Air purification method of household air purification robot

Abstract

The invention provides an air purification method of a household air purification robot. The air purification method comprises the following steps that a microprocessor plans a patrol path of a moving unit; transmitting the detection data to a microprocessor, and performing fusion processing on the detection data through a multi-modal data fusion algorithm; combining historical detection data and sensor real-time detection data after fusion processing, and using a clustering algorithm to identify an air quality hot spot area; the air quality hot spot area, all rooms of a family and obstacles serve as nodes, a shortest path algorithm is used for conducting purification path planning of the moving unit, the moving unit covers the air quality hot spot area with the shortest path, and the microprocessor controls the moving unit to move according to the planned purification path; the microprocessor timely starts the air purification unit according to real-time detection data of the sensor. The purification efficiency and the robot utilization rate are improved, and blind movement and repeated purification are reduced.

Description

Technical Field

[0001] The invention relates to an air purification method, in particular to an air purification method of a household air purification robot, and belongs to the technical field of air purification. Background Art

[0002] With the increase of air pollution and the spread of infectious diseases, the monitoring and purification of indoor air quality in homes has become particularly important. In recent years, with the advancement of robotics technology, intelligent and mobile household air purification robots have become a research hotspot in the field of indoor air purification. However, the existing household air purification robots still have some limitations in home applications. Some household air purification robots adopt a fixed position design, which limits the purification range and can only improve the air quality of local spaces in the home, making it difficult to meet the purification needs of large areas and multiple rooms. Although some air purification robots have the ability to move autonomously, the movement path is often preset or highly random, lacking path planning, resulting in blind movement and repeated purification, reducing purification efficiency. Summary of the invention

[0003] Based on the above background, an object of the present invention is to provide an air purification method for a household air purification robot to solve the problems described in the background technology.

[0004] In order to achieve the above-mentioned object of the invention, the present invention provides the following technical solutions:

[0005] An air purification method for a household air purification robot, the household air purification robot comprising a robot body and an air quality identification unit, an air purification unit, a mobile unit and a control unit arranged on the robot body, the air quality identification unit comprising a multi-parameter air quality sensor, a laser particle counter and a bioaerosol sensor, the air purification unit comprising a primary filter, an activated carbon filter, a high-efficiency particulate air filter and an ultraviolet sterilization lamp, the mobile unit is used to move the robot body, and the control unit comprises a microprocessor electrically connected to the air quality identification unit, the air purification unit and the mobile unit; the air purification method comprises the following steps:

[0006] The microprocessor plans the patrol path of the mobile unit, controls the mobile unit to move along the patrol path, and stays at the data collection location for a certain period of time;

[0007] At each data collection location, a multi-parameter air quality sensor is used to detect the concentration of various air pollutants in different areas of the home. A laser particle counter is used to measure the number of particles of different sizes in the air of different areas of the home to detect the concentration of particles in the air. A bioaerosol sensor is used to detect bioaerosols in the air of different areas of the home. The detection data of the multi-parameter air quality sensor, the laser particle counter and the bioaerosol sensor are transmitted to the microprocessor, and the detection data are fused and processed by a multi-modal data fusion algorithm.

[0008] Combining historical detection data with real-time sensor detection data after fusion processing, a clustering algorithm is used to identify air quality hotspots;

[0009] Air quality hotspots, home rooms and obstacles are used as nodes, and the shortest path algorithm is used to plan the purification path of the mobile unit so that it covers the air quality hotspots with the shortest path. The microprocessor controls the mobile unit to move along the planned purification path.

[0010] The microprocessor starts the air purification unit in a timely manner according to the real-time detection data of the sensor, so that the air passes through the primary filter, activated carbon filter, high-efficiency particulate air filter and ultraviolet germicidal lamp according to the purification strategy, thereby removing pollutants and pathogenic microorganisms in the air.

[0011] Preferably, the multimodal data fusion algorithm comprises the following steps:

[0012] Each test data is standardized, and the mathematical expression of the standardized processing is:

[0013]

[0014] In the formula, is the standardized data, x i is the detection data of the i-th sensor, μ i is the mean value of the i-th sensor, σ i is the standard deviation of the i-th sensor;

[0015] The standardized data of each sensor is weighted fused, and the mathematical expression of the weighted fusion is:

[0016]

[0017] Where z is the initial fusion feature vector obtained after weighted fusion, w i represents the weight coefficient of the i-th sensor, and its constraints are

[0018] The preliminary fused feature vector is input into the deep learning model for further fusion and the air quality index is output.

[0019] Preferably, the deep learning model is a multi-layer perceptron model, and the loss function of the multi-layer perceptron model is:

[0020]

[0021] Where θ is the weight matrix and bias vector of each layer of the multilayer perceptron model, N is the number of training samples, and y i is the true air quality index value of the i-th sample, Represents the air quality index value predicted by the multi-layer perceptron model;

[0022] θ is updated through the back-propagation algorithm and the optimizer to minimize the loss function L(θ).

[0023] Preferably, the clustering algorithm is an improved K-means clustering algorithm, and the improved K-means clustering algorithm comprises the following steps:

[0024] Normalize the air quality index value, timestamp, and spatial coordinates and map them to the [0,1] interval;

[0025] Calculate the mean and standard deviation of the air quality index values ​​in the historical detection data set, select samples with air quality index values ​​greater than μ+σ as candidate initial cluster centers, and randomly select multiple samples from the candidate initial cluster centers as initial cluster centers;

[0026] For each sample in the real-time sensor detection data after fusion processing, the weighted distance between it and each cluster center is calculated using the following formula: Where, d ij is the weighted distance between the i-th sample and the j-th cluster center, x i is the air quality index value of the i-th sample, t i is the timestamp of the i-th sample, s i is the spatial coordinate of the i-th sample, is the air quality index value of the jth cluster center, is the timestamp of the j-th cluster center, is the spatial coordinate of the jth cluster center; each sample is assigned to the cluster center with the closest weighted distance to it to form a new cluster; the center of each cluster is updated, and the update calculation formula is: In the formula, m j is the cluster center, C j is the jth cluster;

[0027] Calculate the difference between the current cluster center and the previous cluster center. If the difference is less than the preset threshold or the maximum number of iterations is reached, stop the iteration; otherwise, repeat the previous clustering step.

[0028] For each final cluster, the mean and variance of the air quality index values ​​of its internal samples are calculated as the representative indicators of the cluster. Based on the representative indicators and spatial distribution of the cluster, the air quality hotspots are identified.

[0029] As a preferred method, the air quality hotspots, the rooms in the home and the obstacles are taken as nodes and specifically include:

[0030] The microprocessor stores the floor plan of each room of the home, takes the center point of each room as a node, and the connecting paths between the rooms as edges, and constructs a graph including a node set and an edge set;

[0031] The center point of the air quality hotspot area and the obstacles in each room of the home are added to the graph as nodes. For the nodes representing obstacles, the edge weights between them and the adjacent rooms are set to larger values.

[0032] Preferably, the shortest path algorithm is an improved Dijkstra algorithm, and the improved Dijkstra algorithm comprises the following steps:

[0033] Each node in the graph has an air quality index corresponding to the node, and each edge has a corresponding length weight;

[0034] For each node in the graph, initialize the shortest path estimate, that is, the shortest path length from the current position of the mobile unit as the starting point to the node; initialize the shortest path estimate of the starting point to zero; initialize the predecessor node of each node to none; initialize the unvisited node set Q to all nodes;

[0035] Select the node u with the smallest shortest path estimate from the unvisited node set Q, and remove the node u from the unvisited node set Q; for each neighbor node v of node u j ∈Q, calculate the distance from the starting point through node u to node v j The path length is calculated as:

[0036] d new =d[u]+w uj +α·AQI j ;

[0037] Where d[u] is the shortest path estimate for node u, w uj For nodes u and v j The length weight of the edge between them, α is the air quality weight factor, AQI j For node v j The air quality index value; if d new <d[v j ], then update node vj The shortest path estimate of , and set its predecessor node to node u;

[0038] Repeat the previous step until the target node, i.e., the center point of the air quality hotspot area, has been visited, or the unvisited node set Q is empty;

[0039] Starting from the target node, trace back to the starting point according to the predecessor node information, generate the optimal path, and output the optimal path from the starting point to the target node as the purification path.

[0040] Preferably, the robot body is provided with a laser radar sensor, and when the microprocessor controls the mobile unit to move according to the planned purification path, the air purification method further comprises the following steps:

[0041] During the movement of the robot body, the environmental information of the home room is continuously collected through the lidar sensor, and the environmental map is updated in real time using the synchronous positioning and mapping algorithm. When a new obstacle is detected or the position of the original obstacle changes, the path re-planning is triggered. The improved Dijkstra algorithm is re-executed using the updated environmental map and obstacle information to generate a new purification path.

[0042] Preferably, the purification strategy comprises:

[0043] The air quality index is obtained by fusing the real-time detection data of the sensor;

[0044] If the current air quality index is lower than the first threshold, the purification unit is put into the primary working state, and only the primary filter works to maintain air cleanliness;

[0045] If the current air quality index is between the first threshold and the second threshold, the purification unit is put into a normal working state, and the primary filter and the high efficiency particulate air filter work simultaneously;

[0046] If the current air quality index is between the second threshold and the third threshold, the purification unit is put into an enhanced working state, and the primary filter, the activated carbon filter and the high-efficiency particulate air filter work simultaneously;

[0047] If the current air quality index is higher than the third threshold, the purification unit is in super-strong working state, and the primary filter, activated carbon filter, high-efficiency particulate air filter and ultraviolet germicidal lamp work at the same time.

[0048] Compared with the prior art, the present invention has the following advantages:

[0049] An air purification method for a household air purification robot of the present invention performs fusion processing on detection data of each sensor through a multimodal data fusion algorithm, effectively integrates information from different sensors, and analyzes historical detection data and real-time detection data through a clustering algorithm, thereby effectively identifying hot spots with poor air quality in an indoor environment. Compared with traditional random or full-area purification methods, the purification efficiency and robot utilization rate are improved; the air purification method takes air quality hot spots, rooms and obstacles as nodes, and uses the shortest path algorithm to perform path planning for the mobile unit, so that the robot covers all air quality hot spots with the shortest path, significantly improving the purification coverage efficiency and reducing blind movement and repeated purification. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0051] Figure 1 It is a schematic diagram of the structure of the household air purification robot in the present invention;

[0052] Figure 2 is a flow chart of the air purification method of the present invention;

[0053] Figure 3 is a flow chart of the multimodal data fusion algorithm of the present invention;

[0054] Figure 4 is a flow chart of the improved K-means clustering algorithm in the present invention;

[0055] Figure 5 It is a flow chart of the improved Dijkstra algorithm in the present invention. DETAILED DESCRIPTION

[0056] The technical solution of the present invention is further described in detail below through specific embodiments and in conjunction with the accompanying drawings. It should be understood that the implementation of the present invention is not limited to the following embodiments, and any form of modification and / or change made to the present invention will fall within the protection scope of the present invention.

[0057] In the present invention, unless otherwise specified, all parts and percentages are weight units, and the equipment and raw materials used can be purchased from the market or are commonly used in the art. The methods in the following embodiments, unless otherwise specified, are conventional methods in the art. The components or equipment in the following embodiments, unless otherwise specified, are universal standard parts or components known to those skilled in the art, and their structures and principles are known to those skilled in the art through technical manuals or conventional experimental methods.

[0058] The embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present invention. However, one or more embodiments may also be implemented by those skilled in the art without these specific details.

[0059] An embodiment of the present invention discloses an air purification method of a household air purification robot.

[0060] like Figure 1 As shown, the household air purification robot includes a robot body and an air quality identification unit, an air purification unit, a mobile unit and a control unit arranged on the robot body. The air quality identification unit includes a multi-parameter air quality sensor, a laser particle counter and a bioaerosol sensor, the air purification unit includes a primary filter, an activated carbon filter, a high-efficiency particulate air filter and an ultraviolet sterilization lamp, the mobile unit is used to move the robot body, and the control unit includes a microprocessor electrically connected to the air quality identification unit, the air purification unit and the mobile unit.

[0061] like Figure 2 As shown, the air purification method comprises the following steps:

[0062] S1, the microprocessor plans the patrol path of the mobile unit, controls the mobile unit to move along the patrol path, and stays at the data collection location for a certain period of time;

[0063] S2. At each data collection location, the concentration of various air pollutants in different areas of the home is detected by a multi-parameter air quality sensor, the number of particles of different particle sizes in the air of different areas of the home is measured by a laser particle counter to detect the concentration of particles in the air, and the bioaerosol sensor is used to detect the bioaerosol in the air of different areas of the home. The detection data of the multi-parameter air quality sensor, the laser particle counter and the bioaerosol sensor are transmitted to the microprocessor, and the detection data are fused through a multimodal data fusion algorithm;

[0064] S3, combining historical detection data and fused real-time sensor detection data, using clustering algorithms to identify air quality hotspots;

[0065] S4, taking the air quality hotspot area, each room of the home and obstacles as nodes, using the shortest path algorithm to plan the purification path of the mobile unit, so that it covers the air quality hotspot area with the shortest path, and the microprocessor controls the mobile unit to move according to the planned purification path;

[0066] S5. The microprocessor starts the air purification unit in a timely manner according to the real-time detection data of the sensor, so that the air passes through the primary filter, activated carbon filter, high-efficiency particulate air filter and ultraviolet sterilization lamp according to the purification strategy, thereby removing pollutants and pathogenic microorganisms in the air.

[0067] like Figure 3 As shown, in step S2, the multimodal data fusion algorithm includes the following steps:

[0068] S201, perform standardization processing on each test data, and the mathematical expression of the standardization processing is:

[0069]

[0070] In the formula, is the standardized data, x i is the detection data of the i-th sensor, μ i is the mean value of the i-th sensor, σ i is the standard deviation of the i-th sensor;

[0071] S202, weighted fusion is performed on the standardized data of each sensor, and the mathematical expression of weighted fusion is:

[0072]

[0073] Where z is the initial fusion feature vector obtained after weighted fusion, w i represents the weight coefficient of the i-th sensor, and its constraints are

[0074] S203: Input the preliminary fused feature vector into the deep learning model for further fusion, and output the air quality index.

[0075] Among them, the deep learning model is a multi-layer perceptron model, and its forward propagation process is expressed as:

[0076]

[0077] In the formula, h 1 ,h 2 ,…,h l represents the hidden layer of the model, σ represents the activation function, which is selected as ReLU in this embodiment, and W 1 ,W 2 ,…,Wl ,W out represents the weight matrix of each layer of the model, b 1 ,b 2 ,…,b l ,b out The bias vectors representing each layer of the model, the weight matrix and the bias vector are used as learnable parameters and are optimized through model training.

[0078] The loss function of the multi-layer perceptron model is:

[0079]

[0080] Where θ is the weight matrix and bias vector of each layer of the multilayer perceptron model, N is the number of training samples, and y i is the true air quality index value of the i-th sample, Represents the air quality index value predicted by the multi-layer perceptron model;

[0081] θ is updated through the back-propagation algorithm and the optimizer to minimize the loss function L(θ).

[0082] The above steps of the multimodal data fusion algorithm combine weighted averaging and deep learning. First, the multi-sensor data is preliminarily fused through weighted averaging, and then the fused data is input into the deep learning model for feature extraction and further fusion. Finally, the deep learning model outputs the air quality index AQI, which serves as an important basis for the subsequent steps of identifying air quality hotspots.

[0083] like Figure 4 As shown, in step S3, the clustering algorithm is an improved K-means clustering algorithm, and its input includes the historical air quality data set Time weight α, space weight β, maximum number of clusters K max , and its output is the clustering result C = C 1 ,C 2 ,…,C K Among them, x i represents the air quality index value (AQI value) of the i-th sample, t i represents the timestamp of the i-th sample, s i Represents the spatial coordinates of the i-th sample.

[0084] The improved K-means clustering algorithm includes the following steps:

[0085] S301, air quality index value x i , timestamp t i , spatial coordinates s i Normalize it and map it to the interval [0,1];

[0086] S302, calculating the mean and standard deviation of the air quality index values ​​in the historical detection data set D, selecting samples with air quality index values ​​greater than μ+σ as candidate initial cluster centers, and randomly selecting multiple samples from the candidate initial cluster centers as initial cluster centers; where μ is the sample mean of the historical detection data set D, and σ is the sample standard deviation of the historical detection data set D;

[0087] S303, for each sample in the sensor real-time detection data after fusion processing, calculate its weighted distance with each cluster center, and the calculation formula is: Where, d ij is the weighted distance between the i-th sample and the j-th cluster center, x i is the air quality index value of the i-th sample, t i is the timestamp of the i-th sample, s i is the spatial coordinate of the i-th sample, is the air quality index value of the jth cluster center, is the timestamp of the j-th cluster center, is the spatial coordinate of the jth cluster center; each sample is assigned to the cluster center with the closest weighted distance to it to form a new cluster; the center of each cluster is updated, and the update calculation formula is: In the formula, m j is the cluster center, C j is the jth cluster;

[0088] S304, calculating the difference between the current cluster center and the previous cluster center, the calculation formula is:

[0089]

[0090] If the difference is less than the preset threshold or reaches the maximum number of iterations, the iteration is stopped, otherwise, step S303 is repeated;

[0091] S305. For each final cluster, the mean and variance of the air quality index values ​​of its internal samples are calculated as representative indicators of the cluster, and the air quality hotspot areas are identified based on the representative indicators and spatial distribution of the cluster.

[0092] Among them, the maximum number of clusters K max The determination method is to calculate the clustering quality index SSE under the current cluster number k according to the clustering result of step S303. k , the calculation formula is:

[0093]

[0094] Then, calculate the current number of clusters k and SSE k The ratio r k ;

[0095] After that, draw kr k Curve, take the k value corresponding to the inflection point of the curve as the maximum cluster number K max .

[0096] The improved K-means clustering algorithm combines time weight, space weight and the determination of adaptive cluster number. By introducing time and space information, the algorithm can better capture the spatiotemporal correlation of air quality data, thereby obtaining better clustering results.

[0097] like Figure 5 As shown, in step S4, the air quality hotspot area, each room in the home and obstacles are taken as nodes and specifically include:

[0098] S401, the microprocessor stores the floor plan of each room of the home, takes the center point of each room as a node, and the connecting paths between the rooms as edges, and constructs a graph including a node set and an edge set;

[0099] S402: Add the center point of the air quality hotspot area and the obstacles in each room of the home as nodes to the graph. For the node representing the obstacle, set the edge weight between it and the adjacent room to a larger value.

[0100] In step S4, the shortest path algorithm is an improved Dijkstra algorithm, and the improved Dijkstra algorithm includes the following steps:

[0101] S403, making each node in the graph have an air quality index corresponding to the node, and each edge have a corresponding length weight;

[0102] S404. For each node v in the graph i ∈V, initialize the shortest path estimate d[v i ]=∞, that is, the shortest path length from the current position of the mobile unit to the node; the shortest path estimate of the initial starting point is zero, d[s]=0; the predecessor node of each node is initialized to none, prev[v i ]=None; Initialize the unvisited node set Q to all nodes, Q=V;

[0103] S405, select a node u with the smallest shortest path estimation value from the unvisited node set Q, and remove the node u from the unvisited node set Q. The selection formula is:

[0104]

[0105] In the formula, α is the air quality weight factor;

[0106] For each neighbor node v of node uj ∈Q, calculate the distance from the starting point through node u to node v j The path length is calculated as:

[0107] d new =d[u]+w uj +α·AQI j ;

[0108] Where d[u] is the shortest path estimate for node u, w uj For nodes u and v j The length weight of the edge between them, α is the air quality weight factor, AQI j For node v j The air quality index value; if d new <d[v j ], then update node v j The shortest path estimate of , and set its predecessor node to node u;

[0109] S406, repeating step S405 until the target node, i.e., the center point of the air quality hotspot area, has been visited, or the unvisited node set Q is empty;

[0110] S407 , starting from the target node, backtracking to the starting point according to the predecessor node information, generating an optimal path, and outputting the optimal path from the starting point to the target node as a purification path.

[0111] The improved Dijkstra algorithm introduces the air quality index AQI as a selection criterion in the node selection stage, and balances the importance of path length and air quality through the air quality weight factor. If α is large, the algorithm will give priority to nodes with high air quality priority, and if α is small, the algorithm will give priority to path length. When calculating the path length, the algorithm considers both the edge length weight and the node's air quality index AQI, thereby generating an optimal path that comprehensively considers path length and air quality. Therefore, the optimal path generated by the improved Dijkstra algorithm is not only the shortest path, but also takes into account the air quality along the way.

[0112] Furthermore, the robot body is also provided with a laser radar sensor. When the microprocessor controls the mobile unit to move according to the planned purification path, the air purification method further includes the following steps:

[0113] During the movement of the robot body, the environmental information of the home room is continuously collected through the lidar sensor, and the environmental map is updated in real time using the synchronous positioning and mapping algorithm. When a new obstacle is detected or the position of the original obstacle changes, the path re-planning is triggered. The improved Dijkstra algorithm is re-executed using the updated environmental map and obstacle information to generate a new purification path.

[0114] In step S5, the purification strategy includes:

[0115] The air quality index is obtained by fusing the real-time detection data of the sensor;

[0116] If the current air quality index is lower than the first threshold, the purification unit is put into the primary working state, and only the primary filter works to maintain air cleanliness;

[0117] If the current air quality index is between the first threshold and the second threshold, the purification unit is put into a normal working state, and the primary filter and the high efficiency particulate air filter work simultaneously;

[0118] If the current air quality index is between the second threshold and the third threshold, the purification unit is put into an enhanced working state, and the primary filter, the activated carbon filter and the high-efficiency particulate air filter work simultaneously;

[0119] If the current air quality index is higher than the third threshold, the purification unit is in super-strong working state, and the primary filter, activated carbon filter, high-efficiency particulate air filter and ultraviolet germicidal lamp work at the same time.

[0120] The air purification method of this embodiment uses a multimodal data fusion algorithm to fuse the detection data of each sensor, effectively integrates the information of different sensors, and analyzes historical detection data and real-time detection data through a clustering algorithm, thereby effectively identifying hot spots with poor air quality in the indoor environment. Compared with traditional random or full-area purification methods, it improves purification efficiency and robot utilization. This air purification method uses air quality hot spots, rooms and obstacles as nodes and uses the shortest path algorithm to plan the path of the mobile unit, so that the robot can cover all air quality hot spots with the shortest path, significantly improve the purification coverage efficiency, and reduce blind movement and repeated purification. This air purification method uses multi-stage air purification units such as primary filters, activated carbon filters, high-efficiency particulate air filters and ultraviolet germicidal lamps to comprehensively remove multiple pollutants in the air.

[0121] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core ideas of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made to the present invention without departing from the principles of the present invention, and these improvements and modifications also fall within the scope of protection of the claims of the present invention.

Claims

1. An air purification method for a household air purification robot, Features: The household air purification robot comprises a robot body, an air quality identification unit, an air purification unit, a mobile unit and a control unit arranged on the robot body, the air quality identification unit comprises a multi-parameter air quality sensor, a laser particle counter and a bioaerosol sensor, the air purification unit comprises a primary filter, an activated carbon filter, a high-efficiency particulate air filter and an ultraviolet sterilization lamp, the mobile unit is used to move the robot body, and the control unit comprises a microprocessor electrically connected to the air quality identification unit, the air purification unit and the mobile unit; the air purification method comprises the following steps: The microprocessor plans the patrol path of the mobile unit, controls the mobile unit to move along the patrol path, and stays at the data collection location for a certain period of time; At each data collection location, a multi-parameter air quality sensor is used to detect the concentration of various air pollutants in different areas of the home. A laser particle counter is used to measure the number of particles of different sizes in the air of different areas of the home to detect the concentration of particles in the air. A bioaerosol sensor is used to detect bioaerosols in the air of different areas of the home. The detection data of the multi-parameter air quality sensor, the laser particle counter and the bioaerosol sensor are transmitted to the microprocessor, and the detection data are fused and processed by a multi-modal data fusion algorithm. Combining historical detection data with real-time sensor detection data after fusion processing, a clustering algorithm is used to identify air quality hotspots; Air quality hotspots, home rooms and obstacles are used as nodes, and the shortest path algorithm is used to plan the purification path of the mobile unit so that it covers the air quality hotspots with the shortest path. The microprocessor controls the mobile unit to move along the planned purification path. The microprocessor starts the air purification unit in a timely manner according to the real-time detection data of the sensor, so that the air passes through the primary filter, activated carbon filter, high-efficiency particulate air filter and ultraviolet germicidal lamp according to the purification strategy, thereby removing pollutants and pathogenic microorganisms in the air.

2. The air purification method of a household air purification robot according to claim 1, Features: The multimodal data fusion algorithm comprises the following steps: Each test data is standardized, and the mathematical expression of the standardized processing is: In the formula, is the standardized data, x i is the detection data of the i-th sensor, μ i is the mean value of the i-th sensor, σ i is the standard deviation of the i-th sensor; The standardized data of each sensor is weighted fused, and the mathematical expression of the weighted fusion is: Where z is the initial fusion feature vector obtained after weighted fusion, w i represents the weight coefficient of the i-th sensor, and its constraints are w i ≥0; The preliminary fused feature vector is input into the deep learning model for further fusion and the air quality index is output.

3. The air purification method of a household air purification robot according to claim 2, Features: The deep learning model is a multi-layer perceptron model, and the loss function of the multi-layer perceptron model is: Where θ is the weight matrix and bias vector of each layer of the multilayer perceptron model, N is the number of training samples, and y i is the true air quality index value of the i-th sample, Represents the air quality index value predicted by the multi-layer perceptron model; θ is updated through the back-propagation algorithm and the optimizer to minimize the loss function L(θ).

4. The air purification method of a household air purification robot according to claim 1, Features: The clustering algorithm is an improved K-means clustering algorithm, and the improved K-means clustering algorithm comprises the following steps: Normalize the air quality index value, timestamp, and spatial coordinates and map them to the [0,1] interval; Calculate the mean and standard deviation of the air quality index values ​​in the historical detection data set, select samples with air quality index values ​​greater than μ+σ as candidate initial cluster centers, and randomly select multiple samples from the candidate initial cluster centers as initial cluster centers; For each sample in the real-time sensor detection data after fusion processing, the weighted distance between it and each cluster center is calculated using the following formula: Where, d ij is the weighted distance between the i-th sample and the j-th cluster center, x i is the air quality index value of the i-th sample, t i is the timestamp of the i-th sample, s i is the spatial coordinate of the i-th sample, is the air quality index value of the jth cluster center, is the timestamp of the j-th cluster center, is the spatial coordinate of the jth cluster center; each sample is assigned to the cluster center with the closest weighted distance to it to form a new cluster; the center of each cluster is updated, and the update calculation formula is: In the formula, m j is the cluster center, C j is the jth cluster; Calculate the difference between the current cluster center and the previous cluster center. If the difference is less than the preset threshold or the maximum number of iterations is reached, stop the iteration; otherwise, repeat the previous clustering step. For each final cluster, the mean and variance of the air quality index values ​​of its internal samples are calculated as the representative indicators of the cluster. Based on the representative indicators and spatial distribution of the cluster, the air quality hotspots are identified.

5. The air purification method of a household air purification robot according to claim 1, Features: The air quality hotspots, home rooms, and obstacles are used as nodes, including: The microprocessor stores the floor plan of each room of the home, takes the center point of each room as a node, and the connecting paths between the rooms as edges, and constructs a graph including a node set and an edge set; The center point of the air quality hotspot area and the obstacles in each room of the home are added to the graph as nodes. For the nodes representing obstacles, the edge weights between them and the adjacent rooms are set to larger values.

6. The air purification method of a household air purification robot according to claim 5, Features: The shortest path algorithm is an improved Dijkstra algorithm, and the improved Dijkstra algorithm comprises the following steps: Each node in the graph has an air quality index corresponding to the node, and each edge has a corresponding length weight; For each node in the graph, initialize the shortest path estimate, that is, the shortest path length from the current position of the mobile unit as the starting point to the node; initialize the shortest path estimate of the starting point to zero; initialize the predecessor node of each node to none; initialize the unvisited node set Q to all nodes; Select the node u with the smallest shortest path estimate from the unvisited node set Q, and remove the node u from the unvisited node set Q; for each neighbor node v of node u j ∈Q, calculate the distance from the starting point through node u to node v j The path length is calculated as: d new =d[u]+w uj +α·AQI j ; Where d[u] is the shortest path estimate for node u, w uj For nodes u and v j The length weight of the edge between them, α is the air quality weight factor, AQI j For node v j The air quality index value; if d new <d[v j ], then update node v j The shortest path estimate of , and set its predecessor node to node u; Repeat the previous step until the target node, i.e., the center point of the air quality hotspot area, has been visited, or the unvisited node set Q is empty; Starting from the target node, trace back to the starting point according to the predecessor node information, generate the optimal path, and output the optimal path from the starting point to the target node.

7. The air purification method of a household air purification robot according to claim 6, Features: The robot body is provided with a laser radar sensor. When the microprocessor controls the mobile unit to move according to the planned purification path, the air purification method further includes the following steps: During the movement of the robot body, the environmental information of the home room is continuously collected through the lidar sensor, and the environmental map is updated in real time using the synchronous positioning and mapping algorithm. When a new obstacle is detected or the position of the original obstacle changes, the path re-planning is triggered. The improved Dijkstra algorithm is re-executed using the updated environmental map and obstacle information to generate a new purification path.

8. The air purification method of a household air purification robot according to claim 1, Features: The purification strategy includes: The air quality index is obtained by fusing the real-time detection data of the sensor; If the current air quality index is lower than the first threshold, the purification unit is put into the primary working state, and only the primary filter works to maintain air cleanliness; If the current air quality index is between the first threshold and the second threshold, the purification unit is put into a normal working state, and the primary filter and the high efficiency particulate air filter work simultaneously; If the current air quality index is between the second threshold and the third threshold, the purification unit is put into an enhanced working state, and the primary filter, the activated carbon filter and the high-efficiency particulate air filter work simultaneously; If the current air quality index is higher than the third threshold, the purification unit is in super-strong working state, and the primary filter, activated carbon filter, high-efficiency particulate air filter and ultraviolet sterilization lamp work at the same time.

Images