Monitoring carpet based on single-chip microcomputer

Abstract

The invention discloses a monitorable carpet based on a single-chip microcomputer. The carpet is of a multi-layer composite structure and comprises a separable surface layer, a first safety protection layer, a flexible sensing layer, an intelligent control layer, a second safety protection layer and a bottom layer from top to bottom. The flexible sensing layer is integrated with a distributed thin film pressure sensor array and a millimeter wave radar module and is used for synchronously acquiring pressure and radar signals; the intelligent control layer comprises a main control module, a signal conditioning circuit, a self-powered module and a communication module. The sensor signals are fused and dynamically compensated through a collaborative calibration algorithm, the problems of signal crosstalk, temperature excursion and non-uniform time sequence are effectively solved, high-precision and high-reliability gait analysis and recognition of abnormal behaviors such as tumble are achieved, meanwhile, through self-power-supply and modular design, the sensor has the advantages of being low in power consumption and easy to maintain, and the application range is wide. The system is especially suitable for intelligent monitoring and safety early warning in a family environment.

Description

Technical Field

[0001] This invention belongs to the field of smart home and health monitoring technology, and in particular relates to a monitorable carpet based on a microcontroller. Background Technology

[0002] With the development of smart homes and health monitoring, monitorable carpets have become a research hotspot. Current technologies mainly rely on pressure sensors, infrared sensors, or millimeter-wave radar. Pressure sensors are the most widely used, but traditional silicon-based materials have inherent drawbacks: environmental temperature fluctuations introduce measurement errors, which existing compensation algorithms struggle to completely eliminate; more importantly, signal crosstalk is a significant problem in multi-sensor arrays, severely limiting measurement accuracy and reliability. While millimeter-wave radar technology avoids the susceptibility of optical sensors to dust and stray light interference, its ability to identify static targets when used alone is limited, making it difficult to achieve high-precision, continuous tracking of multiple targets.

[0003] Currently, although there have been attempts to combine pressure sensors with millimeter-wave radar, an effective collaborative calibration mechanism is lacking. The signals from heterogeneous sensors are difficult to unify in terms of timing and reference, making deep fusion impossible. This results in pressure signal drift, crosstalk, and radar identification errors not being corrected or compensated for by the system. Consequently, simple combinations not only fail to improve performance but may also increase the system's false alarm rate and instability due to data conflicts.

[0004] Therefore, there is an urgent need for a smart carpet solution that can address the problem of collaborative calibration and deep fusion of multimodal signals from both the hardware and algorithm levels. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a microcontroller-based monitorable carpet that can achieve high-precision, low-latency personnel monitoring, while also possessing good environmental adaptability, low power consumption, easy maintenance, and economy.

[0006] To achieve the above objectives, the present invention is implemented using the following technical solution: A microcontroller-based monitorable carpet, comprising, from top to bottom: a separable surface layer; a first security protection layer; a flexible sensing layer; an intelligent control layer; a second security protection layer; and a bottom layer; The detachable surface layer is a textile fabric layer, which is detachably connected to the first safety protection layer via Velcro. The first and second safety protection layers are made of EPDM rubber, and their edges are sealed and bonded to the adjacent layers through a vulcanization process. Temperature and humidity sensors are embedded in the layers, and an IP67 waterproof lead wire mechanism is provided. The flexible sensing layer includes a flexible substrate layer, a thin-film pressure sensor, and a millimeter-wave radar module. The intelligent control layer includes a main control module, a signal conditioning circuit, a self-powered module, and a communication module; The bottom layer is a wear-resistant and non-slip rubber layer.

[0007] Specifically, the flexible substrate of the flexible sensing layer is made using a three-dimensional weaving process, and its material is a piezoresistive material composed of graphene nanosheets and polyimide fibers in a 3:7 ratio.

[0008] Specifically, the thin-film pressure sensors are arranged in an 8×8 array, with a sensor diameter of 1.5cm and a spacing of 2cm between adjacent sensors, forming a flexible pressure sensing matrix with a thickness of 1mm.

[0009] Specifically, the main control module is a microcontroller, which is configured to: synchronously acquire and fuse signals from the pressure sensor array and the millimeter-wave radar module for human body status recognition; and control the power supply of the system to be cut off in response to signals from the temperature sensor and humidity sensor.

[0010] Specifically, the signal conditioning circuit includes a microcontroller U1, inverters U2A-U2D, and a differential amplifier circuit containing operational amplifiers U3A and U3B; The square wave signal output terminal of the microcontroller U1 is simultaneously connected to the input terminals of U2A and U2C in the inverter group; U2A and U2B are connected in sequence to form the first signal conditioning channel. Its output is amplified by the operational amplifier U3B and then connected to the first ADC acquisition port of the microcontroller U1. U2C and U2D are connected in sequence to form a second signal conditioning channel. Its output is amplified by the operational amplifier U3A and then connected to the second ADC acquisition port of the microcontroller U1. The input terminal of the U2D is connected to the sensing electrode of the carpet; The first and second signal conditioning channels are equipped with resistors, capacitors and diodes to form a signal shaping, filtering and amplification network.

[0011] Specifically, the self-powered module includes piezoelectric energy harvesting units distributed in the carpet layer structure, and an energy storage circuit electrically connected to the piezoelectric energy harvesting units; the piezoelectric energy harvesting units are used to convert the mechanical energy of stepping into electrical energy, and after processing by the energy storage circuit, power the system.

[0012] A microcontroller-based method for monitoring carpets includes the following steps: S1. Signal Acquisition: Pressure signals and radar signals are acquired synchronously through a pressure sensor and a millimeter-wave radar. S2. Signal filtering: The acquired raw analog signal is subjected to hardware RC filtering and software filtering based on Kalman filtering algorithm in sequence. The hardware RC filtering is configured to suppress environmental electromagnetic interference, and the software filtering is used to fuse pressure and radar signals and perform dynamic calibration. S3. Feature Extraction: Extract key features such as pressure distribution map, gait cycle, movement speed and trunk micro-motion from the filtered and calibrated pressure signal and radar signal; S4. Status Judgment: Abnormal status is identified based on the extracted key features; when a sudden change in pressure distribution, rapid drop in body height, and prolonged static state are detected, it is determined to be a fall event, and then the alarm information is uploaded to the cloud and the guardian's terminal through the communication module.

[0013] Furthermore, the communication module in the state determination in S4 supports WiFi 6 and Zigbee 3.0 dual protocol stacks as well as the Matter protocol.

[0014] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: 1. This invention solves the problem of unifying the timing and reference of pressure signals and radar signals through an innovative collaborative calibration algorithm, enabling the two sensors to complement each other (direct positioning by pressure sensing and micro-motion sensing by radar), significantly improving the accuracy of functions such as fall detection and intrusion detection, effectively identifying abnormal lingering behaviors such as "standing still for more than 30 seconds", issuing alarms in a timely manner, and providing an effective safety monitoring method for preventing sudden illnesses of elderly people at home.

[0015] 2. This invention effectively suppresses signal crosstalk and environmental electromagnetic interference through wavelet denoising and Kalman filtering, and dynamically compensates for temperature drift, ensuring the long-term stability and accuracy of monitoring data in different home environments.

[0016] 3. The modular layered structure and detachable design of this invention facilitate cleaning and maintenance. The introduction of the self-powered module reduces the dependence on external power sources. While achieving high performance, the overall cost is controlled through optimized sensor layout and algorithm design, making it a potential candidate for large-scale home adoption. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the structure of the monitorable carpet in Example 1.

[0018] Figure 2 This is a flowchart of the monitoring method for the monitorable carpet in Example 2. Detailed Implementation

[0019] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.

[0020] Example 1: A microcontroller-based carpet monitoring system This embodiment mainly describes the overall hardware structure of the carpet monitoring system of the present invention and the way it implements basic monitoring functions.

[0021] refer to Figure 1 The monitorable carpet provided in this embodiment has a structure that, from top to bottom, includes a separable surface layer (1), a first security protection layer (2), a flexible sensing layer (3), an intelligent control layer (4), a second security protection layer (5), and a bottom layer (6).

[0022] The detachable surface layer (1) is a textile fabric layer, which is detachably connected to the first safety protection layer 2 by Velcro, making it easy to clean separately.

[0023] The first safety protection layer (2) and the second safety protection layer (5) are made of EPDM rubber, and their edges are sealed and bonded to the adjacent layers through a vulcanization process. Temperature sensors and humidity sensors are embedded in the layers, and an IP67 waterproof lead wire mechanism is provided. The flexible sensing layer (3) includes a flexible substrate layer made of graphene / polyimide composite piezoresistive material. Multiple thin-film pressure sensors are distributed in an 8×8 array on this substrate layer, with an adjacent sensor spacing of 2 cm, forming a flexible pressure sensing matrix. Simultaneously, a millimeter-wave radar module is also integrated into this layer for detecting human movement and micro-motions. The intelligent control layer (4) includes a main control module, a signal conditioning circuit, a self-powered module, and a communication module; The bottom layer (6) is a wear-resistant and slip-resistant rubber layer.

[0024] Specifically, the main control module uses an STM32H743 microcontroller, which is electrically connected to the thin-film pressure sensor, millimeter-wave radar module, temperature sensor, and humidity sensor, forming a core hardware platform for realizing synchronous acquisition and collaborative processing of multimodal sensor signals.

[0025] Specifically, the piezoelectric material of the self-powered module is PZT piezoelectric ceramic with a thickness of 0.5mm. It can generate 3.2mJ of energy with a single step, and the carpet is self-powered by collecting piezoelectric energy.

[0026] Specifically, the communication module supports both WiFi and Zigbee protocol stacks for uploading data to the cloud or mobile terminals.

[0027] Specifically, the signal conditioning circuit serves as a dedicated interface between the main control module and the pressure sensor array in the flexible sensing layer (3). The following are some of the components required for the signal conditioning circuit: Microcontroller U1; Inverters U2A, U2B, U2C, U2D; Operational amplifier circuits: including the first operational amplifier circuit and the second operational amplifier circuit, the first operational amplifier circuit contains the inverting amplifier U3B, and the second operational amplifier circuit contains the inverting amplifier U3A, both of which are LM358; Resistors R1-R6; Capacitors C1-C3; Diodes D1-D2; Carpet and electrodes.

[0028] Specifically, the main connection method of the signal conditioning circuit is as follows: Connect the PD6 port of the microcontroller U1 to the input terminals of inverters U2A and U2C to output a square wave pulse signal. Then connect inverter U2A to inverter U2B through resistor R1, with diode D1 connected in parallel with R1. The input terminal of inverter U2B is grounded through capacitor C1. The output terminal of inverter U2B is connected to the inverting input of inverting amplifier U3B through resistors R2 and R3 in series. The connection point of R2 and R3 is grounded through filter capacitor C2. Connect inverter U2C to inverter U2D through resistor R4, with diode D2 connected in parallel with R4. The output terminal of inverter U2D is connected to the inverting input of inverting amplifier U3A through resistors R5 and R6 in series. The connection point of R5 and R6 is connected to the power supply terminal VCC through filter capacitor C3. The input terminal of inverter U2D is connected to the electrode of the carpet. The output of the first operational amplifier circuit is connected to the PC0 port of the microcontroller U1, the output of the second operational amplifier circuit is connected to the PC1 port of the microcontroller U1, and the PB3, PB4, and PB5 ports of the microcontroller U1 are used to output control commands.

[0029] Working principle: Upon program startup, the microcontroller U1 outputs a square wave pulse signal. One signal is transmitted sequentially through inverters U2A and U2B to the first operational amplifier circuit for amplification and processing before returning to microcontroller U1. The other signal is transmitted sequentially through inverters U2C and U2D to the second operational amplifier circuit for amplification and processing before also returning to microcontroller U1. Microcontroller U1 analyzes the voltage difference between the two returned signals and outputs control commands based on the analysis results, thereby monitoring people on the carpet, etc. The data is then uploaded to a mobile cloud platform so family members can monitor the home.

[0030] The data for Embodiment 1 of the present invention are shown in Table 1.

[0031] Table 1

[0032] Example 2: A microcontroller-based method for monitoring carpets. Example 2 elaborates on the core collaborative calibration algorithm flow of the microcontroller-based carpet monitoring system described in Example 1.

[0033] like Figure 2 As shown, the monitoring method specifically includes the following steps: S1: Signal Acquisition After power-on initialization, the system enters a low-power standby mode, periodically woken up by an independent timer to scan the sensor array of the flexible sensing layer (3) to monitor for triggering. When the pressure or radar sensor detects activity, the system immediately wakes up and synchronously collects pressure, radar signal, and ambient temperature and humidity data at a sampling frequency of 50kHz. If the ambient temperature is >60℃ or the humidity is >85%RH, the system cuts off power and enters safe sleep mode; if the environment is normal, the process continues. S2, Signal Filtering: The acquired raw analog signal is first filtered by a hardware RC filter in the signal conditioning circuit. A two-stage RC low-pass filter circuit and an operational amplifier are then used for amplification to initially suppress high-frequency noise and improve signal quality. Under safe environmental conditions, the main control module applies real-time dynamic calibration based on a Kalman filter algorithm to the hardware-conditioned signal. This dynamic calibration algorithm fuses pressure and radar signals and compensates for factors such as temperature drift in the pressure sensor, ensuring that the dynamic calibration error is less than 0.3%. S3, Feature Extraction: Using calibrated pressure sensor array data, precise positioning of the footsteps is achieved through row and column scanning. Simultaneously, key features are extracted from the calibrated pressure and radar signals, including pressure distribution patterns, gait cycles, movement speed, and trunk micro-movements. S4. Status Judgment: The main control module performs anomaly detection based on extracted features. When it detects a sudden change in pressure distribution, a rapid decrease in body height accompanied by a prolonged period of stillness, it determines it as a fall event. If the situation is normal, the system stores the data in local Flash memory, supporting 7-day cyclic storage before returning to the monitoring status. If a fall is detected, an interrupt is immediately triggered, data backtracking is performed, data from the 10 seconds prior to the incident is extracted, and an alarm command is generated. The alarm information is transmitted to the cloud and the guardian's terminal via the communication module, and can be linked with other smart devices to achieve cross-brand alarm functionality.

[0034] After completing the above steps, the system processing flow ends.

[0035] Specifically, the sampling frequency of the signal acquisition in S1 is 50kHz.

[0036] Specifically, the filtering window for the filtering process in S2 is 20ms.

[0037] Specifically, the collaborative calibration algorithm in S2 separates the sensor signal through orthogonal coding technology, uses time-division multiplexing technology, and performs data calibration based on the temperature compensation model y=0.987x+0.023T-0.015 (R²=0.996), so that the system maintains a measurement accuracy of ±1.2% in an environment of -10℃ to 60℃.

[0038] Specifically, the communication module in the state determination of S4 supports WiFi 6 and Zigbee 3.0 dual protocol stacks and supports the Matter protocol.

[0039] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A microcontroller-based monitorable carpet, characterized in that, include: The following layers are stacked sequentially from top to bottom: a separable top layer (1); a first safety protection layer (2); a flexible sensing layer (3); an intelligent control layer (4); a second safety protection layer (5); and a bottom layer (6). The detachable outer layer (1) is a textile fabric layer, which is detachably connected to the first safety protection layer (2) by Velcro. The first safety protection layer (2) and the second safety protection layer (5) are made of EPDM rubber, and their edges are sealed and bonded to the adjacent layers through a vulcanization process. Temperature sensors and humidity sensors are embedded in the layers, and an IP67 waterproof lead wire mechanism is provided. The flexible sensing layer (3) includes a flexible substrate layer, a thin-film pressure sensor, and a millimeter-wave radar module; The intelligent control layer (4) includes a main control module, a signal conditioning circuit, a self-powered module, and a communication module; The bottom layer (6) is a wear-resistant and slip-resistant rubber layer.

2. The microcontroller-based monitorable carpet according to claim 1, characterized in that, The flexible base layer of the flexible sensing layer (3) is made of a three-dimensional braiding process and is a piezoresistive material composed of graphene nanosheets and polyimide fibers in a ratio of 3:

7.

3. The monitorable carpet based on a single-chip microcomputer according to claim 1, characterized in that, The thin-film pressure sensors are arranged in an 8×8 array, with a sensor diameter of 1.5cm and a spacing of 2cm between adjacent sensors, forming a flexible pressure sensing matrix with a thickness of 1mm.

4. The microcontroller-based monitorable carpet according to claim 1, characterized in that, The main control module is a microcontroller, which is configured to: synchronously acquire and fuse signals from the pressure sensor array and the millimeter-wave radar module for human body status recognition; and control the power supply of the system to be cut off in response to signals from the temperature sensor and humidity sensor.

5. The microcontroller-based monitorable carpet according to claim 1, characterized in that, The signal conditioning circuit includes a microcontroller U1, inverters U2A-U2D, and a differential amplifier circuit containing operational amplifiers U3A and U3B. The square wave signal output terminal of the microcontroller U1 is simultaneously connected to the input terminals of U2A and U2C in the inverter group; U2A and U2B are connected in sequence to form the first signal conditioning channel. Its output is amplified by the operational amplifier U3B and then connected to the first ADC acquisition port of the microcontroller U1. U2C and U2D are connected in sequence to form a second signal conditioning channel. Its output is amplified by the operational amplifier U3A and then connected to the second ADC acquisition port of the microcontroller U1. The input terminal of the U2D is connected to the sensing electrode of the carpet; The first and second signal conditioning channels are equipped with resistors, capacitors and diodes to form a signal shaping, filtering and amplification network.

6. The microcontroller-based monitorable carpet according to claim 1, characterized in that, The self-powered module includes piezoelectric energy harvesting units distributed in the carpet layer structure, and an energy storage circuit electrically connected to the piezoelectric energy harvesting units; the piezoelectric energy harvesting units are used to convert the mechanical energy of stepping into electrical energy, and after being processed by the energy storage circuit, they supply power to the system.

7. A monitoring method for a monitorable carpet based on any one of claims 1-6, characterized in that, Includes the following steps: S1. Signal Acquisition: Pressure signals and radar signals are acquired synchronously through a pressure sensor and a millimeter-wave radar. S2. Signal filtering: The acquired raw analog signal is subjected to hardware RC filtering and software filtering based on Kalman filtering algorithm in sequence. The hardware RC filtering is configured to suppress environmental electromagnetic interference, and the software filtering is used to fuse pressure and radar signals and perform dynamic calibration. S3. Feature Extraction: Extract key features such as pressure distribution map, gait cycle, movement speed and trunk micro-motion from the filtered and calibrated pressure signal and radar signal; S4. Status Judgment: Abnormal status is identified based on the extracted key features; when a sudden change in pressure distribution, rapid drop in body height, and prolonged static state are detected, it is determined to be a fall event, and then the alarm information is uploaded to the cloud and the guardian's terminal through the communication module.

8. The monitoring method for carpets according to claim 7, characterized in that, The communication module in the state determination of S4 supports WiFi 6 and Zigbee 3.0 dual protocol stacks and Matter protocol.

Images