Electric Seat Anti-pinch Control Method and Device

Abstract

This invention provides a method and device for anti-pinch control of electric seats. The method includes real-time acquisition of current, speed, and physical angle during motor operation; when the current motor current exceeds a dynamic current threshold or the transient collision energy exceeds a collision energy threshold, and the current speed is less than a expected reference speed, the motor is controlled to stop or reverse. The dynamic current threshold is determined comprehensively based on a preset angle reference ammeter, load mass compensation, ambient temperature compensation, and a mechanical wear self-learning factor updated with the number of opening and closing cycles; the transient collision energy is calculated only when the transient current change rate exceeds a dynamic slope threshold, at which point an integral observation window is activated. The electric seat anti-pinch control method and device provided by this invention achieve anti-pinch functionality during the operation of the electric seat.

Description

Technical Field

[0001] This invention relates to the field of smart home control, and more specifically, to a method and device for preventing electric seat pinching. Background Technology

[0002] With the global home furnishing market's intelligent transformation, products such as motorized reclining chairs, smart chaise lounges, and height-adjustable tables have become core components for enhancing living comfort in modern homes. These devices typically use DC motors to drive complex scissor arms, linkage mechanisms, and telescopic supports to achieve functions such as backrest tilting and leg support height adjustment. However, when adjusting the position of a reclining chair, significant shearing forces and compression spaces are created between the mechanical linkages. If children are playing nearby or pets attempt to crawl under the reclining mechanism, severe crush injuries can easily occur once the device is in operation.

[0003] Existing anti-pinch solutions are mainly divided into two categories: contact and non-contact. Non-contact solutions often use infrared sensors, ultrasonic sensors, or millimeter-wave radar to detect object intrusion. Although these solutions can detect risks in advance, infrared sensors are susceptible to interference from ambient light and have difficulty distinguishing between normal decorative accessories on the seat and real obstacles, resulting in a very high false alarm rate. In addition, adding such high-precision sensors significantly increases the material cost of the product and the complexity of production and debugging.

[0004] Contact-based solutions typically install pressure sensors or capacitive sensing strips along the edges of the seat or in areas prone to pinching. While pressure sensor solutions can directly report physical pressure, they suffer from the following drawbacks in practical applications: First, the sensors are usually complexly wired and prone to aging and open circuits due to frequent pressure over time; second, the sensors can only monitor specific coverage areas, resulting in significant blind spots; and finally, because the soft material of the seat absorbs some pressure, by the time the sensor detects pressure reaching a preset danger threshold, the obstacle has often already been substantially damaged—the so-called "dull knife effect."

[0005] In summary, the relevant anti-pinch solutions suffer from low accuracy and high hardware costs. Summary of the Invention

[0006] This application provides an anti-pinch control method and device for electric seats, which solves the problems of low accuracy and high hardware cost.

[0007] In a first aspect, the present invention provides an anti-pinch control method for an electric seat, the electric seat including a rotating part and a motor, the motor being used to drive the rotating part to rotate; when the current current of the motor is greater than a dynamic current threshold and the current speed of the motor is less than a expected reference speed, the motor stops rotating or rotates in the opposite direction.

[0008] This invention uses two indicators—motor current exceeding a threshold and speed falling below expectations—to effectively distinguish between normal load fluctuations and actual physical collisions, thereby improving the accuracy of anti-pinch triggering.

[0009] Preferably, the dynamic current threshold is determined based on the reference current, load mass compensation constant, speed compensation term, ambient temperature compensation, and mechanical wear self-learning factor; wherein, the reference current is obtained by acquiring the current angle and finding the angle based on the current angle. The reference ammeter is determined, wherein the angle-reference ammeter is used to characterize the relationship between the angle and the reference current.

[0010] This invention comprehensively considers compensation from five dimensions, including reference current, load, speed, temperature, and wear, enabling the system to adapt to complex mechanical environment changes and significantly reducing the false alarm rate.

[0011] Preferred angle The reference ammeter is determined by controlling the electric seat to perform a complete opening and closing action in an unloaded state during initialization or reset, and recording the current at 1-degree angular intervals.

[0012] This invention eliminates nonlinear load interferences such as gravitational torque, leverage ratio, and spring assistance caused by seat angle changes by initializing unloaded modeling.

[0013] The preferred dynamic current threshold is also determined based on static safety redundancy; the calculation formula for the dynamic current threshold is as follows:

[0014]

[0015] in, For dynamic current threshold, As the reference current, Let be the load mass compensation constant. For speed compensation, For mechanical wear self-learning factor, For ambient temperature compensation, For static safety redundancy.

[0016] This invention uses quantitative formulas to calculate and introduce static safety redundancy, effectively absorbing circuit noise and instantaneous power fluctuations, and preventing the system from being falsely triggered by minor vibrations.

[0017] Preferably, the load mass compensation constant is determined based on the real-time current, reference current, ambient temperature compensation, speed compensation term, and mechanical wear self-learning factor within the load mass compensation evaluation window; the calculation formula for the load mass compensation constant is as follows:

[0018]

[0019] in, For the real-time current within the load quality compensation assessment window, The reference current within the load quality compensation assessment window, For environmental temperature compensation within the load quality compensation assessment window, This refers to the speed compensation within the load quality compensation assessment window; the load quality compensation assessment window is a preset time period after the start-up surge blanking period ends.

[0020] This invention automatically identifies changes in user weight or load within a specific window during the startup phase, ensuring that the system maintains consistent anti-pinch sensitivity regardless of the user's size.

[0021] Preferably, the mechanical wear self-learning factor is updated as the number of opening and closing cycles changes. The updating of the mechanical wear self-learning factor as the number of opening and closing cycles changes includes: when a preset number of complete opening and closing cycles have been completed and the system is in an unloaded operating state, collecting real-time current at each angular position during the unloaded operating state, and calculating the root mean square deviation between the real-time current and the reference current; if the root mean square deviation exceeds a preset circuit background noise threshold, then the updated mechanical wear self-learning factor is equal to the sum of the original mechanical wear self-learning factor and a preset infinitesimal weight step size.

[0022] This invention enables the system to evolve over time, absorbing the resistance drift of the transmission mechanism caused by long-term use and extending the maintenance-free cycle of the product.

[0023] Preferably, the current speed of the motor being less than the expected reference speed includes: the deviation between the current speed of the motor and the expected reference speed being lower than the anomaly determination threshold; wherein, the expected reference speed is the sum of the command expected speed and the physical model correction amount.

[0024] In a second aspect, the present invention provides an anti-pinch control method for an electric seat. The electric seat includes a rotating part and a motor, the motor driving the rotating part to rotate. When the current current of the motor is greater than a dynamic current threshold and the current speed of the motor is less than a expected reference speed, the motor stops rotating or rotates in the reverse direction. The dynamic current threshold is determined based on a reference current, a comprehensive compensation constant, and a speed compensation term. The reference current is obtained by acquiring the current angle and finding the angle based on the current angle. The reference ammeter is determined, where the angle is... The reference ammeter is used to characterize the relationship between the angle and the reference current; the load quality compensation constant is determined based on the real-time current, reference current, and speed compensation item within the comprehensive evaluation window; the comprehensive evaluation window is a preset time period after the start-up surge blanking period ends.

[0025] The second aspect of the invention simplifies the derivation of physical operators by comprehensively evaluating the effects of temperature, wear, and load in a single step, thereby reducing reliance on high-performance hardware and additional sensors and saving material costs.

[0026] In a third aspect, the present invention proposes an anti-pinch control method for an electric seat: an anti-pinch control method for an electric seat, the electric seat including a rotating part and a motor, the motor being used to drive the rotating part to rotate, the method including: when the transient collision energy is greater than the collision energy threshold and the current speed of the motor is less than the expected reference speed, the motor stops rotating or rotates in the reverse direction; the transient collision energy is determined by calculating the deviation energy accumulated when the transient current change rate exceeds the dynamic slope threshold within the integral observation window.

[0027] The third aspect of the present invention utilizes the rate of change of current for advanced prediction, which can capture weak features in the early stage of a collision and achieve damage-free perception of soft objects.

[0028] Preferably, the integral observation window is opened when the transient current change rate exceeds the dynamic slope threshold, thereby accurately locating the starting point of abnormal current fluctuations. Time-domain observation ensures that the system only performs integral calculations on energy waveforms with collision characteristics.

[0029] Preferably, the transient current change rate is determined by linearly fitting the current within the sliding window using the least squares method; the dynamic slope threshold is determined by... Determined; among them, For dynamic slope threshold, The baseline value is obtained through simulated impact experiments. From the current perspective The mechanical stiffness compensation factor is as follows: denoted as , where is the standard deviation of the ripple slope, and v is the current speed of the motor.

[0030] This invention uses linear fitting to filter high-frequency noise from brush commutation and dynamically adjusts the slope threshold based on angular stiffness and rotational speed to ensure accuracy in judgment under different motion states.

[0031] Preferred transient collision energy ;in, Let L be the transient current change rate measured in the i-th sampling, and L be the length of the integration observation window. The sampling period is defined as follows. This invention upgrades the single slope limit violation judgment to a feature energy accumulation judgment, effectively filtering out occasional instantaneous spike signals or electrostatic interference.

[0032] Preferably, the collision energy threshold Determined; among them, The baseline value is obtained through simulated impact experiments. From the current perspective The mechanical stiffness compensation factor is as follows: This is for speed compensation.

[0033] The benchmark value determined by the present invention through simulated impact experiments Scientific protection standards that conform to the elasticity of human tissue have been established, thereby improving the safety level of protection.

[0034] Preferably, the current speed of the motor being less than the expected reference speed includes: the deviation between the current speed of the motor and the expected reference speed being lower than the anomaly determination threshold; wherein, the expected reference speed is the sum of the command expected speed and the physical model correction amount.

[0035] This invention introduces speed feedback as a secondary confirmation based on energy determination, ensuring that obstacle avoidance action is only performed when real physical resistance is generated, thereby further enhancing anti-interference capability.

[0036] The fourth aspect of the present invention relates to an anti-pinch control device for an electric seat. The electric seat includes a rotating part and a motor, the motor being used to drive the rotating part to rotate. The device includes: a parameter acquisition module for acquiring the current current of the motor, the current speed, the current angle of the rotating part, and the ambient temperature; a storage module for storing an angle-reference ammeter, a mechanical wear self-learning factor, and a preset window length parameter; a logic calculation and judgment module for judging whether condition one is satisfied, condition one being determined by judging whether the current current of the motor is greater than a dynamic current threshold or whether the transient collision energy is greater than a collision energy threshold; and judging whether condition two is satisfied, condition two being determined by judging whether the current speed of the motor is less than a expected reference speed; and a drive control module for outputting a control signal to drive the motor to stop or reverse rotation when judgment conditions one and two are satisfied.

[0037] According to an embodiment of the present invention, the anti-pinch control system parameterizes and incorporates factors such as seat angle, load, temperature, and wear into the model, enabling the system to accurately distinguish real pinching events and significantly reduce the false alarm rate.

[0038] According to an embodiment of the present invention, the anti-pinch control system utilizes transient slope By combining an energy integrator, weak features in the early stages of a collision are captured, resulting in a faster response than traditional methods. Furthermore, the single slope exceeding limit is upgraded to feature energy accumulation judgment, filtering out instantaneous spike signals. A dynamic slope threshold is introduced, and an angle compensation factor is obtained by looking up a table. and speed compensation This allows the threshold to be dynamically adjusted according to changes in the equivalent stiffness of the sofa mechanism, maintaining optimal anti-pinch sensitivity at different angles and rotational speeds. Attached Figure Description

[0039] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope of the embodiments of the present invention.

[0040] Figure 1 This is an overall hardware logic block diagram of the anti-pinch system provided by the present invention;

[0041] Figure 2 The generation angle of Embodiment 1 of the present invention A schematic flowchart of a reference ammeter;

[0042] Figure 3 The wear factor of Embodiment 1 of the present invention Logic diagram of the self-learning update algorithm;

[0043] Figure 4 This is a flowchart of the software algorithm logic of Embodiment 1 of the present invention;

[0044] Figure 5 This is a flowchart of the software algorithm logic of Embodiment 2 of the present invention;

[0045] Figure 6 This is a flowchart of the software algorithm logic of Embodiment 3 of the present invention.

[0046] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0047] 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 embodiments of the present invention, and not all embodiments. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed embodiments of the present invention, but merely to illustrate selected embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the embodiments of the present invention.

[0048] The embodiments of the present invention will be described in detail below with reference to specific examples.

[0049] In recent years, the technology of using motor drive current for obstacle detection has gradually gained attention. The core physical basis of this approach is the direct proportionality between the output torque of a DC motor and its armature current. When the rotating part encounters an obstacle during operation, the motor load torque increases rapidly, causing a momentary surge in the drive current. This solution requires no modification to the seat's appearance, is simple to install and debug, and offers high stability.

[0050] However, in the actual implementation of current detection, several challenges need to be addressed due to the complexity of the mechanical structure of electric seats: During the process of the seat retracting and unfolding fully, it is affected by the combined effects of gravitational torque, changes in leverage ratio, and spring assistance. Therefore, its current reference curve is not a horizontal line, but a function that fluctuates dramatically with the angle and position. When the motor is running at high speed, the vibration generated by gear meshing will be superimposed on the current signal as a high-frequency AC component, easily triggering false alarms. The surge current of a DC motor at startup is much higher than the normal operating value; without targeted suppression, the system may fail to start normally. With increasing usage years, the lubrication of the seat's transmission components decreases, and the overall operating resistance will slowly drift, causing the preset fixed threshold to fail.

[0051] Figure 1 This is a general hardware logic block diagram of the anti-pinch system provided by the present invention. (See attached diagram.) Figure 1 This invention provides a transient sensing anti-pinch system based on adaptive physical modeling. Its hardware architecture mainly consists of a central processing unit, a current sampling and signal conditioning module, a position and speed feedback module, an H-bridge power drive module, an environmental perception and alarm auxiliary module, and a non-volatile storage module. The specific technical implementation and logical connection of each module are as follows:

[0052] The central processing unit (CPU) employs a high-performance 32-bit embedded processor with a built-in high-resolution multi-channel analog-to-digital converter (ADC) for current sampling. Its computing power supports high-frequency sampling processing, achieving a sampling frequency of 2kHz through direct memory access. The current sampling and signal conditioning module converts the virtual pressure from the motor into a digital signal that the processor can process. A precision sampling resistor is connected in series at the low end of the motor drive circuit, and the operational amplifier circuit amplifies and filters the voltage signal across the sampling resistor. After acquiring the raw ADC data, the processor runs a 20-point fast iterative median sorting filtering algorithm to effectively eliminate random pulse glitches generated by brush commutation. The output current I is used for subsequent model calculations. The position and speed feedback module integrates a dual-channel Hall encoder. The signal is directly connected to the processor's timer input capture pin or the encoder's dedicated interface. It calculates the real-time speed v and physical position angle by measuring the number of pulses per unit time or the time interval between pulses. The H-bridge power drive module controls the motor's operation. It supports pulse width modulation (PWM) control to adjust the seat's running speed. The hardware circuitry supports fast electronic braking logic, enabling immediate stopping of rotation or reverse braking upon triggering the anti-pinch protection. An environmental sensing and alarm-assisted temperature sensing module uses a built-in or external temperature sensor to acquire the ambient temperature (T) in real time, providing physical characteristic compensation parameters for the algorithm. The alarm device integrates a buzzer and other audible and visual signal devices, emitting an audible and visual warning signal to alert the user when the system enters the anti-pinch protection locking state. Non-volatile memory stores pre-calibrated angles. Reference ammeter and self-learning updated wear factor Physical model parameters, etc.

[0053] Example 1:

[0054] Figure 2 This is a schematic flowchart illustrating the generation of an angle-reference ammeter according to Embodiment 1 of the present invention. Angle The purpose of the reference ammeter is to eliminate nonlinear load interference caused by the angle of the electric seat. The rotating part described in this invention refers to the mechanical mechanism in an electric seat used to achieve functions such as backrest tilting and leg support height adjustment. Its specific forms include, but are not limited to, functional frames, scissor frames, linkage mechanisms, or telescopic supports. During the movement of the rotating part of the electric seat, its physical load is affected by the following factors: the larger the backrest angle, the greater the torque generated by the gravitational component with the angle. The variations in the extension ratios of the scissor-mounted frame at different stages result in inconsistent equivalent resistance at the motor end. Furthermore, some seat base frames are equipped with assist springs, whose elasticity releases at different positions, causing significant current troughs or peaks. To address this issue, this invention proposes establishing a position reference function. The specific calibration steps are as follows: Step 201: Trigger the no-load modeling mode: Trigger the no-load modeling mode when the seat comes off the production line or when the user resets the system; Step 202: Cyclic sampling: Make the rotating part perform a complete action from closing to unfolding and then to fully closing under no-load conditions; Step 203: Data recording: The processor records data every... The angle position is recorded as a current sample point; step 204: look up and store the table: generate a table containing the angle. The reference ammeter is stored in non-volatile memory. During actual operation, the processor retrieves the current angle value in real time using a lookup table. .

[0055] Figure 3 The wear factor of Embodiment 1 of the present invention The self-learning update algorithm logic diagram. After several years of use, electric seats may experience slow vertical drift due to the evaporation of lubricating oil in the transmission mechanism, dust accumulation, or minor deformation of the metal linkage structure. If the system always uses the fixed reference current from the factory... Older equipment will frequently trigger false alarms due to increased normal mechanical resistance, leading to decreased product usability. This invention introduces a self-learning algorithm for periodic updates. By utilizing a long-term observation window, resistance drift caused by mechanical structure aging is absorbed, thereby achieving consistent sensitivity throughout the product's entire lifecycle. (Combined with attached...) Figure 3 The self-learning update process in this embodiment consists of the following steps: Step 301: Counting trigger: The system maintains a non-volatile opening and closing counter. Only when the system has accumulated 50 complete opening and closing cycles will a wear factor calculation and update instruction be triggered; Step 302: Operating condition determination: Determine whether the update is performed under no-load conditions; Step 303: Data recording: During the update cycle, the processor records the filtered current sample points corresponding to each angle position of the entire stroke in real time. Step 304: Calculation of root mean square deviation: Calculate the no-load operating current and the original reference current during these cycles. root mean square deviation .

[0056]

[0057] Step 305: Determine if there is a continuous positive aging trend. Does it exceed the preset circuit background noise threshold? Step 306: If aging is determined, the system will adjust the wear factor. Increase the weight step size by a small increment and store it in memory. Step 307: If determined to be a temporary load or normal fluctuation, maintain the original value. This invention employs a micro-adjustment mechanism similar to neural network weight updates. Through this long-term integral accumulation, the safety threshold slowly increases as the seat ages. Wear factor let The entire unit is moved upwards slowly, thereby extending the product's maintenance-free period and preventing frequent false alarms from old seats.

[0058] Figure 4 This is a flowchart illustrating the software algorithm logic of Embodiment 1 of the present invention. (In conjunction with...) Figure 4 Each time a run command is received and the motor is started, the following specific steps are executed:

[0059] Start-up surge blanking 401: At the initial moment of motor start-up via pulse width modulation, the system avoids extremely high and unstable start-up surge current through a software blanking mechanism to prevent the algorithm from making misjudgments before the motor enters steady state.

[0060] Data Acquisition and Preprocessing 402: Acquire current I, real-time rotational speed v, and physical position at a fixed frequency. Ambient temperature T.

[0061] Load quality compensation Assessment 403: Due to the baseline model It is established under no-load conditions. In order to eliminate the interference of gravity distribution caused by different user weights on the anti-pinch judgment, the system introduces load mass compensation. Evaluation Mechanism. Immediately after the surge blanking phase ends, the central processing unit (processor) opens a load quality compensation evaluation window, preferably with a duration of 100ms-300ms. During this window, the seat rotation has not yet experienced significant physical displacement or approached a potential compression point within a safe range. The processor continuously acquires the current signal after 20-point fast iterative median sorting and filtering, and calculates the average steady-state operating current within this window. Calculate the current stable operating current and the corresponding angle reference current. Load compensation constant between .

[0062]

[0063] Once determined during the startup phase This value will be used as a global correction parameter for this specific running cycle: the system will obtain The load is superimposed in real-time onto the baseline envelope for the entire subsequent travel. This means that if the current user is heavier, the system will raise the overall threshold; conversely, if the load is lighter, the threshold will remain low. This calculation process is triggered every time the motor runs. This method ensures that even if a user of different weight is replaced during operation, or if items of different weights are placed on the seat, the system can accurately adapt to the load.

[0064] Real-time calculation of judgment threshold 404: The processor calculates the real-time threshold for the current time t. The calculation formula is as follows:

[0065]

[0066] in, based on Find the reference current using the angle-reference ammeter. .

[0067] in, This is a compensation mechanism to address speed fluctuations. The motor's operating speed directly affects the background noise of the current reading. As the speed increases, the high-frequency mechanical noise generated by the radial runout of the gearbox gears interferes with the current reference. Furthermore, the radial runout amplitude caused by the non-roundness of the gearbox gears increases with the speed, generating high-frequency mechanical noise that also interferes with the current reference. This embodiment establishes a linear compensation function:

[0068]

[0069] in, and The vibration coupling coefficient is obtained by fitting the data using the least squares method. To suppress noise generated by mechanical vibration, A 20-point fast iterative median sorting filter is used to output a stable current or velocity observation. This is achieved by weighted summation or averaging of the velocity characteristics within a small time window to counteract high-frequency noise caused by gear jitter.

[0070] Preferably, Temperature-based compensation mechanisms could also be considered. The viscosity of the grease in the mechanical transmission components of the electric seat is affected by ambient temperature. In low-temperature environments, the grease thickens, significantly increasing the inherent resistance of the mechanism and causing the current reference curve to drift. Compensation logic: Establish a temperature compensation operator. This scheme introduces a temperature compensation operator based on an improved Arrhenius equation. The processor uses its built-in temperature sensor to obtain the ambient temperature and adjusts the judgment threshold in real time.

[0071]

[0072] Among them, A (environmental impact factor) and The viscosity-temperature sensitivity coefficient is obtained by performing no-load opening and closing experiments at different temperatures, recording the operating current reference of the motor, and then using the least squares method to substitute the measured data into the equation for nonlinear fitting. R is the molar gas constant. These constants are pre-programmed into the non-volatile memory of each controller as factory configuration parameters.

[0073] Among them, static security redundancy Used to absorb circuit background noise and instantaneous power fluctuations, preventing false triggering. Its determination process is as follows: Background noise test: During the no-load modeling stage, record the peak-to-peak fluctuation coefficient of the current signal throughout its entire stroke. Based on the experimentally measured circuit thermal noise and brush commutation noise, The value range is usually set to 2-6 times the standard deviation of the filtered signal.

[0074] Step 405: Determine the real-time current. Whether it deviates from the dynamic threshold: If the real-time current deviates within a current deviation judgment window, Then determine the real-time current. Deviation from dynamic threshold. The current deviation judgment window is a time-based debouncing threshold set to filter out commutation sparks or momentary mechanical vibrations.

[0075] If the real-time current If the speed deviates from the dynamic threshold, proceed to step 406: determine if the rotational speed v is less than the expected reference speed. Relying solely on current to determine anti-pinch performance is insufficient. If the user simply changes their seating position, although the current I increases, if the rotational speed v remains stable within the expected range of PWM control, the system interprets it as an increase in load rather than clamping, thus not triggering a shutdown. When a true anti-pinch event occurs, due to a sudden increase in resistance, the motor's real-time rotational speed v will show a significant, unexpected decrease. When the system detects a sudden increase in current, if it also detects a rotational speed less than the expected reference speed, it confirms a physical collision and immediately triggers electronic braking and reversing actions. This unexpected negative change is not simply deceleration, but a mismatch between the actual physical feedback and the system's expected command. The determination formula is as follows:

[0076]

[0077] Actual measured speed It uses a dual-channel Hall encoder integrated on the motor for measurement. The command indicates the expected speed.

[0078] The implementation method is determined by the PWM duty cycle output by the processor. When the adjustment speed is manually changed, This is the corresponding preset value. Expected speed of the instruction. A first-order inertial filter or ramp function is introduced during the speed regulation phase to simulate the actual physical acceleration response curve of the motor and eliminate the time-domain lag between command and feedback. Anomaly detection threshold. Used to absorb circuit noise and instantaneous power fluctuations during normal operation, it is usually set to 3 to 6 times the standard deviation of the filtered signal.

[0079] Because the angle-adjusting motor is affected by gravitational torque, changes in the linkage lever ratio, and spring assistance during operation, the system needs to retrieve the normal velocity offset corresponding to the current angle using a lookup table method. (Physical model correction amount) The generated dynamic parameter table is calibrated using the following steps: During system initialization or production line calibration, the no-load modeling mode is triggered, and the seat is controlled to perform a complete opening and closing motion, utilizing a Hall encoder to... The rotational speed of each sample point is recorded at angular intervals, and the central processing unit calculates the velocity characteristics corresponding to each angle. The generated angle-velocity lookup table is stored in the storage module for real-time retrieval during operation.

[0080] If the system determines that the rotational speed v is less than the expected reference speed, that is, it determines that clamping or mechanical jamming has occurred, then proceed to step 407: emergency response and active obstacle avoidance. Once the anti-pinch judgment circuit (whether static or transient) outputs a trigger signal, the processor immediately executes at least one of the following standardized obstacle avoidance actions: (1) Emergency electronic braking: The system instructs the H-bridge circuit to short-circuit the two ends of the motor through the low-side MOSFET, using the induced current of the motor to generate a huge reverse braking torque. (2) Active pressure relief and retraction: After braking is completed, the processor controls the H-bridge circuit to switch polarity, instructing the motor to immediately rotate in the opposite direction for about 1.5 seconds at 50%-70% of the rated power, driving the rotating part to retract about 30mm-50mm. The purpose of this action is to release the stress at the physical compression point instantly. For example, when the tail of a pet is caught in the scissor rack, simply stopping the motor is not enough, the mechanical tension will still cause continuous compression, while active retraction provides an escape gap for the clamped object. (3) Alarm and status lock: The alarm device emits a continuous pulse sound signal to alert the user. The system enters a protection lock state, and will not continue to execute the original forward squeeze command unless the user manually presses the reverse button.

[0081] Example 2:

[0082] Figure 5 This is a flowchart illustrating the software algorithm logic of Embodiment 2 of the present invention. This embodiment is a low-cost implementation of Embodiment 1. Parts not described in detail are basically the same as in Embodiment 1. The following description focuses only on the improvements / differentiated steps in Embodiment 2; other repetitive parts will not be elaborated upon. Combined with... Figure 5 Each time a run command is received and the motor is started, the following specific steps are executed:

[0083] Data acquisition and preprocessing step 502: Acquire the motor's current I, real-time speed v, and physical position at a fixed frequency. .

[0084] Comprehensive Compensation Assessment 503: Immediately after the surge blanking phase ends, the central processing unit (processor) opens a comprehensive compensation assessment window, preferably with a duration of 100ms-300ms. During this window, the seat rotation has not yet experienced significant physical displacement or approached a potential compression point within a safe range. The processor continuously acquires the current signal after 20-point fast iterative median sorting and filtering, and calculates the average steady-state operating current within this window. Calculate the current stable operating current and the corresponding angle reference current. and speed compensation item The comprehensive compensation constant between .

[0085]

[0086] By using the comprehensive compensation constant Defined as In fact, the temperature compensation mechanism in Example 1 is used. Wear factors and load quality compensation All of these factors are incorporated into a comprehensive compensation constant. The comprehensive compensation constant is essentially a mapping of the total static error of the system at the initial moment of this operation. Since the ambient temperature, mechanical wear factors, and user load do not change abruptly during a single opening and closing process, all the aforementioned nonlinear disturbance terms can be converted into the comprehensive compensation constant by performing a single calculation within the comprehensive evaluation window. The advantage of Example 2 compared to Example 1 is that it simplifies the algorithm, eliminates complex physical operator derivations and long-period root mean square (RMS) self-learning calculations, greatly saving processor computing resources and storage space; it eliminates the need for built-in or external temperature sensors, reducing the system's bill of materials and avoiding system risks caused by sensor failure. However, the disadvantage is that this scheme is highly dependent on the stability of the comprehensive compensation evaluation window operation.

[0087] Real-time calculation of judgment threshold 504: The processor calculates the real-time threshold for the current time t. The calculation formula is as follows, without considering the temperature compensation mechanism. and wear factor :

[0088]

[0089] Example 3:

[0090] The transient slope in this embodiment The method can make advance predictions and is used to identify minute deformations of soft objects such as fingers or pet limbs at the moment of contact. It can make predictions in advance before the absolute value of the current increases, thus achieving harmless perception.

[0091] Figure 6 This is a flowchart illustrating the software algorithm logic of Embodiment 3 of the present invention. This embodiment is an additional implementation based on other embodiments. The parts not described in detail are basically the same as those in other embodiments. The following text only focuses on the improvements / differentiated steps in Embodiment 3, and the remaining repetitive parts will not be elaborated.

[0092] In the data acquisition and preprocessing stage 602, the current current value of the motor is obtained through the sampling resistor, and the Hall encoder pulse / frequency is used to calculate the motor speed v and the current displacement angle in real time. The problem in the current sampling stage is that the commutation sparks of the brushed DC motor and the H-bridge PWM chopping generate huge high-frequency glitches in the current loop. In calculus, differentiation greatly amplifies these high-frequency noises. To suppress this interference at its physical source, this embodiment uses a PWM center-aligned mode for analog-to-digital conversion trigger control: the advanced timer is configured in center-aligned counting mode for sampling time selection. In this mode, the ADC trigger signal is precisely configured to be generated when the counter reaches its overflow value (i.e., the midpoint of the PWM duty cycle). Based on the inductive physical characteristics of the motor windings, the current fluctuates in a triangular wave pattern within each PWM cycle. The current value at the conduction midpoint is closest to the average current value of that cycle, and the slope of the current change is most stable at this point. To further avoid the electromotive force fluctuations generated by the H-bridge drive circuit during dead-time switching and the local current oscillations caused by the diode freewheeling phase, this system introduces a microsecond-level trigger delay compensation parameter based on the timer midpoint trigger. This parameter ensures that the ADC's sampling and holding action avoids the transient oscillation region during the switching of power devices, accurately falling within the absolutely stable region of the current waveform. To achieve continuous monitoring of the transient rate of change of current, the system constructs a sliding window buffer queue in memory, and a first-in-first-out calculation queue of length L, preferably with 8-12 sampling points. After each ADC conversion, the system pushes the latest current sample value onto the stack and removes the oldest sample. Simultaneously, this set of current samples is compared with the current physical angle. Spatiotemporal correlation mapping is performed. Before proceeding to the next step of the calculation, median filtering or amplitude limiting filtering algorithms are applied to the sampled data to identify and remove pulse interference within the window.

[0093] Real-time calculation of the transient rate of change of motor current In stage 603, the traditional two-point difference derivative algorithm (i.e. Extremely sensitive to high-frequency glitches, even with a 20-point median filter, the output slope curve still exhibits severe jitter, easily triggering false alarms. This invention abandons the two-point difference method and introduces least squares (OLS) linear fitting within the time-domain feature observation window to estimate the current change rate in real time. The processor maintains a sliding window of length L (e.g., 8-12 sampling points), performs a first-order linear regression on the data points within the window, and the slope of the resulting line is the current linear regression result. .

[0094]

[0095] in, For the sampling time point, The mean over the window time. For sampling current, The average current within the window is used. This method smooths out the random high-frequency noise generated by brush commutation within the limits of processor resources, ensuring the accuracy of characteristic slope determination. The dynamic slope threshold is calculated in real time. In stage 604, considering the significant differences in the equivalent stiffness and vibration noise of the mechanical linkage under different operating speeds and deployment angles of the seat, this system establishes a dynamic slope threshold. .

[0096]

[0097] This embodiment was determined through the following experimental steps. A silicone module with a Shore hardness of 20 was selected as the standard test material to simulate the muscle elastic resistance of a child's finger or a pet's limb. In the impact test, the seat's rotating part was moved at its rated speed, and the aforementioned silicone module was placed along the path of motion. The first derivative of the current at the moment of contact was recorded using a high-speed oscilloscope. Take 80% to 90% of the minimum effective slope obtained in the simulation experiment as... A baseline value is used to ensure the sensitivity of the prediction. The system acquires the motor speed v in real time and obtains the current angle by looking up a table. Mechanical stiffness compensation factor As a speed compensation. The standard deviation of the ripple slope is obtained by recording the normal current fluctuations caused by gear meshing and motor brush commutation during a complete no-load opening and closing cycle of the seat, and by statistical calculation. Standard deviation of ripple slope This is usually stored as a constant in advance. As the motor speed v increases, the radial runout of the gearbox gears and the high-frequency electromagnetic noise caused by brush commutation will significantly increase. The system can adjust the threshold value in real time based on the rotational speed. Additionally, It must also be significantly higher than the electrical ripple slope of the system itself. The electrical ripple slope is obtained through a no-carrier ripple test, which records the normal current fluctuations generated by gear meshing and brush commutation during the complete opening and closing cycle of the seat, thereby obtaining the standard deviation of the ripple slope.

[0098] When the rate of change of current exceeds the dynamic slope threshold At that time, the energy integrator is activated to calculate the transient collision energy. and collision energy threshold Among them, the collision energy threshold It is calculated using the following formula:

[0099]

[0100] The following experimental method was used during the production line calibration or system initialization phase: A silicone module with a Shore hardness of 20 was used to simulate the muscle elastic resistance of a child's finger or a pet's limb. The seat was impacted with the silicone module at a rated speed, and the current slope from the moment of contact was recorded. Exceed The integrated energy distribution after integration:

[0101]

[0102] Preferably, 0.7-0.8 times the accumulated energy in the first 150ms of the experiment is selected as the collision energy trigger threshold. To prevent false triggering due to speed fluctuations, the system will... Real-time correction is performed. The speed compensation term for the collision energy threshold is also included. Typically, a linear function is used, and its determination method is similar to that of Example 1. That is, during the seat's no-load operation phase, the background integral energy generated by the system at different speeds v is recorded. The experimentally collected data points are fitted using the least squares method to obtain coefficients, which are then used as factory configuration parameters and pre-stored in non-volatile memory. Calibration is also completed during the production line calibration phase or the system initialization phase. During the system's initial no-load operation, the central processing unit records the reference current while simultaneously calculating the position at each angle using a sliding window. The mean rate of change of no-load current under the given conditions is denoted as the natural noise slope. Selecting a reference point: The angle position where the operation is most stable throughout the entire stroke is selected as the reference point, and the corresponding natural noise slope is denoted as... The compensation factor for each angle is obtained using the following formula and stored in non-volatile memory.

[0103]

[0104] To eliminate false triggering caused by single-point or sporadic electrostatic interference, this embodiment upgrades the single slope exceeding the limit judgment to a characteristic energy accumulation judgment. When cross At that time, the energy integrator is activated to calculate. Transient collision energy Defined as:

[0105]

[0106] in, Let L be the transient current change rate measured in the i-th sampling, and L be the length of the integration observation window. The sampling period is denoted as . It can be seen that the transient collision energy is the accumulation of the current slope deviation within the integration observation window. The transient collision energy within the window is calculated if and only if... The collision energy exceeds the set threshold. Only then does the system initially confirm it as a potential collision event. If... If the actual deceleration slope is much greater than the normal deceleration predicted by the physical model (i.e., there is an unexpected negative change), then a real physical compression is confirmed, and the system proceeds to step 607, immediately triggering electronic braking and active pressure relief retraction. If the current slope suddenly increases but the speed v remains stable (consistent with the expected PWM command), then it is determined to be an occasional electromagnetic interference or no-load vibration. The system only records the abnormal data for subsequent self-learning, but does not interrupt the motor operation.

[0107] This embodiment utilizes transient slope By combining an energy integrator, it is possible to capture weak features in the initial stage of a collision, identifying minute deformations at the moment of contact with soft objects, thus predicting and stopping the system before physical damage occurs. Furthermore, the simple slope exceeding limit judgment has been upgraded to feature energy accumulation judgment, filtering out instantaneous spike signals and eliminating false triggers. A dynamic slope threshold is introduced. The angle compensation factor is obtained by looking up a table. By calculating the rotational speed compensation term, the threshold can be dynamically adjusted to follow the changes in the equivalent stiffness of the sofa mechanism, maintaining optimal anti-pinch sensitivity at different angles and rotational speeds.

[0108] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the technical solutions of the present invention have been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for preventing pinching of an electric seat, characterized in that, The electric seat includes a rotating part and a motor, the motor being used to drive the rotating part to rotate. The method includes: when the current current of the motor is greater than the dynamic current threshold and the current speed of the motor is less than the expected reference speed, the motor stops rotating or rotates in the reverse direction.

2. The method according to claim 1, characterized in that, The dynamic current threshold is determined based on a reference current, a load mass compensation constant, a speed compensation term, an ambient temperature compensation, and a mechanical wear self-learning factor; wherein, the reference current is obtained by acquiring the current angle and finding the angle based on the current angle. The reference ammeter is determined, wherein the angle is... A reference ammeter is used to characterize the relationship between angle and reference current.

3. The method for preventing pinching of an electric seat according to claim 2, characterized in that, The angle The reference ammeter is determined by controlling the electric seat to perform a complete opening and closing action in an unloaded state during initialization or reset, and recording the current at 1-degree angular intervals.

4. The method according to claim 2, characterized in that, The dynamic current threshold is also determined based on static safety redundancy; the calculation formula for the dynamic current threshold is as follows: Among them, the The dynamic current threshold, the As the reference current, the The load mass compensation constant, the For the speed compensation term, the The mechanical wear self-learning factor, the For ambient temperature compensation, the For static safety redundancy.

5. The method according to claim 2 or 4, characterized in that, The load quality compensation constant is determined based on the real-time current, reference current, ambient temperature compensation, speed compensation, and mechanical wear self-learning factor within the load quality compensation evaluation window; the calculation formula for the load quality compensation constant is as follows: Among them, the For the real-time current within the load quality compensation assessment window, the The reference current within the load quality compensation assessment window, the For environmental temperature compensation within the load quality compensation assessment window, the The load quality compensation assessment window is a preset time period after the start-up surge blanking period ends.

6. The method according to claim 2 or 4, characterized in that, The mechanical wear self-learning factor is updated as the number of opening and closing cycles changes; the updating of the mechanical wear self-learning factor as the number of opening and closing cycles changes includes: When the preset number of complete opening and closing cycles is completed and the system is in an unloaded operating state, the real-time current at each angle position during the unloaded operating state is collected, and the root mean square deviation between the real-time current and the reference current is calculated. If the root mean square deviation exceeds the preset circuit background noise threshold, the updated mechanical wear self-learning factor is equal to the sum of the original mechanical wear self-learning factor and the preset micro-weight step size.

7. The method according to claim 1, characterized in that, The current speed of the motor being less than the expected reference speed includes: The deviation between the current speed of the motor and the expected reference speed is lower than the anomaly detection threshold; wherein, the expected reference speed is the sum of the command expected speed and the physical model correction amount.

8. A method for preventing pinching of an electric seat, characterized in that: The electric seat includes a rotating part and a motor, the motor being used to drive the rotating part to rotate; The method includes: when the current current of the motor is greater than the dynamic current threshold and the current speed of the motor is less than the expected reference speed, the motor stops rotating or rotates in the reverse direction; The dynamic current threshold is determined based on the reference current, the comprehensive compensation constant, and the speed compensation term. The reference current is obtained by acquiring the current angle and then looking up the angle based on the current angle. The reference ammeter is determined, wherein the angle is... A reference ammeter is used to characterize the relationship between angle and reference current; The comprehensive compensation constant is determined based on the real-time current, reference current, and speed compensation items within the comprehensive evaluation window; the comprehensive evaluation window is a preset time period after the start-up surge blanking period ends.

9. A method for preventing pinching of an electric seat, characterized in that, The electric seat includes a rotating part and a motor, the motor being used to drive the rotating part to rotate, the method comprising: When the transient collision energy is greater than the collision energy threshold and the current speed of the motor is less than the expected reference speed, the motor stops rotating or rotates in the opposite direction. The transient collision energy is determined by calculating the accumulated deviation energy when the transient current change rate exceeds the dynamic slope threshold within the integration observation window.

10. The method according to claim 9, characterized in that, The integral observation window opens when the transient current change rate exceeds the dynamic slope threshold; the transient current change rate is determined by linearly fitting the current within the sliding window using the least squares method; the dynamic slope threshold is determined by... Determined; wherein, the As a dynamic slope threshold, the The reference value is obtained through simulated impact experiments. From the current perspective The mechanical stiffness compensation factor below, the The ripple slope standard deviation is given by , and v is the current speed of the motor.

11. The method according to claim 9, characterized in that, The transient collision energy ; wherein, the Let L be the transient current change rate measured in the i-th sampling measurement, and L be the length of the integration observation window. The sampling period; the collision energy threshold Determined; wherein, the The reference value is obtained through simulated impact experiments. From the current perspective The mechanical stiffness compensation factor below, the This is for speed compensation.

12. An anti-pinch control device for an electric seat, characterized in that, The electric seat includes a rotating part and a motor, the motor being used to drive the rotating part to rotate; the device includes: The parameter acquisition module is used to acquire the current current, current speed, current angle of the rotating part, and ambient temperature of the motor. The storage module is used to store the angle-reference ammeter, the mechanical wear self-learning factor, and the preset window length parameter; The logic calculation and determination module is used to determine whether condition one is met. Condition one is determined by determining whether the current current of the motor is greater than the dynamic current threshold or whether the transient collision energy is greater than the collision energy threshold. The logic calculation and determination module is also used to determine whether condition two is met. Condition two is determined by determining whether the current speed of the motor is less than the expected reference speed. The drive control module is used to output a control signal to drive the motor to stop or reverse rotation when conditions one and two are met.

Images