Intelligent clothes airing and storing system based on multi-modal perception

Abstract

This invention provides an intelligent clothes drying and storage system based on multimodal perception, including a control module, a sensing module, an execution module, and a communication module. The system achieves clothing sorting, drying, and storage through the synergy of multi-sensor fusion, a clothing recognition model, and automatic actuators. Based on data fusion and intelligent decision-making, the system determines whether the current environment is suitable for drying clothes and uses mechanisms such as motors, slide rails, circulating chains, or steel wire ropes to complete the drying and automatic storage of clothes. The system integrates IoT control functions, supporting remote control via Wi-Fi, enabling intelligent clothes drying and storage, and improving the level of home intelligence and the user's quality of life.

Description

Technical Field

[0001] This invention relates to the field of smart home technology, and in particular to a smart clothes drying and storage system based on multimodal perception. Background Technology

[0002] Clothes racks, as everyday tools for drying and airing clothes, are widely used in people's daily lives. With the development of technology and the improvement of people's living standards, clothes racks have also evolved from traditional fixed racks to today's intelligent clothes drying systems with multiple functions. However, existing clothes drying systems receive relatively fixed amounts of sunlight, which can easily lead to either excessive or insufficient sunlight exposure. Clothes are difficult to dry on rainy days and are prone to bacterial growth. In addition, existing systems still rely on manual drying and storage, which is inconvenient for the elderly, children, or those with weaker physical abilities.

[0003] Therefore, given the low level of intelligence of existing clothes drying racks, insufficient environmental / clothes status sensing functions, the need for manual collection of clothes and judgment of dryness / wetness, and the failure to effectively reduce manpower input, there is an urgent need to design a more intelligent clothes drying rack. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide an intelligent clothes drying and storage system based on multimodal perception.

[0005] The technical solution adopted in this invention is: A smart clothes drying and storage system based on multimodal sensing includes a control module, a sensing module, an execution module, and a communication module, wherein... The control module is used for program execution and module control, including receiving data, executing intelligent decisions, and sending control commands. The sensing module, electrically connected to the control module, is used to collect real-time data on the clothes drying environment and the status of the clothes, and to identify the type of clothes. The sensing module includes a clothes drying environment monitoring module and a clothes type and dry / wet identification module, wherein: The clothes drying environment monitoring module consists of sensors for monitoring various indoor and outdoor environmental parameters. The sensors used for monitoring the environment are all commercially available devices for sensing environmental parameters such as wind force, temperature, humidity, and rain. The clothes drying environment monitoring module is connected to the control module by lines or signals, transmits the various environmental data collected by the sensors to the control module, and compares them with the built-in thresholds of the control module to determine whether the current environment is suitable for drying clothes. The thresholds can be set by customization or can be adaptively changed according to the current season and location. The clothing category and wet / dry identification module consists of a camera, a humidity probe, and a gravity hook. The camera, humidity probe, and gravity hook are connected to the control module via wiring or signal. The camera is used to capture images of the clothing and hangers. During use, the camera can be suspended from the ceiling and aligned with the rotating wheel. When the wheel carries the clothing past the camera, it ensures a clear image of the current clothing while minimizing the entry of preceding or following clothing items into the camera's field of view. The captured image information is then transmitted to the control module, which identifies the clothing type based on a clothing neural network recognition model. The humidity probe is used to sweep across the surface of the clothing; when the clothing comes into contact with the fixed humidity probe, the probe reports the change in humidity of the clothing to the control module. The gravity hook is fixed to the clothesline and has a hook for hanging clothes hangers. The hook consists of a pressure sensor or spring that can react to gravity and a mechanical housing, thereby obtaining the gravity change of the clothes as water evaporates during the drying process and feeding it back to the control module. If a pressure sensor is used, it can be used with a Bluetooth module to transmit gravity data to the control module. If a spring is used, a scale and pointer can be brought out. The gravity change data is identified and obtained by the control module based on the image information captured by the camera. The control module also records the drying time of the clothes on each hook based on the camera. When the drying time reaches a preset threshold, it can be determined that the clothes are dry. Thus, by combining the three dimensions of humidity, gravity and time, a high accuracy judgment of the dryness and wetness of the clothes can be achieved. The execution module includes a rotary motor, two or more rotating wheels, a circulating clothesline, an electric telescopic rod, a swing slide rail, several branch slide rails, and a clothing storage cabinet. The rotary motor is connected to and controlled by the control module. The rotary motor is connected to one of the rotating wheels as the driving wheel. The circulating clothesline is tensioned on the rotating wheels. The rotary motor drives the connected driving wheel to rotate, and then the belt drive of the circulating clothesline drives the other driven wheels to rotate. During use, the rotating wheels can be suspended from the ceiling. The hooks of several gravity hooks are fixed at intervals to the circulating clothesline. During use, clothes are placed on regular hangers, and the hangers are hooked onto the gravity hooks. Under the control of the control module, the rotary motor drives the clothes on the gravity hook to rotate along the circulating clothesline, so that the clothes are evenly exposed to wind and sunlight, improving drying efficiency. The electric telescopic rod is fixed to the side of the rotating wheel and is fixed after the posture and position are adjusted so that when the electric telescopic rod is pushed out diagonally upward, it can enter the hook of the clothes hanger, but not the hook of the gravity hook. The electric telescopic rod extends and pushes the clothes hanger with clothes, so that the clothes hanger automatically disengages from the hook that continues to rotate. One end of the swing slide rail is fixed to the end of the electric telescopic rod, and the other end swings to connect with one of the branch slide rails as needed. The ends of each branch slide rail are set in the corresponding clothes storage cabinet. The communication module, electrically connected to the control module, enables remote system control and data acquisition, facilitating data interaction between the control module and the remote control terminal. It supports remote operation and can utilize modules with network communication capabilities, such as commercial network cards, SIM card modules, or Wi-Fi modules. If a Wi-Fi module is used, it can connect to a home Wi-Fi router or hotspot under the control of the control module, access the network to log in to a cloud platform (such as Onenet or Huawei Cloud), obtain cloud platform commands, upload data to the cloud platform, and access the cloud platform via a mobile application to obtain system data and operate the clothes drying system, thereby realizing IoT functionality. The mobile application should include, but is not limited to, the following functional modules: 1. Parameter status display: displaying current environmental monitoring data such as temperature and humidity, the quantity and type of clothes being dried, and their dryness / wetness status; 2. Clothes drying and storage operation: controlling the system to store specified or all dried clothes, stop or adjust the rotation speed, etc.; 3. System built-in parameter settings: allowing modification of system parameter thresholds. The mobile application can be developed based on a mobile operating system platform, or it can be developed as a WeChat mini-program with related application functions using WeChat developer tools.

[0006] Preferably, in the above-mentioned intelligent clothes drying and storage system based on multimodal perception, the control module is an embedded system chip or a microcomputer, the microcomputer is a Raspberry Pi or Arduino development board, and the embedded system chip can be an STM32 series or RK3588 series microcontroller.

[0007] Preferably, the above-mentioned intelligent clothes drying and storage system based on multimodal perception monitors the clothes drying environment based on multimodal fusion of multiple sensors, wherein the multiple sensors are various environmental monitoring sensors such as commercial temperature and humidity sensors, wind sensors, rain sensors, pressure sensors and / or dust sensors.

[0008] Preferably, the above-mentioned intelligent clothes drying and storage system based on multimodal perception uses a sensor group to sense drying conditions. When the control module analyzes and determines that drying is not suitable based on the current sensing environment parameters, it sends an early warning signal and controls the execution module to trigger the clothes storage action.

[0009] Preferably, the above-mentioned intelligent clothes drying and storage system based on multimodal perception includes a neural network recognition model pre-trained with images of tagged clothing in the control module. The camera of the perception module collects images and inputs them into the neural network recognition model trained with sample clothing images. The neural network recognition model consists of a convolutional neural network that can extract image information and a fully connected layer network that can output categories. It can identify the collected images with high accuracy and determine the type of clothing.

[0010] Preferably, in the above-mentioned intelligent clothes drying and storage system based on multimodal perception, the gravity hook consists of a hook body capable of suspending clothes hangers and a pressure sensor or spring capable of responding to gravity. The pressure sensor or spring is installed inside the hook body. During use, the gravity hook is fixed to the clothesline. If a pressure sensor is used, the gravity hook uses the pressure sensor to obtain the change in gravity of the clothes as water evaporates during the drying process and feeds it back to the control module. When a pressure sensor is used, it can be used in conjunction with a Bluetooth module to transmit gravity data to the control module. If a spring is used, a scale and pointer can be brought out, and the data on the change in gravity can be identified and obtained by the control module based on the image information collected by the camera.

[0011] Preferably, the above-mentioned intelligent clothes drying and storage system based on multimodal perception performs multimodal fusion dry and wet state detection. The multimodal fusion dry and wet state detection includes setting a gravity sensor on the clothes hanging assembly to determine the dry and wet state of the clothes by detecting changes in the weight of the clothes; the clothes passing over a humidity probe to measure the humidity and compare it with a reference humidity to determine the dry and wet state; and a camera recording the type of clothes and the drying time, and comparing it with an experienced reference threshold to confirm the dry and wet state.

[0012] Preferably, in the above-mentioned intelligent clothes drying and storage system based on multimodal perception, the circulating clothesline is a closed-loop chain or a steel wire rope.

[0013] Preferably, in the above-mentioned intelligent clothes drying and storage system based on multimodal perception, the electric telescopic rod is fixed to the side of the rotating wheel, extending obliquely upwards to lift the clothes hanger, causing it to detach from the gravity hook fixed to the clothesline. The clothes hanger slides along the electric telescopic rod under the action of gravity to the swing rail connected to the rear of the electric telescopic rod. One end of the swing rail is fixed to the end of the electric telescopic rod, and the other end swings as needed to connect movably with one of the branch rails. The ends of each branch rail are placed in the corresponding clothing storage cabinet. A swingable servo motor is fixed on the swing rail. The servo motor is connected to the control module. The control module controls the servo motor to swing according to the type and dryness of the clothes, switching the swing rail to the branch rail with different endpoints for connection, so that the clothes hanger with clothes slides into the designated wardrobe under the action of gravity, ultimately completing the closed-loop automated "drying and storage integration" process from drying to storage. During use, the electric telescopic rod, swing slide rail, and branch slide rail can be suspended and fixed to the ceiling, or fixed to the floor or wall support frame. The specific position can be adjusted according to the position of the gravity hook fixed to the clothesline.

[0014] Preferably, in the above-mentioned intelligent clothes drying and storage system based on multimodal perception, the remote control terminal is a smartphone or tablet computer, which can realize the functions of querying drying status, sending drying instructions, triggering storage instructions, and setting system parameters through a dedicated APP. The control module feeds back the execution status to the remote control terminal in real time through the Wi-Fi communication unit.

[0015] The beneficial effects of this invention are: The aforementioned intelligent clothes drying and storage system based on multimodal perception is an intelligent clothes drying system grounded in the fields of intelligent control and mechanical transmission, as well as sensor technology and data fusion. Through multi-sensor fusion, clothing recognition models, and automatic actuators, it achieves the drying, sorting, and storage of clothes, and supports IoT control. The system uses a microcontroller or microcomputer as its core control module, integrating sensors for temperature, humidity, wind, rain, and gravity to perceive external weather conditions and the dryness / wetness of clothing in real time. Simultaneously, it incorporates cameras and convolutional neural networks to accurately identify clothing categories. Through data fusion and intelligent decision-making, the system determines whether drying is suitable and, using mechanisms such as motors, slide rails, circulating chains, or steel cables, completes the natural drying and automatic storage of clothes. It also supports remote control via Wi-Fi, achieving intelligent clothes drying and storage, improving the level of home intelligence and the quality of life for users. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the control method for the intelligent clothes drying and storage system based on multimodal perception described in this invention.

[0017] Figure 2 This is a schematic diagram of the intelligent clothes drying and storage system based on multimodal perception described in this invention. In the diagram, 1-control module, 2-sensor, 3-camera, 4-humidity probe, 5-gravity hook, 6-rotor, 7-circulating clothesline, 8-electric telescopic rod, 9-swinging slide rail, 10-branch slide rail, 11-clothes storage cabinet, 12-cloud platform, 13-remote control terminal.

[0018] Figure 3 This is the clothing recognition neural network model of the intelligent clothes drying and storage system based on multimodal perception described in this invention.

[0019] Figure 4 This is a schematic diagram of the gravity hook of the intelligent clothes drying and storage system based on multimodal perception described in this invention. In the diagram, 5-gravity hook, 5-1-spring, 5-2-measuring scale, and 5-3-hook body. Detailed Implementation

[0020] To enable those skilled in the art to more clearly understand the technical solution of the present invention, the technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0021] Example 1 like Figure 1-4 As shown, the intelligent clothes drying and storage system based on multimodal perception includes a control module, a sensing module, an execution module, and a communication module, wherein... The control module 1 is an embedded system chip or microcomputer used to store and run programs. It is responsible for data reception, fusion analysis, decision-making and instruction issuance, and coordinating the collaborative operation of various modules. For example, it receives data, executes intelligent decisions and sends control instructions. To save costs and improve integration, the STM32 series embedded chip system can be preferred. The sensing module, electrically connected to the control module, is used to collect real-time data on the clothes drying environment and the status of the clothes, and to identify the type of clothes. The sensing module includes a clothes drying environment monitoring module and a clothes type and dry / wet identification module, wherein: The clothes-drying environment monitoring module consists of sensors 2 used to monitor various indoor and outdoor environmental parameters. These sensors are commercially available devices for sensing environmental parameters such as wind, temperature, humidity, and rain, including temperature and humidity sensors, wind sensors, rain sensors, and dust sensors. Outdoor sensors are fixed outside the window to acquire multiple parameters of the current outdoor clothes-drying environment; indoor sensors can be fixed to the ceiling or wall to monitor the indoor environment. Each sensor is electrically connected to the control module, transmitting the collected environmental data to the control module and comparing it with built-in thresholds to determine if the current environment is suitable for drying clothes and whether to trigger the clothes-collecting action. The thresholds can be user-defined or adaptively changed based on the current season and location. The clothing category and wet / dry identification module consists of a camera 3, a humidity probe 4, and a gravity hook 5. The camera, humidity probe, and gravity hook are connected to the control module via wiring or signal. The camera is used to capture images of clothing and hangers. During use, the camera can be suspended from the ceiling, facing the rotating wheel. When the wheel moves clothing past the camera's shooting position, it ensures a clear image of the current clothing and tries to prevent the preceding or following clothing from entering the camera's field of view. The captured image information is then transmitted to the control module. The control module has a built-in neural network recognition model pre-trained with images of tagged clothing. This neural network recognition model consists of a convolutional neural network for image information extraction and a fully connected layer network for outputting categories, enabling it to identify the type of clothing in the captured images. The gravity hook is fixed to the clothesline during use. It consists of a hook body for hanging clothes hangers and a pressure sensor or spring for detecting gravity. The pressure sensor or spring is located within the hook body. If a pressure sensor is used, the gravity hook uses it to obtain the change in gravity of the clothes as moisture evaporates during drying and feeds this data back to the control module. When using a pressure sensor, a Bluetooth module can be used to transmit gravity data to the control module. If a spring is used, a scale and pointer can be extended, and the gravity change data can be identified and obtained by the control module based on image information captured by the camera. The humidity probe is used to pass over the clothes; that is, when the clothes move with the transmission mechanism and come into contact with the fixed humidity probe, the humidity probe feeds back the change in humidity of the clothes to the control module. Simultaneously, the control module records the drying time of the clothes on each hook based on the camera. When the drying time reaches a preset drying time threshold, the clothes are considered dry. By integrating information from gravity, humidity, and time, a high accuracy rate in determining the dryness or wetness of the clothes is achieved.

[0022] The execution module includes a rotary motor, two rotating wheels 6, a circulating clothesline 7, an electric telescopic rod 8, a swing slide rail 9, multiple branch slide rails 10, and a clothing storage cabinet 11. The rotary motor is connected to the control module and controlled by the control module. The rotary motor is rotatably connected to one of the rotating wheels. The circulating clothesline is a steel wire rope (or a closed-loop chain), tensioned at the edges of the two rotating wheels. The rotary motor drives the connected rotating wheels to rotate, and then the belt drive of the circulating clothesline drives the other rotating wheels to rotate. During use, the rotating wheel is suspended from the ceiling, and the hook bodies of several gravity hooks are fixed at intervals on the circulating clothesline. During use, clothes are placed on regular hangers, which are hooked onto the gravity hooks. Under the control of the control module, the rotating motor is started. As the rotating motor moves, the clothes on the gravity hooks rotate together with the circulating clothesline, realizing periodic rotation of positions, so that the clothes are evenly exposed to wind and sunlight, improving the overall drying efficiency of the clothes. The electric telescopic rod is fixed to the side of the rotating wheel. After the posture and position are adjusted, it is fixed so that when the electric telescopic rod is pushed out diagonally upward, it can enter the hook of the hanger, but not the hook of the gravity hook. The electric telescopic rod extends and pushes the hanger with clothes, so that it is detached from the gravity hook fixed on the clothesline. The hanger slides along the electric telescopic rod under the action of gravity to the swing slide rail connected to the rear of the electric telescopic rod. One end of the swing slide rail is fixed to the end of the electric telescopic rod, and the other end swings to connect to one of the branch slide rails as needed. The ends of each branch slide rail are set in the corresponding clothes storage cabinet. A swingable servo motor is fixed to the swing rail. The servo motor is connected to the control module. The control module controls the servo motor to swing according to the type and dryness of the clothes, switching the swing rail to different branch rails for connection. This allows the clothes hanger with clothes to slide into the designated wardrobe under gravity, thus completing a closed-loop automated "drying and storing" process. During use, the electric telescopic rod, swing rail, and branch rail can be suspended upside down from the ceiling, or fixed to the floor or wall support frame. The specific position can be adjusted according to the position of the gravity hook fixed to the clothesline. The communication module, electrically connected to the control module, enables remote system control and data acquisition. It facilitates data interaction between the control module and the remote control terminal 13 (smartphone or tablet), supporting remote operation. It can utilize commercially available network cards, SIM card modules, or Wi-Fi modules capable of network communication. If a Wi-Fi module is used, it can connect to a home Wi-Fi router or hotspot under the control of the control module, access the network, log in to the cloud platform 12 (such as Onenet, Huawei Cloud, etc.), obtain cloud platform commands, upload data to the cloud platform, and access the cloud platform via a mobile application to obtain system data and operate the clothes drying system, realizing IoT functionality. The mobile application should include, but is not limited to, the following functional modules: 1. Parameter status display: Displaying current environmental monitoring data such as temperature and humidity, the quantity and type of clothes being dried, and related parameters such as dryness and wetness status; 2. Clothes drying and storage operation: Controlling the system to store specified or all dried clothes, stop or adjust the rotation speed, etc.; 3. System built-in parameter settings: Allowing modification of system parameter thresholds. The mobile application can be developed and built based on a mobile operating system platform, and can also be built based on WeChat developer tools, and can contain relevant application functions as a WeChat mini program.

[0023] Specifically, the aforementioned camera can be a low-cost camera such as the OV2640. The camera's orientation is adjusted according to the hook position to ensure complete image capture of the clothing. The camera can be suspended from the ceiling, oriented towards the rotating wheel. When the wheel moves the clothing past the camera, it should clearly capture the current clothing while minimizing the entry of images of preceding or following clothing into the field of view. The camera communicates with the embedded chip via the SCCB protocol, transmitting the captured clothing images to a pre-trained clothing recognition neural network model within the control module for clothing recognition. The neural network recognition model consists of a convolutional neural network for image information extraction and a fully connected layer network for outputting categories. Specifically, the model can be built based on open-source AI frameworks such as Keras or PyTorch, utilizing well-known high-accuracy open-source convolutional neural network models for image recognition (CNNnet), such as VGGnet or MobileNet, by setting it to retain only the initial convolutional network portion for image feature extraction. Preferably, the MobileNetV3 network is used, as this model not only has high accuracy but also a small size, making it suitable for embedded chip applications. Then, the convolutional part of the first half of the model, which extracts graphic features, is followed by a clothing classification network customized according to the usage scenario to construct the neural network clothing recognition model, such as... Figure 3As shown. The clothing classification network is a fully connected layer network set according to the expected classification categories. The input parameters are determined by the image recognition model used, and the output parameters are the clothing categories to be classified. As an example, clothing can be divided into three categories: tops, pants, and skirts. In actual applications, the number of categories can be adjusted according to the specific scenario. However, the more detailed the classification, the higher the requirements for the network model. It can be adjusted according to the actual application needs. After determining the clothing categories, a fully connected layer network with the corresponding number of neurons is built according to the input and output parameters. The softmax classifier is selected as the activation function of the output layer, thus constructing the neural network model for clothing recognition. During model training, it can be trained based on rich clothing annotation datasets such as open source DeepFashion or Fashionpedia. These datasets have more detailed classifications. Before training, the categories of the dataset can be selected and merged according to the target classification, and then divided into training and test sets in a 4:1 ratio for training and testing the model. Since it is based on a well-trained open source model, this training process is efficient and has excellent accuracy. To apply this model to an embedded system, C code conversion is required. The trained model is exported as a .h5 file. The X-Cube-AI extension package in STM32CubeMX is used to convert the trained model into C code recognizable by the microcontroller. Finally, the file is deployed and the program is burned into the microcontroller. In actual operation, the clothes drying area uses a camera to capture images of the clothes in real time. These images are then input into the clothing type recognition model to obtain the corresponding clothing category label, providing a basis for subsequent sorting and storage.

[0024] The clothing wetness / dryness recognition module integrates three types of data—gravity, humidity, and drying time—to improve accuracy and robustness. Regarding weight changes, wet clothing gradually decreases in weight as moisture evaporates during drying, eventually stabilizing. The system sets a threshold during measurement; when the difference in gravity between two consecutive measurements is less than this threshold, the clothing is considered essentially dry. As a preferred solution, a gravity hook 5... Figure 4As shown, two sets of springs 5-1 are provided. Measuring scales 5-2 are located on both sides of the springs on the hook body 5-3. These two sets of springs can accommodate the measurement of weight changes in both lightweight and heavy clothing. When the clothing is light, the larger spring will not show a significant change in reading; the weight change can be determined solely by the reading of the smaller spring. When the clothing is heavy, the smaller spring's reading reaches its maximum, and only the reading of the larger spring needs to be observed. A pre-reserved limiting space is provided under the smaller spring, ensuring that even when heavy clothing is hung, the smaller spring remains within its elastic deformation range and will not be damaged. Furthermore, the center of gravity of this gravity hook structure is positioned directly below the clothesline, with limiting protection to prevent the hanger from falling off the hook. During use, several of these gravity hooks are evenly distributed and directly fixed to the clothesline. The weight change parameters of the clothing can then be obtained by taking photos with a camera. Specifically, the camera can capture images showing the gravity dial reading, which can be identified using OpenCV open-source functions to obtain the gravity change information. Optionally, the spring can be replaced with Bluetooth and a pressure sensor. Bluetooth will feed back the information measured by the pressure sensor to the control module in real time to obtain the gravity change, which will slightly increase the energy consumption. Regarding changes in clothing humidity, when the clothing moves to the preset detection position with the transmission mechanism, the clothing will come into contact with a humidity probe fixed to the wall. That is, the humidity probe will brush past the edge of the clothing. The probe is located slightly below the hanger; as long as there is clothing hanging, it will come into contact with the humidity probe. Multiple probes at different heights can also be set to adapt to humidity measurement of clothing of different lengths, thereby obtaining the clothing's humidity information. The difference between the measured humidity and the reference ambient humidity is compared to determine the current dryness of the clothing. If the difference is less than a threshold, the clothing is considered dry. Regarding the drying time, a camera can record the drying time of different clothing on the clothesline, and the dryness is confirmed by referring to the experienced drying time of different clothing. The above three criteria are used for comprehensive decision-making through weighted fusion or logical AND / OR rules. Only when multiple conditions are met is the clothing finally confirmed to be dry, thus triggering the storage process.

[0025] The clothes drying module includes a circulating drying mechanism. To improve the evenness of drying and avoid damage or odor problems caused by localized sun exposure or insufficient sunlight, a circulating drying structure is adopted, such as... Figure 2 As shown. Specifically, two rotatable wheels are installed upside down on the ceiling. The rotating motor is controlled by the control module and directly drives one of the wheels. A closed-loop chain or steel wire rope is tensioned between the two wheels to serve as a clothesline. The control module controls one of the wheels to drive the clothesline in a cyclical motion. Several hooks with gravity measuring devices are fixed at equal intervals on the clothesline, and clothes are hung on the hooks via hangers. When the motor starts, all the clothes move at a uniform speed along the circular path with the clothesline, achieving periodic rotation and ensuring that each piece of clothing receives even sunlight and ventilation, improving overall drying efficiency. At the same time, this cyclical mechanism is also key to delivering the clothes to the aforementioned designated position for the camera to capture information, or to the position of the electric push rod described later to lift the clothes.

[0026] After confirming the dryness / wetness and category of the clothing, the system initiates an automatic sorting and storage process. The storage mechanism mainly consists of a slide rail assembly, an electric telescopic rod, and a swing slide rail driven by a deflection servo motor. (See attached image.) Figure 2 The electric telescopic rod is fixed to the side of the rotating wheel and can be suspended from the ceiling or supported on the floor. Its specific position needs to be adjusted according to the position of the gravity hook fixed to the clothesline, ensuring that when the telescopic rod is pushed upwards, it enters the hook of the clothes hanger, but not the gravity hook. When it senses that clothing needs to be stored and the clothing has moved to a designated position along the clothesline, it drives the corresponding telescopic rod to push upwards, causing the clothes hanger with clothing to detach from the gravity hook and slide along the telescopic rod under gravity to the swing rail connected to the back of the telescopic rod. The other end of the swing rail is fixed to a swingable servo motor controlled by the control module. The servo motor dynamically adjusts the swing rail according to the type and dryness / wetness of the clothing, ensuring it precisely aligns with the corresponding branch rail. This allows the clothes hanger with clothing to finally slide into the designated wardrobe under gravity, completing automatic sorting and collection.

[0027] Another key module for the intelligent clothes drying system is the communication module, which includes a Wi-Fi module and a mobile application. This specific implementation uses a Wi-Fi networking module, specifically an ESP8266 device. The control module, based on the Wi-Fi module, executes AT commands to connect to the router and access the network, logs into the cloud platform (such as Onenet, Huawei Cloud, etc.) via the MQTT protocol, uploads local system parameters to the cloud platform, and retrieves data from the cloud platform. When accessing via mobile phone, it can connect to the cloud server via the HTTP protocol to interact and operate the clothes drying system. To facilitate operation, a corresponding app will be developed; preferably, a corresponding WeChat mini-program can be developed and published using WeChat developer tools, making development simpler and application more convenient. The mobile application should include, but is not limited to, the following functional modules: 1. Parameter status display: Displaying current environmental monitoring data such as temperature and humidity, the quantity, type, and dry / wet status of the clothes being dried; 2. Clothes drying and storage operation: Controlling the system to store specified or all dried clothes, stop or change the rotation speed, etc.; 3. Built-in system parameter settings: Allowing modification of system parameter thresholds. Create a user interface and window on the WeChat mini program that connects to the cloud platform, allowing users to obtain system status via their mobile phones and issue commands to the system.

[0028] In summary, the specific steps of the above-mentioned application method for the intelligent clothes drying and storage system based on multimodal perception are as follows: First, the relevant components are arranged as described above, including a rotating wheel fixed to the ceiling, a connected rotary motor, a clothesline tensioned on the wheel, gravity hooks fixed to the clothesline, and interconnected electric telescopic rods, swing rails, branch rails, and corresponding clothing storage cabinets fixed to the ceiling, walls, or floor; as well as sensors distributed indoors and outdoors for monitoring environmental parameters, and a camera fixed to the ceiling for clear image capture of clothing, ensuring that each component operates normally under the control of the control module. When clothes need to be dried, the rotary motor stops, and the clothes to be dried are hung on several evenly distributed gravity hooks. The motor is started, and the system begins to operate. The system automatically determines whether the current environment is suitable for drying clothes based on the sensing module. If suitable, it continues to operate and cycle through drying. If unsuitable, the system controls the execution model to collect all the clothes into the designated wardrobe using the telescopic rods and swing rails. During the drying process, the system automatically acquires the humidity, weight, drying time, and type of clothing on each hook. Based on these conditions, when the clothing is deemed dry, the system adjusts the swing rail, and the telescopic rod lifts the garment and slides it into the designated wardrobe. Notably, parameter and status acquisition, as well as clothing storage control, can all be performed via a mobile application.

[0029] As can be seen, this system achieves a closed-loop automation of the entire process from "identification—judgment—execution—classification," significantly improving user convenience, comfort, and home intelligence. This system boasts a high degree of automation and intelligence, preventing damage caused by uneven drying or unsuitable drying environments, reducing clothing loss rates, optimizing the living environment, and enhancing comfort; it also reduces manual operation, saving labor costs; improves clothing drying efficiency, saving time and space, and possesses low-carbon and environmentally friendly advantages.

[0030] The above-described embodiments are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A smart clothes drying and storage system based on multimodal perception, characterized in that: It includes a control module, a sensing module, an execution module, and a communication module, among which, The control module is used for program execution and module control; The sensing module includes a clothes drying environment monitoring module and a clothing category and dry / wet identification module, wherein: The clothes drying environment monitoring module consists of sensors for monitoring various indoor and outdoor environmental parameters. The clothes drying environment monitoring module is connected to the control module by lines or signals, transmits the various environmental data collected by the sensors to the control module, and compares them with the built-in thresholds of the control module to determine whether the current environment is suitable for drying clothes. The clothing category and dry / wet identification module consists of a camera, a humidity probe, and a gravity hook. The camera, humidity probe, and gravity hook are respectively connected to the control module circuitry or signal to identify the type of clothes being dried and their dry / wet status. The execution module includes a rotary motor, two or more rotating wheels, a circulating clothesline, an electric telescopic rod, a swing slide rail, several branch slide rails, and a clothing storage cabinet. The rotary motor is connected to and controlled by the control module. The rotary motor is connected to one of the rotating wheels, which acts as the driving wheel. The circulating clothesline is tensioned on the rotating wheel. The rotary motor drives the driving wheel to rotate, and then the belt drive of the circulating clothesline drives the other rotating wheels, which act as driven wheels, to rotate. The hook bodies of several gravity hooks are fixed at intervals on the circulating clothesline. The electric telescopic rod is fixed to the side of the rotating wheel and is used to extend and lift the clothes hanger with clothes on it, so that the clothes hanger automatically disengages from the hook that continues to rotate. One end of the swing slide rail is fixed to the end of the electric telescopic rod, and the other end swings to connect with one of the branch slide rails as needed. The ends of each branch slide rail are placed in the corresponding clothing storage cabinet to realize the circulating drying and intelligent classification and storage of clothes. The communication module is electrically connected to the control module and can perform remote system control and data transmission and reception. It is used to realize data interaction between the control module and the remote control terminal and supports remote operation.

2. The intelligent clothes drying and storage system based on multimodal perception according to claim 1, characterized in that: The control module includes, but is not limited to, embedded system chips or microcomputers. The microcomputers include, but are not limited to, Raspberry Pi or Arduino development boards. The embedded system chips include, but are not limited to, STM32 series or RK3588 series microcontrollers.

3. The intelligent clothes drying and storage system based on multimodal perception according to claim 2, characterized in that: The circulating clothesline includes, but is not limited to, closed-loop chains or steel wire ropes, which move in a circular motion driven by a motor to evenly dry clothes.

4. The intelligent clothes drying and storage system based on multimodal perception according to claim 1, characterized in that: The gravity hook consists of a hook body for hanging clothes hangers and a pressure sensor or spring for detecting the gravity it bears. The pressure sensor or spring is located inside the hook body.

5. The intelligent clothes drying and storage system based on multimodal perception according to claim 1, characterized in that: One end of the swing slide rail is fixed to the end of the electric telescopic rod, and the other end swings to connect with one of the branch slide rails as needed. The ends of each branch slide rail are placed in the corresponding clothing storage cabinet. A swingable servo motor is fixed on the swing slide rail, and the servo motor is connected to the control module circuit.

6. The intelligent clothes drying and storage system based on multimodal perception according to claim 1, characterized in that: The control module has a built-in neural network recognition model that is pre-trained using images of labeled clothing. The camera of the perception module captures images and inputs them into the neural network recognition model trained with sample clothing images. The neural network recognition model consists of a convolutional neural network for extracting image information and a fully connected layer network for category output, so as to identify the type of clothing in the captured image with high accuracy.

7. The intelligent clothes drying and storage system based on multimodal perception according to claim 1, characterized in that: This system monitors the clothes drying environment using a multimodal fusion of multiple sensors, including commercial temperature and humidity sensors, wind sensors, rain sensors, pressure sensors, and / or dust sensors.

8. The intelligent clothes drying and storage system based on multimodal perception according to claim 1, characterized in that: Multimodal fusion dry and wet state detection includes setting a gravity sensor on the clothing hanging assembly to determine the dry and wet state of the clothing by detecting changes in the weight of the clothing; the clothing passing over a humidity probe to measure the humidity and comparing it with a reference humidity to determine the dry and wet state; and recording the clothing type and drying time through a camera and comparing it with an experienced reference threshold to confirm the dry and wet state.

9. The intelligent clothes drying and storage system based on multimodal perception according to claim 1, characterized in that: The remote control terminal includes, but is not limited to, smartphones or tablets.

Images