An extremely low frequency near field coupling type life state sensing device and method

Abstract

The application discloses a kind of based on coil impedance feature extraction's extremely low frequency near field coupling type life state perception device and method.The present application is not only suitable for door lock, switch panel and other static near field wake-up scene, and the stable extremely low frequency field is established in fixed position to realize human existence detection;More can be extended and applied to mobile robot platform.Because the extremely low frequency field has good penetration and anti-metal interference ability, the present application can be directly integrated on the surface of robot metal shell, without additional radar or vision sensor windowing, and the living obstacle avoidance and anti-collision in close range (such as <20cm) can be realized.(See figure 1)The characteristics of 'combination of static and dynamic' make the present scheme have cross-platform general perception ability.

Description

Technical Field

[0001] This invention relates to the field of non-contact life detection technology, specifically to an extremely low-frequency near-field coupling life state sensing device and method. The system and method can be integrated into devices such as robots, door handles, and switch panels. Background Technology

[0002] Existing life detection technologies (such as millimeter-wave radar and infrared thermal imaging) have limitations in scenarios with thick obstructions or requiring extremely high privacy protection. Millimeter-wave radar suffers severe attenuation when penetrating walls, and thermal imaging is easily affected by ambient temperature. Therefore, there is an urgent need for a life sensing solution based on the principle of electromagnetic near-field coupling, which is insensitive to obstructions and protects privacy. It should be noted that while existing capacitive proximity sensors (such as those published in CN111473809A and CN206060714U) and capacitive touch IPs (such as the Cypress CapSense series) can achieve human proximity detection, their detection principle is based on changes in capacitance between electrodes, which can only acquire a single capacitance characteristic, and the operating frequency band is usually in the kHz-MHz range. In addition, capacitive solutions are relatively sensitive to ambient humidity, temperature, and surface liquids, and cannot acquire impedance phase information. This invention employs coil impedance detection under extremely low frequency (1-30Hz) excitation, simultaneously extracting the impedance amplitude |Z| and phase difference θ as dual characteristic parameters. Detection is achieved through the specific coupling response of biological tissue to the electromagnetic field of this frequency band and the micro-motion modulation caused by vital signs. This invention differs fundamentally from capacitive detection in principle, detection quantity, and operating frequency band, and also possesses advantages such as strong resistance to environmental interference, ability to detect non-metallic living organisms, and applicability to metal shell surfaces. Summary of the Invention

[0003] The present invention aims to overcome the defects of the prior art and provide an extremely low frequency near-field coupling life state sensing device.

[0004] To achieve the above objectives, the present invention adopts the following technical solution: A very low frequency near-field coupled life state sensing device includes a microcontroller, a driving circuit, a transmitting coil, and an impedance sampling circuit; The impedance sampling circuit acquires the voltage and current changes of the coil through a series sampling resistor Rs; A signal extraction module and a determination module. Preferably, both the signal extraction module and the determination module are implemented through the analog-to-digital converter (ADC) interface and the running control software inside the microcontroller (MCU). Specifically, the signal extraction module is used to extract the ratio R / X of the real to imaginary parts of the impedance and the impedance phase angle θ; the determination module has a built-in adaptive baseline comparison algorithm, which specifically executes the following steps: 1) Calculate the real-time feature parameters \Delta |Z| and \Delta \theta within the current sampling period, where \Delta |Z| = |Z_{current}| - |Z_{baseline}|, and \Delta \theta = \theta_{current} -\theta_{baseline}. 2) Compare the calculated feature parameters with the preset threshold range; if \Delta |Z| and \Delta\theta simultaneously satisfy the preset joint judgment condition (\Delta |Z|>Z_{threshold} and |\Delta \theta|>\theta_{threshold}), then output a high-level signal to determine that a living organism exists.

[0005] "Through extensive simulation experiments and fitting of measured data, this invention has determined a set of excitation frequency combinations that have the best signal-to-noise ratio and anti-interference capability in a specific extremely low frequency band (such as 1-30Hz), for example [12.5Hz, 25Hz, 50Hz]. At the same time, the optimal weighting coefficients [w1, w2, w3] for each frequency point have been determined." The microcontroller generates an extremely low frequency excitation signal from 1Hz to 30Hz; the driving circuit converts the signal into a driving current; and the transmitting coil generates an alternating magnetic field in the near field. The impedance sampling circuit obtains the coil current by measuring the voltage drop across the series resistor Rs, and calculates the impedance amplitude |Z| and phase difference θ by combining the coil voltage. The microcontroller compares the impedance characteristic indicator value with a preset baseline to determine the presence and micro-motion state of the living organism.

[0006] In another specific application scenario, the sensing device described in this invention can be integrated into a mobile robot or a companion robot.

[0007] This invention can be widely applied to door handles, switch panels, and various electronic devices requiring close-range wake-up. In particular, due to the excellent anti-metal interference capability of the extremely low-frequency near-field coupling effect, this invention is also applicable to robot shell surfaces to achieve close-range live obstacle avoidance and human-machine interaction. By combining multi-frequency excitation with adaptive baseline comparison, this solution significantly improves detection stability in complex environments while meeting extremely low power consumption requirements.

[0008] Specifically, the transmitting coil and impedance sampling circuit are installed inside the chassis or torso of the robot.

[0009] When the robot is in a static standby or mobile cruising state, the system continuously runs an excitation sampling process in the background. Specifically, the adaptive baseline comparison refers to the microcontroller continuously collecting impedance amplitude and phase data for a preset duration (e.g., 30 seconds) during the system initialization phase or in a static environment without living interference, and calculating their statistical average values ​​as the initial baselines |Z_{baseline}| and \theta_{baseline}. During system operation, if a slow and continuous unidirectional drift of the feature parameters is detected (usually caused by slow changes in ambient temperature), the system will automatically trigger a baseline update mechanism to compensate for the baseline shift caused by slow environmental changes, thereby reducing the false alarm rate of the system. Specifically, the excitation sampling process includes the following steps: S1: The microcontroller outputs a sinusoidal excitation signal of a specific frequency (such as 12.5Hz) through the DAC (digital-to-analog converter) interface; S2: The drive circuit receives the signal and amplifies it into an AC current sufficient to drive the transmitting coil; S3: The impedance sampling circuit acquires the voltage drop across the series resistor R_s in real time to obtain the coil current I, and simultaneously acquires the voltage V across the coil. S4: The microcontroller performs digital filtering (such as moving average filtering) on ​​the acquired voltage and current signals to eliminate high-frequency noise, and then calculates the current impedance characteristic index values ​​|Z| and θ, preparing to enter the judgment stage.

[0010] When the robot approaches a user (such as an elderly person, a child, or a pet) within a preset distance range, the values ​​of |Z| and θ change. Based on this, the robot determines that there is a living being in front of it and triggers the corresponding interaction logic (such as slowing down to avoid obstacles, stopping movement, or issuing a voice greeting).

[0011] By combining ultra-low frequency sensing technology with robots, the problem of traditional visual sensors failing in privacy protection and dark environments has been solved.

[0012] In further system-level applications, the ultra-low frequency near-field coupled life presence sensing device of the present invention It can serve as a "bottom-level trigger node" in intelligent sensing networks, working collaboratively with other high-dimensional sensing modalities to build... Hierarchical sensing system.

[0013] Specifically, the device can be triggered or woken up after determining the presence of a living being and outputting an indication signal. The second sensing module performs refined detection. This second sensing module may include, but is not limited to, millimeter-wave radar. (For ranging and micro-motion tracking), infrared thermal imaging module (for body temperature distribution scanning), ultrasonic sensor Devices (for medium interface detection) or photoelectric volumetric plethysmography (PPG) modules (for refined blood oxygen / heart rate monitoring).

[0014] For example, in a specific linkage logic, the ultra-low frequency sensing device is responsible for all-weather low-power monitoring. The system detects whether life exists in the area; once the detection is successful, the main control unit activates the millimeter-wave radar. The module can locate or identify the pose of the target; it can also wake up the PPG module and guide it to align with the target area. To obtain more accurate physiological parameters (such as blood glucose trends, blood oxygen saturation, etc.). Furthermore, the wake-up logic also includes a priority arbitration mechanism. When the ultra-low frequency sensing device triggers a wake-up signal, the system's main control unit first determines whether the current high-dimensional sensor (such as the PPG module) is in a ready state; if it is in a deep sleep state, it first sends a wake-up command and waits for its initialization to complete, then sends an alignment command via the I2C or SPI bus to ensure synchronous acquisition of multimodal data in the time dimension.

[0015] Through this layered architecture of "coarse judgment of extremely low frequency near-field + detailed exploration of high-dimensional modes", the problem of transmission is solved. Overcoming the shortcomings of single sensors in terms of privacy, power consumption, and penetration, while paving the way for future integration of more complex biometric features. The recognition algorithm (such as a non-invasive blood glucose estimation model based on multimodal data fusion) has reserved system interfaces and Data channel. Those skilled in the art will understand that the core of this invention lies in providing a stable and reliable "presence". The trigger signal, specific upper-level algorithms, and application scenarios can be flexibly configured according to actual needs. Attached Figure Description Figure 1 is a schematic diagram of the structure of an ultra-low frequency near-field coupled life state sensing device of the present invention.

[0016] Figure 1 The numbers in the text are as follows: (100) Extremely low frequency excitation source (102) Transmit / near field coil (including sampling resistor Rs) (104) Differential sampling and amplification module (106) Processing and judgment module (calculate |Z| / θ, baseline comparison) (108) Feedback / reference baseline reference S presence indication output.

Claims

1. A very low frequency near-field coupled life state sensing device, characterized in that, include: Microcontrollers are used to generate extremely low frequency excitation signals in the range of 1 Hz to 30 Hz. A driving circuit, connected to the output terminal of the microcontroller, is used to convert the excitation signal into a driving current; A transmitting coil, connected to the driving circuit, is used to generate an alternating electromagnetic field in the near field range; An impedance sampling circuit is used to collect the voltage across the transmitting coil or the current flowing through the transmitting coil in order to obtain the impedance change that characterizes the electromagnetic field coupling state. A signal extraction module, connected to the impedance sampling circuit, is used to extract characteristic indicator values ​​based on the impedance change. The determination module, connected to the signal extraction module, is used to compare the feature indicator values ​​with a preset baseline to determine the existence status of the living organism.

2. The apparatus according to claim 1, characterized in that, The impedance sampling circuit is specifically used to collect the voltage drop across the sampling resistor connected in series in the transmitting coil circuit.

3. The apparatus according to claim 1, characterized in that, The characteristic indicators include the magnitude change or phase shift of the impedance.

4. The apparatus according to claim 1, characterized in that, The determination module is also used to further distinguish the micro-movement state of the organism based on the fluctuation frequency of the characteristic indicator value.

Images