A remote monitoring method and system for warehouse clearance robots based on the Internet of Things

Abstract

The present invention is applicable to the field of remote monitoring technology, and in particular relates to a remote monitoring method and system for a warehouse clearance robot based on the Internet of Things. The method comprises: utilizing a preset Internet of Things video monitoring device to collect video monitoring data in the cabin, and taking out a snapshot, delineating the use range of the warehouse clearance robot, dividing the use range into a number of blocks, creating nodes corresponding to the blocks one by one, synchronizing the snapshots to the corresponding nodes, extracting foreground features in the snapshots, and covering the nodes. The present invention can greatly improve the smoothness of warehouse clearance tasks and improve warehouse clearance efficiency by generating a task chain. By calculating edge coordinates, it can identify the location of residues, optimize the warehouse clearance path, and greatly improve the operation accuracy and warehouse clearance efficiency. By verifying the residues, it can further improve the warehouse clearance effect, reduce errors and repeated warehouse clearance, and ensure that the warehouse clearance process is more accurate and efficient.

Description

Technical Field

[0001] The present invention relates to the field of remote monitoring technology, and in particular to a remote monitoring method and system for a warehouse clearance robot based on the Internet of Things. Background Art

[0002] Transport ships typically use equipment such as bucket wheel reclaimers and grab cranes to load and unload cargo. However, for some special types of cargo, such as high-value rare earths, grains prone to mold, and coal with high pollution risks, these cargoes often leave residues during the loading and unloading process. If these residues are not cleared in a timely manner, they will not only cause economic losses but also lead to cross-contamination between different cargoes, contaminate the ship's hull, and affect navigation safety.

[0003] However, most of the existing warehouse cleaning robots use a predetermined path planning method and cannot dynamically respond to sudden environmental changes. When working in a large space, it is easy to cause problems such as repeated cleaning or missed cleaning. Therefore, "how to use video surveillance equipment to provide warehouse cleaning guidance for warehouse cleaning robots" is the technical problem that the present invention needs to solve. Summary of the Invention

[0004] The purpose of the present invention is to provide a remote monitoring method and system for a warehouse clearance robot based on the Internet of Things, so as to solve the problem of "how to use video monitoring equipment to provide warehouse clearance guidance for the warehouse clearance robot" raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A remote monitoring method for a warehouse cleaning robot based on the Internet of Things, the method comprising:

[0007] Using pre-set IoT video surveillance equipment, video surveillance data from the cabin is collected and captured as snapshots. This defines the range of use for the warehouse clearance robot, which is then divided into several blocks. Nodes corresponding to the blocks are created, and the snapshots are synchronized to the corresponding nodes. Foreground features in the snapshots are extracted and overlaid on the nodes. All nodes are integrated to generate a task chain, a position coordinate system is constructed, and the edge coordinates of the foreground features in each node are calculated using an edge detection algorithm.

[0008] Obtain access to the laser generating device pre-integrated on the warehouse clearance robot, read the reflected light signal at the edge coordinates, and perform residue verification on all nodes in turn. If the verification fails, the corresponding node is deleted from the task chain;

[0009] Locating the real-time position of the warehouse cleaning robot, generating a movement path with the real-time position as the starting point and the preset position as the end point, generating waypoints using the edge coordinates, inserting the waypoints into the movement path, and sending the movement path to the warehouse cleaning robot;

[0010] Embed a monitoring mechanism into the task chain, update the snapshot, and adjust the task chain, wherein the adjustment at least includes: merging and splitting;

[0011] Furthermore, the steps of using a preset IoT video surveillance device to collect video surveillance data in the cabin, taking snapshots, defining the use range of the clearance robot, and dividing the use range into a plurality of blocks include:

[0012] Obtain the equipment distribution in the cabin and divide the restricted areas;

[0013] Based on the restricted area, a virtual fence is generated, and the virtual fence is marked in the snapshot;

[0014] Determine whether the real-time position exceeds the virtual fence, and if so, trigger a preset alarm mechanism.

[0015] Furthermore, the steps of creating nodes corresponding to the blocks one by one, synchronizing the snapshots to the corresponding nodes, extracting foreground features in the snapshots, overwriting the nodes, integrating all the nodes, and generating a task chain include:

[0016] Collect historical data on residues in the cabin, label and preprocess it, and generate a training set;

[0017] Creating an object detection model and optimizing it using the training set;

[0018] The snapshot is input into the target detection model, and the foreground features and background features are output.

[0019] Furthermore, the step of constructing a position coordinate system and calculating edge coordinates of foreground features in each node using an edge detection algorithm includes:

[0020] Determine a reference point, define an origin and an axis direction, correct the snapshot, integrate the origin and axis coordinates, and construct a coordinate system;

[0021] The edge detection algorithm is used to delineate the boundaries of the foreground features in each block, and the edge coordinates of the boundaries are read out.

[0022] Furthermore, the steps of obtaining permission to use the laser generating device pre-integrated on the clearance robot, reading the reflected light signal at the edge coordinates, and sequentially performing residue verification on all nodes include:

[0023] Configuring a scanning priority for each block and setting a scanning frequency to cluster the reflected light signal into a smooth portion and a rough portion;

[0024] It is determined whether the reflected light signal at the edge coordinate is a rough portion, and if so, the verification is passed.

[0025] Furthermore, the steps of locating the real-time position of the warehouse cleaning robot, generating a movement path with the real-time position as the starting point and the preset position as the end point, generating waypoints using the edge coordinates, inserting the waypoints into the movement path, and sending the movement path to the warehouse cleaning robot include:

[0026] Integrate the starting point, the waypoints, and the end point to generate a movement path, and determine deviation factors, wherein the deviation factors include at least: power level, progress, and obstacles;

[0027] The deviation value between the real-time position and the moving path is calculated. If the deviation value is greater than a preset threshold, an alarm message is generated and sent to a preset terminal.

[0028] Furthermore, the steps of embedding a monitoring mechanism into the task chain, updating the snapshot, and adjusting the task chain include:

[0029] Collecting fused data from multi-level IoT sensors and integrating it into the task chain, activating the monitoring mechanism, and monitoring the status of the warehouse cleaning robot;

[0030] The generation time of the snapshot is recorded, the snapshot is updated based on a preset frequency, and the status monitoring results and the snapshot are integrated to adjust the task chain.

[0031] Furthermore, the system includes:

[0032] A computing module is configured to utilize preset IoT video surveillance equipment to collect video surveillance data within the cabin, extract snapshots, delineate the usage range of the warehouse clearance robot, divide the usage range into a number of blocks, create nodes corresponding to the blocks, synchronize the snapshots to the corresponding nodes, extract foreground features from the snapshots, overlay the nodes, integrate all nodes, generate a task chain, construct a position coordinate system, and utilize an edge detection algorithm to calculate the edge coordinates of the foreground features in each node;

[0033] A deletion module is used to obtain the use permission of the laser generating device pre-integrated on the clearance robot, read the reflected light signal at the edge coordinate, and perform residue verification on all nodes in turn. If the verification fails, the corresponding node is deleted from the task chain;

[0034] a sending module, configured to locate the real-time position of the warehouse cleaning robot, generate a movement path with the real-time position as a starting point and a preset position as an end point, generate waypoints using the edge coordinates, insert the waypoints into the movement path, and send the movement path to the warehouse cleaning robot;

[0035] The adjustment module is used to embed a monitoring mechanism into the task chain, update the snapshot, and adjust the task chain, wherein the adjustment at least includes: merging and splitting.

[0036] Furthermore, the calculation module includes:

[0037] a marking unit, configured to obtain the equipment distribution in the cabin, divide the restricted area, generate a virtual fence based on the restricted area, and mark the virtual fence in the snapshot;

[0038] a determination unit, configured to determine whether the real-time position exceeds a virtual fence, and if so, trigger a preset alarm mechanism;

[0039] An output unit is used to collect historical data of residues in the cabin, annotate and preprocess them, generate a training set, create a target detection model, optimize it using the training set, input the snapshots into the target detection model, and output foreground features and background features;

[0040] The reading unit is used to determine the reference point, define the origin and axis direction, correct the snapshot, integrate the origin and axis coordinates, construct a coordinate system, use the edge detection algorithm to delineate the boundary of the foreground feature in each block, and read the edge coordinates of the boundary.

[0041] Furthermore, the deletion module includes:

[0042] A clustering unit, configured to configure a scanning priority for each block and set a scanning frequency, and cluster the reflected light signal into a smooth portion and a rough portion;

[0043] The verification unit is used to determine whether the reflected light signal at the edge coordinate is a rough portion, and if so, the verification is passed.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] By collecting video surveillance data and dividing it into several blocks, the scope of use can be managed in a refined manner, greatly improving the accuracy of clearing warehouses. By taking snapshots, a data basis and clearing guidance can be provided for the clearing robots. At the same time, the clearing robots can be monitored in real time and emergencies can be handled in a timely manner. By generating task chains, the smoothness of clearing tasks can be greatly improved, and the efficiency of clearing warehouses can be improved. By calculating edge coordinates, the location of residues can be identified, and the clearing path can be optimized, which greatly improves the operation accuracy and clearing efficiency. By verifying the residues, the clearing effect can be further improved, errors and repeated clearing can be reduced, and the clearing process can be ensured to be more accurate and efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A flowchart of a remote monitoring method for a warehouse cleaning robot based on the Internet of Things provided by an embodiment of the present invention;

[0047] Figure 2 A block diagram of the first sub-process of the remote monitoring method for a warehouse cleaning robot based on the Internet of Things provided in an embodiment of the present invention;

[0048] Figure 3 A block diagram of the second sub-process of the remote monitoring method for a warehouse cleaning robot based on the Internet of Things provided by an embodiment of the present invention;

[0049] Figure 4 A block diagram of the third sub-process of the remote monitoring method for a warehouse cleaning robot based on the Internet of Things provided by an embodiment of the present invention;

[0050] Figure 5 A fourth sub-flow diagram of the remote monitoring method for a warehouse cleaning robot based on the Internet of Things provided in an embodiment of the present invention;

[0051] Figure 6 A block diagram of the composition of a warehouse cleaning robot remote monitoring system based on the Internet of Things provided by an embodiment of the present invention;

[0052] Figure 7 A block diagram of the composition of the computing module in the remote monitoring system for warehouse cleaning robots based on the Internet of Things provided by an embodiment of the present invention;

[0053] Figure 8 A block diagram of the deletion module in the remote monitoring system for warehouse cleaning robots based on the Internet of Things provided by an embodiment of the present invention;

[0054] Figure 9 A block diagram of the components of the sending module in the remote monitoring system for a warehouse cleaning robot based on the Internet of Things provided by an embodiment of the present invention;

[0055] Figure 10 This is a block diagram of the composition of the adjustment module in the IoT-based warehouse clearance robot remote monitoring system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0057] In Example 1, Figure 1 The implementation process of the remote monitoring method for a warehouse cleaning robot based on the Internet of Things provided by an embodiment of the present invention is shown and described in detail below:

[0058] S100: Utilize the preset IoT video surveillance equipment to collect video surveillance data in the cabin, and take snapshots to define the use range of the clearance robot. The use range is divided into several blocks, and nodes corresponding to the blocks are created. The snapshots are synchronized to the corresponding nodes, and the foreground features in the snapshots are extracted and overwritten. All nodes are integrated to generate a task chain, construct a position coordinate system, and use an edge detection algorithm to calculate the edge coordinates of the foreground features in each node.

[0059] The IoT video surveillance equipment installed on the deck of the transport ship collects video surveillance data from the cabin, extracts key frames, and generates snapshots, where the snapshots are state diagrams after cargo loading and unloading. The scope of use of the clearance robot in the cabin is determined, and the scope of use is the area that needs to be cleared. The scope of use is divided into several blocks, a node is created for each block, and the snapshot part corresponding to the block is synchronized to the node. Image processing technology is used to extract foreground features in the snapshot, where the foreground features are the foreground part in the snapshot, and the foreground part is used to cover the snapshot in the node. All nodes are integrated to generate a task chain to ensure that each block can be cleared in sequence. A global position coordinate system is constructed to determine the spatial relationship of each block. Edge detection algorithms (such as Canny or Sobel algorithms) are used to process the foreground features contained in each node, and the edge coordinates of the foreground features are calculated and extracted.

[0060] S200: Obtain the use authority of the laser generating device pre-integrated on the clearance robot, read the reflected light signal at the edge coordinates, and perform residue verification on all nodes in turn. If the verification fails, the corresponding node will be deleted from the task chain.

[0061] Deploy a laser generating device in the clearance machine, obtain the use permission of the laser generating device, activate the laser generating device with the use permission, emit the laser, emit the laser to the edge coordinate point by point, record the intensity and pattern of the reflected light signal, and combine the preset signal threshold to determine whether there is any residue. If there is no residue in the block, delete the corresponding node from the task chain.

[0062] S300: Locate the real-time position of the warehouse-clearing robot, generate a moving path with the real-time position as the starting point and the preset position as the end point, generate waypoints using the edge coordinates, insert the waypoints into the moving path, and send the moving path to the warehouse-clearing robot.

[0063] Using video surveillance equipment, the real-time position of the clearance robot in the position coordinate system is determined, and a moving path is generated with the real-time position as the starting point and the preset position as the end point. The preset position can be the starting point, the charging position, or the end position, etc. On this basis, the edge coordinates are determined as the waypoints, the moving path is adjusted, and the adjusted moving path is sent to the clearance robot.

[0064] S400: Embed a monitoring mechanism into the task chain, update the snapshot, and adjust the task chain, wherein the adjustment at least includes: merging and splitting.

[0065] By calling IoT video surveillance equipment in real time, the task area is dynamically monitored. The monitoring mechanism is as follows: real-time monitoring of the clearance robot to determine whether it has deviated, encountered obstacles, or suffered mechanical failures; snapshots are updated at a preset frequency, and the task chain is adjusted. Specific adjustment methods include: deleting processed blocks, merging blocks, and reordering them.

[0066] In Example 2, Figure 2 The implementation process of the remote monitoring method for a warehouse cleaning robot based on the Internet of Things provided by an embodiment of the present invention is shown. The following details the steps of using a preset Internet of Things video monitoring device to collect video monitoring data in the cabin, extract snapshots, define the usage range of the warehouse cleaning robot, and divide the usage range into several blocks.

[0067] S101: Obtain the equipment distribution in the cabin and divide the prohibited area into two areas. Generate a virtual fence based on the prohibited area and mark the virtual fence in the snapshot.

[0068] Use IoT video surveillance equipment to determine the distribution of equipment in the cabin and the restricted areas for the clearance robot in the cabin; divide the restricted areas into dynamic restricted areas (such as areas blocked due to temporary operations) and static restricted areas (such as areas with fixed dangerous equipment); generate virtual fences based on the restricted areas to prevent the clearance robot from entering the virtual fences.

[0069] S102: Determine whether the real-time position exceeds the virtual fence, and if so, trigger a preset alarm mechanism.

[0070] If the real-time position of the clearance robot exceeds the virtual fence, the preset alarm mechanism is triggered, where the alarm mechanism sends an alarm message to a preset terminal, where the preset terminal is the terminal of the person in charge of loading and unloading.

[0071] In Example 3, Figure 2 The implementation process of the remote monitoring method for a warehouse cleaning robot based on the Internet of Things provided by an embodiment of the present invention is shown. The following details the steps of creating nodes corresponding to the blocks, synchronizing the snapshots to the corresponding nodes, extracting foreground features in the snapshots, overwriting the nodes, integrating all nodes, and generating a task chain.

[0072] S103: Collect historical data of residues in the cabin, perform labeling and preprocessing, generate a training set, create a target detection model, and use the training set for optimization. Input the snapshot into the target detection model and output foreground features and background features.

[0073] Collect snapshots of the residues in the cabin at historical moments, manually annotate the location and size of the residues, strengthen the snapshots, perform data augmentation, etc., integrate all the snapshots to generate a training set.

[0074] From the existing data set, a target detection model architecture is selected and initialized. The target detection model architecture is trained and optimized using the training set to obtain a target detection model. The target detection model is mainly used to separate foreground features and background features in the snapshot.

[0075] In Example 4, Figure 2 The implementation process of the remote monitoring method for a warehouse cleaning robot based on the Internet of Things provided by an embodiment of the present invention is shown. The steps of constructing a position coordinate system and calculating the edge coordinates of the foreground features in each node using an edge detection algorithm are described in detail below.

[0076] S104: Determine a reference point, define an origin and axis directions, correct the snapshot, integrate the origin and axis coordinates, construct a coordinate system, use an edge detection algorithm to delineate the boundaries of foreground features in each block, and read the edge coordinates of the boundaries.

[0077] Determine the reference point and select a fixed point as the origin of the coordinate system, where both the reference point and the origin are equipment in the cabin. Determine the direction of the coordinate axis and construct the coordinate system; use edge detection algorithms (such as Canny or Sobel) to process the snapshots in each block, extract the foreground features in the snapshots, and delineate the boundaries of the foreground features; read the edge coordinates of the boundary from the coordinate system, where all the edge coordinates constitute the specific location of the residue.

[0078] In Example 5, Figure 3 The implementation process of the remote monitoring method for a warehouse cleaning robot based on the Internet of Things provided by an embodiment of the present invention is shown. The following details the steps of obtaining permission to use the laser generating device pre-integrated on the warehouse cleaning robot, reading the reflected light signal at the edge coordinates, and sequentially performing residue verification on all nodes.

[0079] S201: configuring a scanning priority for each block and setting a scanning frequency, and clustering the reflected light signal into a smooth portion and a rough portion.

[0080] Determine the scanning priority of each block. If the area composed of edge coordinates in the block accounts for a large proportion, a higher scanning frequency should be set.

[0081] S202: Determine whether the reflected light signal at the edge coordinate is a rough portion. If yes, the verification is successful.

[0082] The reflected light signal in each block is read and divided into a smooth portion and a rough portion, wherein the smooth portion is an area without residue and the rough portion is an area with residue.

[0083] In Example 6, Figure 4 The implementation process of the remote monitoring method for a warehouse cleaning robot based on the Internet of Things provided by an embodiment of the present invention is shown. The following details the steps of locating the real-time position of the warehouse cleaning robot, generating a movement path with the real-time position as the starting point and the preset position as the end point, generating waypoints using the edge coordinates, inserting the waypoints into the movement path, and sending the movement path to the warehouse cleaning robot.

[0084] S301: Integrate the starting point, waypoints and end point to generate a moving path, and determine deviation factors, wherein the deviation factors include at least: power, progress and obstacles.

[0085] Integrate the starting point, transit points and end point to generate a moving path; use IoT video surveillance equipment to read out deviation factors in the moving path.

[0086] S302: Calculate the deviation between the real-time position and the moving path. If the deviation is greater than a preset threshold, generate an alarm message and send the alarm message to a preset terminal.

[0087] The point where the clearance robot begins to deviate from the moving path is marked, and the distance between the real-time position and this point is calculated, defined as the deviation value, and the deviation value is monitored in real time. If the deviation value is greater than the preset threshold, an alarm message is sent to the preset terminal, where the preset terminal is the terminal of the person responsible for loading and unloading.

[0088] In Example 7, Figure 5 The implementation process of the remote monitoring method for a warehouse cleaning robot based on the Internet of Things provided by an embodiment of the present invention is shown. The steps of embedding a monitoring mechanism into the task chain, updating the snapshot, and adjusting the task chain are described in detail below:

[0089] S401: Collect fusion data from multi-level IoT sensors and integrate it into the task chain, activate the monitoring mechanism, and monitor the status of the warehouse cleaning robot.

[0090] Various IoT sensors installed in the cabin (such as temperature and humidity sensors, infrared sensors, etc.) collect data in real time and fuse the collected data to determine the operating status of the clearance robot and judge whether there are any faults such as route deviation and accidental entry into prohibited areas.

[0091] S402: Record the generation time of the snapshot, update the snapshot based on a preset frequency, integrate the status monitoring result and the snapshot, and adjust the task chain.

[0092] Update the snapshot at a preset frequency and adjust the task chain. The specific adjustment methods include: deleting processed blocks, merging and reordering blocks, etc.

[0093] Figure 6 The figure shows a structural block diagram of a warehouse cleaning robot remote monitoring system based on the Internet of Things provided by an embodiment of the present invention. The warehouse cleaning robot remote monitoring system 1 based on the Internet of Things includes:

[0094] The computing module 11 is configured to utilize a preset IoT video surveillance device to collect video surveillance data within the cabin, extract snapshots, define the usage range of the warehouse clearance robot, divide the usage range into a number of blocks, create nodes corresponding to the blocks, synchronize the snapshots to the corresponding nodes, extract foreground features from the snapshots, overlay the nodes, integrate all nodes, generate a task chain, construct a position coordinate system, and calculate the edge coordinates of the foreground features in each node using an edge detection algorithm.

[0095] The deletion module 12 is used to obtain the use permission of the laser generating device pre-integrated on the clearance robot, read the reflected light signal at the edge coordinate, and perform residue verification on all nodes in turn. If the verification fails, the corresponding node is deleted from the task chain;

[0096] The sending module 13 is used to locate the real-time position of the clearance robot, generate a movement path with the real-time position as the starting point and the preset position as the end point, generate waypoints using the edge coordinates, insert the waypoints into the movement path, and send the movement path to the clearance robot;

[0097] The adjustment module 14 is configured to embed a monitoring mechanism into the task chain, update the snapshot, and adjust the task chain, wherein the adjustment includes at least merging and splitting.

[0098] Figure 7 The following is a structural block diagram of a warehouse cleaning robot remote monitoring system based on the Internet of Things provided by an embodiment of the present invention. The computing module 11 includes:

[0099] The marking unit 111 is used to obtain the equipment distribution in the cabin and divide the restricted area, generate a virtual fence based on the restricted area, and mark the virtual fence in the snapshot;

[0100] a determination unit 112, configured to determine whether the real-time position exceeds a virtual fence, and if so, trigger a preset alarm mechanism;

[0101] The output unit 113 is used to collect historical data of residues in the cabin, annotate and preprocess them, generate a training set, create a target detection model, optimize the training set, input the snapshots into the target detection model, and output foreground features and background features;

[0102] The reading unit 114 is used to determine the reference point, define the origin and axis direction, and correct the snapshot, integrate the origin and axis coordinates, construct a coordinate system, use the edge detection algorithm to delineate the boundary of the foreground feature in each block, and read the edge coordinates of the boundary.

[0103] Figure 8 The following is a structural block diagram of a warehouse cleaning robot remote monitoring system based on the Internet of Things provided by an embodiment of the present invention. The deletion module 12 includes:

[0104] The clustering unit 121 is used to configure the scanning priority of each block and set the scanning frequency to cluster the reflected light signal into a smooth part and a rough part;

[0105] The verification unit 122 is configured to determine whether the reflected light signal at the edge coordinate is a rough portion, and if so, the verification is successful.

[0106] Figure 9 The structure diagram of the remote monitoring system for a warehouse cleaning robot based on the Internet of Things provided by an embodiment of the present invention is shown. The sending module 13 includes:

[0107] a deviation unit 131 for integrating the starting point, waypoints, and end point to generate a movement path and determining deviation factors, wherein the deviation factors include at least: power level, progress, and obstacles;

[0108] The sending unit 132 is configured to calculate a deviation between the real-time position and the moving path, and if the deviation is greater than a preset threshold, generate an alarm message and send the alarm message to a preset terminal.

[0109] Figure 10 The following is a structural block diagram of a warehouse cleaning robot remote monitoring system based on the Internet of Things provided by an embodiment of the present invention. The adjustment module 14 includes:

[0110] An activation unit 141 is configured to collect fused data from multi-level IoT sensors and integrate the data into the task chain, thereby activating the monitoring mechanism and performing status monitoring on the warehouse cleaning robot.

[0111] The updating unit 142 is configured to record the generation time of the snapshot, update the snapshot based on a preset frequency, and integrate the status monitoring result and the snapshot to adjust the task chain.

[0112] The calculation module 11 is mainly used to complete step S100, the deletion module 12 is mainly used to complete step S200, the sending module 13 is mainly used to complete step S300, and the adjustment module 14 is mainly used to complete step S400;

[0113] The marking unit 111 is mainly used to complete step S101, the judging unit 112 is mainly used to complete step S102, the output unit 113 is mainly used to complete step S103, and the reading unit 114 is mainly used to complete step S104;

[0114] The clustering unit 121 is mainly used to complete step S201, and the verification unit 122 is mainly used to complete step S202;

[0115] The deviation unit 131 is mainly used to complete step S301, and the sending unit 132 is mainly used to complete step S302;

[0116] The activation unit 141 is mainly used to complete step S401, and the update unit 142 is mainly used to complete step S402.

[0117] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0118] The above-described embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

[0119] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A remote monitoring method for a warehouse cleaning robot based on the Internet of Things, characterized in that: The method comprises: Using pre-set IoT video surveillance equipment, video surveillance data from the cabin is collected and captured as snapshots. This defines the range of use for the warehouse clearance robot, which is then divided into several blocks. Nodes corresponding to the blocks are created, and the snapshots are synchronized to the corresponding nodes. Foreground features in the snapshots are extracted and overlaid on the nodes. All nodes are integrated to generate a task chain, a position coordinate system is constructed, and the edge coordinates of the foreground features in each node are calculated using an edge detection algorithm. Obtain access to the laser generating device pre-integrated on the warehouse clearance robot, read the reflected light signal at the edge coordinates, and perform residue verification on all nodes in turn. If the verification fails, the corresponding node is deleted from the task chain; Locating the real-time position of the warehouse cleaning robot, generating a movement path with the real-time position as the starting point and the preset position as the end point, generating waypoints using the edge coordinates, inserting the waypoints into the movement path, and sending the movement path to the warehouse cleaning robot; Embed a monitoring mechanism into the task chain, update the snapshot, and adjust the task chain, wherein the adjustment at least includes: merging and splitting; The steps of obtaining permission to use the laser generating device pre-integrated on the clearance robot, reading the reflected light signal at the edge coordinates, and performing residue verification on all nodes in turn include: configuring the scanning priority of each block and setting the scanning frequency, clustering the reflected light signal into smooth parts and rough parts; and determining whether the reflected light signal at the edge coordinates is a rough part, and if so, passing the verification.

2. The remote monitoring method of a warehouse cleaning robot based on the Internet of Things according to claim 1 is characterized in that: The steps of using a preset IoT video surveillance device to collect video surveillance data in the cabin, taking snapshots, defining the use range of the warehouse clearance robot, and dividing the use range into a plurality of blocks include: Obtain the equipment distribution in the cabin and divide the restricted areas; Based on the restricted area, a virtual fence is generated, and the virtual fence is marked in the snapshot; Determine whether the real-time position exceeds the virtual fence, and if so, trigger a preset alarm mechanism.

3. The remote monitoring method of a warehouse cleaning robot based on the Internet of Things according to claim 2 is characterized in that: The steps of creating nodes corresponding to the blocks one by one, synchronizing the snapshots to the corresponding nodes, extracting foreground features in the snapshots, overwriting the nodes, integrating all the nodes, and generating a task chain include: Collect historical data on residues in the cabin, label and preprocess it, and generate a training set; Creating an object detection model and optimizing it using the training set; The snapshot is input into the target detection model, and the foreground features and background features are output.

4. The remote monitoring method of a warehouse cleaning robot based on the Internet of Things according to claim 1, characterized in that: The steps of constructing a position coordinate system and calculating edge coordinates of foreground features in each node using an edge detection algorithm include: Determine a reference point, define an origin and an axis direction, correct the snapshot, integrate the origin and axis coordinates, and construct a coordinate system; The edge detection algorithm is used to delineate the boundaries of the foreground features in each block, and the edge coordinates of the boundaries are read out.

5. The remote monitoring method of a warehouse cleaning robot based on the Internet of Things according to claim 1, characterized in that: The steps of locating the real-time position of the warehouse cleaning robot, generating a moving path with the real-time position as a starting point and a preset position as an end point, generating waypoints using the edge coordinates, inserting the waypoints into the moving path, and sending the moving path to the warehouse cleaning robot include: Integrate the starting point, the waypoints, and the end point to generate a movement path, and determine deviation factors, wherein the deviation factors include at least: power level, progress, and obstacles; The deviation value between the real-time position and the moving path is calculated. If the deviation value is greater than a preset threshold, an alarm message is generated and sent to a preset terminal.

6. The remote monitoring method of a warehouse cleaning robot based on the Internet of Things according to claim 1, characterized in that: The steps of embedding a monitoring mechanism into the task chain, updating the snapshot, and adjusting the task chain include: Collecting fused data from multi-level IoT sensors and integrating it into the task chain, activating the monitoring mechanism, and monitoring the status of the warehouse cleaning robot; The generation time of the snapshot is recorded, the snapshot is updated based on a preset frequency, and the status monitoring results and the snapshot are integrated to adjust the task chain.

7. A remote monitoring system for warehouse clearance robots based on the Internet of Things, characterized in that: The system comprises: A computing module is configured to utilize preset IoT video surveillance equipment to collect video surveillance data within the cabin, extract snapshots, delineate the usage range of the warehouse clearance robot, divide the usage range into a number of blocks, create nodes corresponding to the blocks, synchronize the snapshots to the corresponding nodes, extract foreground features from the snapshots, overlay the nodes, integrate all nodes, generate a task chain, construct a position coordinate system, and utilize an edge detection algorithm to calculate the edge coordinates of the foreground features in each node; A deletion module is used to obtain the use permission of the laser generating device pre-integrated on the clearance robot, read the reflected light signal at the edge coordinate, and perform residue verification on all nodes in turn. If the verification fails, the corresponding node is deleted from the task chain; a sending module, configured to locate the real-time position of the warehouse cleaning robot, generate a movement path with the real-time position as a starting point and a preset position as an end point, generate waypoints using the edge coordinates, insert the waypoints into the movement path, and send the movement path to the warehouse cleaning robot; An adjustment module for embedding a monitoring mechanism into the task chain, updating the snapshot, and adjusting the task chain, wherein the adjustment includes at least: merging and splitting; Wherein, the deletion module includes: A clustering unit, configured to configure a scanning priority for each block and set a scanning frequency, and cluster the reflected light signal into a smooth portion and a rough portion; The verification unit is used to determine whether the reflected light signal at the edge coordinate is a rough portion, and if so, the verification is passed.

8. The remote monitoring system for warehouse clearing robots based on the Internet of Things according to claim 7 is characterized in that: The calculation module includes: a marking unit, configured to obtain the equipment distribution in the cabin, divide the restricted area, generate a virtual fence based on the restricted area, and mark the virtual fence in the snapshot; a determination unit, configured to determine whether the real-time position exceeds a virtual fence, and if so, trigger a preset alarm mechanism; An output unit is used to collect historical data of residues in the cabin, annotate and preprocess them, generate a training set, create a target detection model, optimize it using the training set, input the snapshots into the target detection model, and output foreground features and background features; The reading unit is used to determine the reference point, define the origin and axis direction, correct the snapshot, integrate the origin and axis coordinates, construct a coordinate system, use the edge detection algorithm to delineate the boundary of the foreground feature in each block, and read the edge coordinates of the boundary.

Images