and fed back in real time. This process mainly relies on Hall sensor signals from the seat motor: first, the seat motor is driven in one direction to a stall position, then in the opposite direction to a stall position, and the total number of Hall pulses throughout the entire travel is recorded. If this value is within the calibration error range, self-learning is successful, and the system can obtain the current position of each axis of the seat. Throughout the process, the validity of the Hall signals is also checked to further ensure the accuracy and reliability of the position information. To further improve system accuracy, the range of position information is magnified by a factor of 10 to reduce quantization error and improve the resolution of measurement and control.
[0069] To prevent incorrect seat position information due to anti-pinch triggering, the system will only update the position learning value when both of the following conditions are met: first, the seat is at the end of its mechanical travel; second, the motor has indeed stalled. This mechanism avoids seat adjustment malfunctions caused by incorrect learning at the source.
[0070] The system first determines the presence of an occupant based on signals from the seat pressure sensor, dynamically dividing the system into anti-pinch and stall zones and implementing differentiated control strategies. In the anti-pinch zone, an adaptive anti-pinch algorithm is activated for real-time safety monitoring; conversely, in the stall zone at the end of the travel, the system switches to stall protection logic. This solution effectively improves the reliability and safety of anti-pinch control through precise zoned management.
[0071] like Figure 2 As shown, after the seat completes its self-learning process, its main workflow is as follows: It begins when the system detects an adjustment command, and each sensor then synchronously collects current, voltage, pressure, Hall effect, and temperature signals; then, the system performs anti-pinch detection on the above signals; finally, based on the detection results, the system decides whether to perform normal adjustment or activate anti-pinch retraction and alarm.
[0072] Specifically, such as Figure 3 As shown, the system first preprocesses the acquired raw signals to effectively remove noise and interference. Then, it divides the seat's travel distance into an anti-pinch zone and a stall zone. The range of the anti-pinch zone can be flexibly set according to the user's actual needs. Upon receiving a seat adjustment signal, the system continuously monitors the seat's current position and determines whether it has entered the anti-pinch zone. Within the anti-pinch zone, the system uses an adaptive algorithm to monitor the risk of pinching in real time. The core of this approach is a multi-parameter fusion strategy based on a CNN model to achieve more accurate anti-pinch status determination. Specifically, the model first independently analyzes the individual effects of voltage, temperature, and pressure on the Hall pulse, then dynamically calculates the weights of each effect using an attention mechanism, and finally outputs a real-time changing rate of change K value for the anti-pinch Hall pulse through weighted fusion. This method, through multi-source signal fusion and AI prediction, enables the system to effectively adapt to complex and changing working conditions.
[0073] After obtaining the predicted K value, the anti-pinch judgment logic is as follows: Figure 4 As shown, the system sets a maximum Hall pulse cycle anti-pinch force threshold during seat cold starts to ensure a rapid response to potential pinching at startup. Subsequently, the system enters a baseline learning phase, calculating the average Hall pulse cycle during normal seat operation as a baseline value T_avg. During the real-time monitoring phase, the system continuously monitors the current average Hall pulse cycle T_now. Anti-pinch determination is divided into two levels: if T_now exceeds K times T_avg, pinching is determined to have occurred, the seat immediately reverses to a safe distance, and an alarm is triggered; if T_now shows a continuous increasing trend but does not reach the anti-pinch threshold, the system issues a warning, first activating an audible alarm.
[0074] Furthermore, the system employs a redundant safety design. When the Hall signal fails, the system automatically switches to a backup anti-pinch strategy based on current detection. This strategy involves three stages: First, during the system's self-learning phase, it learns the average current value I_avg during normal seat operation under no-load conditions as a benchmark. Second, during real-time monitoring, to eliminate interference from the surge current at the moment of motor startup, the system ignores the current data during the initial startup phase. Finally, after stable operation, if the system detects that the current average current continuously exceeds K*I_avg for a certain duration (e.g., 300ms), it determines that pinching has occurred, immediately triggers reversal and an alarm, thereby ensuring safety redundancy in the event of a main sensor failure.
[0075] It is hereby declared that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.
Claims
1. An adaptive anti-pinch safety protection system for an electric seat, characterized in that: It includes a sensor signal acquisition module, a seat control module, a motor drive module, and an alarm module; Sensor signal acquisition module: used to collect seat operating status and surrounding environment data in real time and in multiple dimensions; used to filter and preprocess the collected data, convert the data into digital information suitable for analysis, and then transmit it to the seat control module; The sensor signal acquisition module collects data including motor drive current, system power supply voltage, ambient temperature, pressure signals distributed on the seat contact surface, and Hall sensor signals used to detect motor speed and position. Seat control module: Used to ensure the control accuracy and operational safety of the seat; Motor drive module: Used to control each seat motor to operate normally according to instructions. If the anti-pinch condition is activated, the seat motor will stop immediately and move in the opposite direction a preset distance to quickly eliminate the danger. Alarm module: Used to issue an audible alarm when anti-pinch behavior is detected or a potential anti-pinch trend is identified, so as to promptly alert the user.
2. The adaptive anti-pinch safety protection system for an electric seat according to claim 1, characterized in that: The system also has location learning and location memory functions; The position learning function means that by driving each seat motor to complete the full stroke, the number of Hall pulse strokes and the operating current of each seat motor are collected and recorded to locate the current position of the seat. The position memory function means that the real-time position data of the seat can be permanently saved; Each seat is equipped with a pressure sensor to detect the occupant's presence in real time and dynamically divide the anti-pinch zone and the stall zone accordingly. The system will update the position learning value when both of the following conditions are met: first, the seat is at the end of its mechanical stroke; second, the motor stalls.
3. The adaptive anti-pinch safety protection system for an electric seat according to claim 2, characterized in that: The seat control module incorporates an adaptive anti-pinch control algorithm and anti-pinch judgment logic based on Hall sensors; The adaptive anti-pinch control algorithm is used for real-time safety monitoring in the anti-pinch area, and dynamically predicts and updates the anti-pinch Hall pulse change rate K value. The anti-pinch judgment logic based on Hall sensor is used to provide stall protection in the stall area at the end of the stroke.
4. The adaptive anti-pinch safety protection system for an electric seat according to claim 3, characterized in that, The steps for obtaining the anti-pinch Hall pulse change rate K value in the adaptive anti-pinch control algorithm are as follows: I. First, the data collected by the sensor signal acquisition module is used as input. The sensor signal acquisition module samples the data in a fixed time window T and converts each parameter into a time series vector, which together constitute the input tensor. II. Subsequently, each sensor sequence is passed through an independent branch of a one-dimensional convolutional neural network to extract its unique temporal features; for any sensor sequence X_i, the convolutional feature extraction formula is: F_i = ReLU(W_i * X_i + b_i) Where * denotes a one-dimensional convolution operation; W_i is the convolution kernel weight; and b_i is the bias term. III. Next, the feature vectors output from all branches are concatenated to obtain the comprehensive feature F_combined. Then, the comprehensive feature is fed into an attention mechanism layer to calculate the normalized attention weight α_i. IV. The weighted fusion output anti-pinch Hall pulse change rate K value is calculated using the following formula: K = Σ(α_i·C_i) Here, C_i is obtained by mapping the feature vector F_i through a fully connected layer.
5. The adaptive anti-pinch safety protection system for an electric seat according to claim 4, characterized in that, The anti-pinch judgment logic based on the Hall sensor is as follows: I. First, set a maximum Hall pulse cycle anti-pinch force threshold when the seat is cold-started to ensure a rapid response to potential clamping at the moment of start-up; II. Subsequently, the benchmark learning phase begins, using the average Hall pulse period during normal seat operation as the benchmark value T_avg; During the real-time monitoring phase, the current average Hall pulse period T_now is continuously detected; The anti-pinch detection is divided into two levels: if T_now exceeds K times T_avg, it is determined that a pinch has occurred, the seat will immediately reverse to a safe distance and trigger an alarm; if T_now shows a continuous increasing trend but does not reach the anti-pinch threshold, a warning will be issued and an audible alarm will be activated first.
6. The adaptive anti-pinch safety protection system for an electric seat according to claim 5, characterized in that: The seat control module has a built-in backup anti-pinch strategy based on current detection. That is, when the Hall sensor fails, the system will automatically activate the backup anti-pinch strategy based on current detection.
7. The adaptive anti-pinch safety protection system for an electric seat according to claim 6, characterized in that: The backup anti-pinch strategy based on current detection: By monitoring the current changes during the entire movement of the seat, the stall and anti-pinch conditions are determined, thereby ensuring the safe operation of the seat in the sensor failure mode.