Method, device and robot for determining an elevator operating state

Abstract

The application discloses a kind of determination method, device and robot of elevator running state.Therein, the method includes: in the case where robot enters elevator is detected, acceleration data collected by acceleration sensor is obtained, wherein robot follows elevator movement and acceleration sensor is set in robot;Acceleration data is analyzed using identification model, and identification result is obtained, wherein identification result is used to determine the motion state of robot;According to identification result, the elevator running state of elevator is determined, wherein elevator running state information is used to control the action of robot.The application solves the technical problem that related technology uses external ultrasonic sensor or sets up camera on robot to assist robot to determine elevator running state, external environment interference is easy in detection process, resulting in inaccurate detection result.

Description

Technical Field

[0001] This application relates to the field of robotics, and more specifically, to a method, apparatus, and robot for determining the operating state of an elevator. Background Technology

[0002] With the development of artificial intelligence technology, robots are gradually being applied to tasks such as inspection, remote sensing, and rescue. In order to enable robots to be used safely in elevators, related technologies use external ultrasonic sensors or cameras installed on the robots to help them judge the elevator's operating status. However, the detection process is easily affected by external environmental interference. For example, at high floors, the ranging capability of ultrasonic sensors is severely attenuated, and the camera's field of view may be obstructed in the elevator, leading to inaccurate detection results.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This application provides a method, apparatus, and robot for determining the operating status of an elevator, at least to solve the technical problem that related technologies rely on external ultrasonic sensors or cameras mounted on the robot to assist in judging the elevator's operating status, which are easily affected by external environmental interference during the detection process, leading to inaccurate detection results.

[0005] According to one aspect of the embodiments of this application, a method for determining the operating state of an elevator is provided, comprising: when a robot is detected to have entered the elevator, acquiring acceleration data collected by an acceleration sensor, wherein the robot moves with the elevator and the acceleration sensor is installed in the robot; analyzing the acceleration data using a recognition model to obtain a recognition result, wherein the recognition result is used to determine the motion state of the robot; and determining the elevator operating state based on the recognition result, wherein the elevator operating state information is used to control the actions of the robot.

[0006] In some embodiments of this application, the recognition model includes an input layer, a recognition network, and an output layer. The recognition network includes a first hidden layer and a second hidden layer. The recognition model is used to analyze acceleration data to obtain recognition results, including: obtaining a preset number of acceleration sequence data prior to the acquisition time of the acceleration data through the input layer, and the historical recognition results corresponding to the previous recognition network of the current recognition network, to obtain input data; extracting features from the input data through the first node in the first hidden layer of the recognition network to obtain a first feature vector, the first hidden layer including multiple first nodes; performing feature mapping on the first feature vector through the second node in the second hidden layer of the recognition network to obtain a second feature vector, the second hidden layer including multiple second nodes, the number of second nodes being greater than the number of first nodes; converting the second feature vector into the robot's motion state through the output layer, and determining the recognition result based on the robot's motion state.

[0007] In some embodiments of this application, feature extraction of input data by identifying the first node in the first hidden layer of the network to obtain a first feature vector includes: obtaining a first weight set corresponding to each first node, the first weight set including the weights corresponding to each acceleration sequence data and the weights corresponding to historical recognition results; weighted summation of the weights in the first weight set with the corresponding data in the input data to obtain a first result; and determining a first feature vector based on the first result, the first feature vector being used to represent the output of all first nodes.

[0008] In some embodiments of this application, a second feature vector is obtained by feature mapping of the first feature vector to the second node in the second hidden layer of the identification network. This includes: obtaining a second weight set corresponding to each second node, the second weight set including the weights corresponding to the outputs of all first nodes respectively; weighting and summing the weights in the second weight set with the outputs of the first nodes respectively in the first feature vector to obtain a second result; and determining the second feature vector based on the second result, the second feature vector being used to represent the outputs of all second nodes.

[0009] In some embodiments of this application, the identification model includes multiple identification networks, which share the same number and parameters of first and second nodes. In each identification network, the parameters differ between the multiple first nodes and between the multiple second nodes.

[0010] In some embodiments of this application, an identification model is used to analyze acceleration data to obtain identification results, including: for a target identification network among multiple identification networks, the target identification network is used to analyze acceleration data to obtain a first identification result and a second identification result, wherein the first identification result is used to reflect the motion state of the robot, and the second identification result is used as the input of the next identification network of the target identification network, and the target identification network is any one of the multiple identification networks.

[0011] In some embodiments of this application, after determining the elevator's operating status based on the identification results, the method further includes: if the elevator operating status information indicates that the elevator is stationary, determining whether the robot meets preset conditions; if the robot meets the preset conditions, controlling the robot to move out of the elevator and stopping the collection of acceleration data.

[0012] According to another aspect of the embodiments of this application, an elevator operating state determination device is also provided, comprising: an acquisition module, configured to acquire acceleration data collected by an acceleration sensor when a robot is detected entering the elevator, wherein the robot follows the elevator and the acceleration sensor is disposed within the robot; an identification module, configured to analyze the acceleration data using an identification model to obtain an identification result, the identification result being used to determine the motion state of the robot; and a determination module, configured to determine the elevator operating state based on the identification result, the elevator operating state information being used to control the robot's actions.

[0013] According to another aspect of the embodiments of this application, a robot is also provided, including: an acceleration sensor, a moving mechanism, a memory, and a processor. The acceleration sensor is used to collect acceleration data, the moving mechanism is used to control the robot to enter and exit an elevator, and the memory is used to store program instructions. The processor is connected to the memory and is used to execute the method for determining the elevator operating state as described above.

[0014] According to another aspect of the embodiments of this application, a non-volatile storage medium is also provided, the non-volatile storage medium including a stored computer program, wherein the device containing the non-volatile storage medium executes the above-mentioned method for determining the elevator operating state by running the computer program.

[0015] According to another aspect of the embodiments of this application, a computer program product is also provided, including computer instructions that, when executed by a processor, implement the above-described method for determining the elevator operating state.

[0016] In this embodiment, by analyzing the data from the robot's built-in accelerometer, the operating status of the elevator can be accurately identified. This achieves the goal of accurately capturing subtle changes in acceleration data through deep learning of the recognition network, thereby improving the accuracy of elevator operating status identification and enhancing the robot's safety and autonomy in the elevator environment. This also solves the technical problem that related technologies that use external ultrasonic sensors or cameras on the robot to assist in judging the elevator operating status are easily affected by external environmental interference during the detection process, leading to inaccurate detection results. Attached Figure Description

[0017] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0018] Figure 1 This is a hardware structure block diagram of a computer terminal for a method of determining the operating status of an elevator according to an embodiment of this application;

[0019] Figure 2 This is a flowchart of a method for determining the operating status of an elevator according to an embodiment of this application;

[0020] Figure 3 This is a schematic diagram of the identification model structure of a method for determining the operating status of an elevator according to an embodiment of this application;

[0021] Figure 4 This is a schematic diagram of the identification network structure of a method for determining the operating status of an elevator according to an embodiment of this application;

[0022] Figure 5 This is a schematic diagram of the identification result of a method for determining the operating status of an elevator according to an embodiment of this application;

[0023] Figure 6 This is a schematic diagram of the overall process of a method for determining the operating status of an elevator according to an embodiment of this application;

[0024] Figure 7 This is a schematic diagram of the structure of an elevator operation status determination device according to an embodiment of this application. Detailed Implementation

[0025] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0026] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0027] To better understand the embodiments of this application, the technical terms involved in the embodiments of this application are explained below:

[0028] ARM-based Industrial Computer (AIC): An industrial control computer based on the ARM architecture, characterized by low power consumption, high performance and high integration. It is widely used in embedded systems, Internet of Things (IoT) devices and robot control, and can handle complex control logic and data processing tasks.

[0029] Mobility Mechanism: A component of a robot responsible for its physical movement, including walking, rolling, crawling, and flying. Common forms include wheeled, legged, and tracked mechanisms. The design of the mobility mechanism directly affects the robot's flexibility, stability, and efficiency, and is a key component enabling the robot to perform specific tasks.

[0030] In the inspection work of modern data center (IDC) rooms, robots are tasked with autonomously using elevators to inspect multiple floors in order to improve inspection efficiency. The key to a robot's successful entry and exit from elevators lies in its ability to accurately determine whether the elevator is stopped. This challenge not only tests the robot's intelligence but also concerns the safety and stability of the inspection process.

[0031] Traditional elevator status detection methods primarily rely on listening to and receiving signals from the elevator control system. However, this reliance on external system notifications poses potential safety hazards. If the external system provides incorrect signals, the robot might perform actions such as entering or exiting the elevator during operation, leading to unpredictable dangers. Therefore, robots need to possess the ability to independently perceive the elevator's operating status as a second layer of protection in the event of data corruption in the elevator control system.

[0032] In related technologies, the main methods for independently judging the robot's state during elevator ride without relying on the elevator's own elevator control system are: (1) by installing ultrasonic sensors in the elevator shaft, directly measuring the elevator's position information through ultrasonic sensors, and thus judging the elevator's operating state; (2) by installing cameras, using deep learning technology to identify the opening and closing status of the elevator doors in the visual image, thereby actually judging the current elevator door status and indirectly obtaining the current elevator's operating state. Of these two technical routes, firstly, installing sensor equipment in the elevator shaft will bring significant construction costs, be difficult to construct, and inconvenient to use. In addition, the shaft environment varies, and the ranging capability of ultrasonic sensors is severely attenuated at high floors. Secondly, although the method of using deep network technology by installing visual cameras on the robot body does not have the problem of decreased detection accuracy at high floors, there is a problem that the current detection results may fail when the camera's field of view is blocked by the elevator.

[0033] To address the aforementioned technical problems, this application provides corresponding solutions, which are detailed below.

[0034] The method for determining the elevator operating status provided in this application can be executed on a mobile terminal, computer terminal, or similar computing device. Figure 1 A hardware block diagram of a computer terminal for determining the operating status of an elevator is shown. Figure 1 As shown, the computer terminal 10 may include one or more processors (shown as 102a, 102b, ..., 102n in the figure) (the processor may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 104 for storing data, and a transmission module 106 for communication functions connected via wired and / or wireless networks. In addition, it may also include: a display, a keyboard, a cursor control device, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the I / O interface), a network interface, and a BUS bus. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0035] It should be noted that the aforementioned one or more processors and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be implemented wholly or partially as software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be wholly or partially integrated into any other element in the computer terminal 10. As involved in the embodiments of this application, the data processing circuits serve as processor control (e.g., selection of a variable resistor termination path connected to an interface).

[0036] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the elevator operating state determination method in this embodiment. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the aforementioned elevator operating state determination method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0037] The transmission module 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 10. In one example, the transmission module 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission module 106 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0038] The display may be, for example, a touchscreen liquid crystal display (LCD) that allows the user to interact with the user interface of the computer terminal 10.

[0039] It should be noted here that, in some optional embodiments, the above... Figure 1The computer terminal shown may include hardware elements (including circuitry), software elements (including computer code stored on a computer-readable medium), or a combination of both hardware and software elements. It should be noted that... Figure 1 This is only one instance of a specific particular instance, and is intended to illustrate the types of components that may exist in the aforementioned computer terminal.

[0040] In the above operating environment, this application provides an embodiment of a method for determining the operating state of an elevator. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0041] Figure 2 This is a flowchart of a method for determining the operating state of an elevator according to an embodiment of this application, such as... Figure 2 As shown, the method includes the following steps:

[0042] Step S202: When the robot is detected to have entered the elevator, the acceleration data collected by the acceleration sensor is acquired, wherein the robot moves with the elevator and the acceleration sensor is installed in the robot.

[0043] In step S202 above, in the IDC data center inspection scenario, the main purpose of the robot entering the elevator is to achieve autonomous movement between floors. When the robot follows the elevator, whether the elevator is stationary, accelerating upwards, decelerating downwards, or translating between floors, the robot will move synchronously with the elevator. Therefore, the robot's acceleration will directly reflect the elevator's motion state.

[0044] An accelerometer can be mounted at the center of the robot's bottom to capture changes in the robot's acceleration in three dimensions (X, Y, and Z axes). During elevator operation, the accelerometer can record the robot's acceleration data as it moves with the elevator. For example,

[0045] (1) Elevator stationary state: When the elevator stops at a certain floor, the robot's acceleration will be close to the Earth's gravitational acceleration (approximately 9.8 m / s²). 2 The acceleration on the Z-axis is relatively stable, while the acceleration on the X and Y axes should be close to zero, indicating no change in acceleration in the horizontal direction.

[0046] (2) Elevator accelerating upward: When the elevator starts to rise, since the acceleration is proportional to the elevator's upward speed, the robot's acceleration sensor will detect a positive Z-axis acceleration, which will be higher than the gravitational acceleration in the stationary state. This indicates that the elevator is accelerating upward.

[0047] (3) Elevator deceleration and descent state: When the elevator descends from the upper floor and gradually decelerates, the robot's acceleration sensor will measure a negative Z-axis acceleration, indicating that the elevator is decelerating and descending, and the magnitude of the value reflects the degree of deceleration.

[0048] (4) Elevator translation state: If the elevator moves horizontally between floors (such as slight shaking when level), the acceleration sensor will capture the corresponding acceleration changes on the X and Y axes, while the acceleration on the Z axis should remain close to the gravitational acceleration of the stationary state.

[0049] Once the robot enters the elevator, the accelerometer begins collecting acceleration data at a fixed frequency (e.g., 100Hz). This data is then input into the robot's internal memory, forming a data sequence. To accurately determine the elevator's operating state, the robot can analyze acceleration data over a period of time, rather than a single instantaneous reading. For example, the robot can feed data from the five most recent sampling times (i.e., the sequence data) into a deep neural network model for analysis. These five data points provide sufficient information to determine whether the elevator is accelerating, decelerating, or in a stable state.

[0050] Step S204: The acceleration data is analyzed using an identification model to obtain identification results, which are used to determine the motion state of the robot.

[0051] In step S204 above, the recognition model refers to a deep neural network model that uses acceleration data to analyze and classify the elevator's operating state. Through training, this model can learn and understand the characteristics of the acceleration data, thereby accurately predicting whether the robot is stationary or in motion.

[0052] In various complex elevator environments, to extract deeper features from acceleration data and improve the accuracy and robustness of elevator operation status recognition, in some embodiments of this application, the recognition model includes an input layer, a recognition network, and an output layer. The recognition network includes a first hidden layer and a second hidden layer. The recognition model is used to analyze acceleration data to obtain recognition results, including: obtaining a preset number of acceleration sequence data prior to the acquisition time of the acceleration data through the input layer, and the historical recognition results corresponding to the previous recognition network of the current recognition network, to obtain input data; extracting features from the input data through the first node in the first hidden layer of the recognition network to obtain a first feature vector, the first hidden layer including multiple first nodes; performing feature mapping on the first feature vector through the second node in the second hidden layer of the recognition network to obtain a second feature vector, the second hidden layer including multiple second nodes, the number of second nodes being greater than the number of first nodes; converting the second feature vector into the robot's motion state through the output layer, and determining the recognition result based on the robot's motion state.

[0053] The input layer is the starting part of the recognition model, used to receive raw data from the accelerometer and historical recognition results from the previous recognition network. For example, the input layer acquires the acceleration data at the current time t, as well as data from the previous four time points, and simultaneously receives the output results from the previous recognition network. The input layer in the recognition model receives more than just the acceleration data at the current time; the model can process time-series data, capturing the trend of acceleration data changes over time, thereby more accurately recognizing the robot's motion state.

[0054] The first hidden layer is the first processing layer in the model (recognition network). It contains multiple first nodes, which are used to perform preliminary feature extraction on the input data. For example, key patterns and features in acceleration data can be identified through weighted sums and activation functions (such as ReLU). Each of the multiple first nodes performs a weighted summation of the input data with different weight parameters, and then transforms it through an activation function to extract different feature representations.

[0055] The second hidden layer, located after the first hidden layer, contains more nodes than the first hidden layer. It performs deeper feature mapping on the first feature vector, thereby capturing more complex patterns. By setting more nodes than the first layer, the model can distinguish between stationary and non-stationary states. Specifically, the nodes in the second hidden layer learn the non-stationary signal changes in acceleration data during elevator start-up and stop, distinguishing between stationary and non-stationary operation (abrupt acceleration signals). In a stationary state, the acceleration signal approximates gravitational acceleration, while in a non-stationary state, the acceleration signal deviates from gravitational acceleration, exhibiting significant fluctuations. Furthermore, this setup facilitates dynamic acceleration trend analysis. Specifically, during elevator operation, the acceleration signal changes with acceleration, deceleration, and stopping. More nodes allow the model to analyze acceleration signal trends. For example, the model can learn the acceleration change patterns throughout the entire process of the elevator accelerating from a standstill to a running speed and then decelerating to a stop. This trend analysis helps to more accurately identify the robot's motion state.

[0056] The output layer is the final layer of the recognition model, responsible for converting the second feature vector output by the second hidden layer into the robot's motion state. For example, it is converted into a value between 0 and 1 using an activation function (such as sigmoid). If the output value is greater than 0.8, the robot determines that it is in a running state; if it is less than or equal to 0.8, especially close to 0, the robot determines that it is in a stationary state.

[0057] To analyze acceleration data from different angles and depths, a multi-layer recognition network can be constructed. Specifically, the recognition model includes multiple recognition networks, which share the same number and parameters of first and second nodes. Within each recognition network, the parameters differ between the multiple first nodes and between the multiple second nodes.

[0058] Each recognition network is an independent processing unit in the recognition model, responsible for analyzing acceleration data and outputting the elevator's operating status. In the design of multi-layer recognition networks, each network can focus on different aspects or levels of acceleration data, thereby improving the overall analytical capability and accuracy of the model.

[0059] The first and second nodes are located in the first and second hidden layers of the recognition network, respectively. Multiple recognition networks share the same number and parameters for the first and second nodes; that is, in a multi-layer recognition network, the nodes in the first and second hidden layers use the same parameter configuration (such as weights and biases) across all networks. For example, a recognition model consisting of 5 recognition networks can be constructed, with each network having 50 nodes in its first hidden layer and 100 nodes in its second hidden layer. Although the number of nodes and parameter structure remain consistent across all networks, the node parameters (weights and biases) of each network are individually adjusted during the training process to adapt to the characteristics and variation patterns of the acceleration data at different time points.

[0060] The design of multi-layer recognition networks enables the model to analyze acceleration data from different angles and depths. The first layer of the recognition network may focus on the basic features of the acceleration data, such as changes in signal amplitude and frequency. As the network depth increases, subsequent recognition networks can learn more advanced and abstract features, such as the time-series trends of the signal, details of non-stationary signals, and abnormal patterns in elevator operation. For example, acceleration data is first input into the first recognition network, processed by the first and second hidden layers, and outputs a preliminary result for identifying the robot's motion state. Subsequently, this result, along with the original acceleration data, is input into the next recognition network for further in-depth feature analysis and state determination. This progressive processing approach helps the model gradually accumulate information about the robot's motion state, and the final recognition result is derived from the comprehensive analysis of multiple recognition networks.

[0061] When constructing a multi-layer recognition network, the acceleration data can be analyzed through the following steps to obtain the recognition result: For the target recognition network among the multiple recognition networks, the acceleration data is analyzed using the target recognition network to obtain the first recognition result and the second recognition result. The first recognition result is used to reflect the motion state of the robot, and the second recognition result is used as the input of the next recognition network of the target recognition network. The target recognition network can be any one of the multiple recognition networks.

[0062] The first recognition result refers to the main output of each recognition network, which reflects the robot's motion state at a specific point in time, i.e., whether the elevator is stationary or running. After all recognition networks have processed the input data sequence, the model fuses the first recognition results of each network to determine the final recognition result.

[0063] The second identification result refers to the auxiliary result generated by each identification network, which is used as input for the next identification network. This facilitates information transfer between networks, enabling the subsequent network to perform more in-depth processing and judgment based on the analysis results of the previous network.

[0064] In some embodiments of this application, the first feature vector is obtained by extracting features from the input data through the first node in the first hidden layer of the identification network. Specifically, the first weight set corresponding to each first node is obtained, and the first weight set includes the weight corresponding to each acceleration sequence data and the weight corresponding to the historical identification results; the weights in the first weight set are weighted and summed with the corresponding data in the input data to obtain a first result; the first feature vector is determined based on the first result, and the first feature vector is used to represent the output of all first nodes.

[0065] During model training, the first node learns a set of weights (the first weight set), including weights for different time points in the acceleration sequence data and weights from the output of the previous recognition network. The output of the first node can be calculated using the following formula:

[0066] n output_i =RELU(n i1 *x t-4 +n i2 *x t-3 +n i3 *x t-2 +n i4 *x t-1 +n i5 *x t +

[0067] n i6 *B t-1 )(Formula 1)

[0068] Where, n output_i Indicates each n i The output of (the first node), which is the result obtained after the first result is transformed by the activation function, n i1 n i2 n i3 n i4 n i5 n i6 For n i The weights of the nodes, i.e., the first set of weights, n i1 up to n i5 The n represents the weights corresponding to the acceleration sequence data. i6 This represents the weight corresponding to the historical recognition results.

[0069] The first hidden layer consists of multiple first nodes, each of which independently performs weighted summation and feature extraction on the input data. These first nodes can process data in parallel. The outputs of all the first nodes in the first hidden layer are combined to form a first feature vector. This first feature vector can be regarded as a compact representation of the acceleration data and historical recognition results, capturing the key features of the time series data, which facilitates deeper feature mapping in the second hidden layer.

[0070] In some embodiments of this application, a second feature vector is obtained by feature mapping of the first feature vector through the second node in the second hidden layer of the identification network. Specifically: a second weight set corresponding to each second node is obtained, the second weight set including the weights corresponding to the outputs of all first nodes respectively; the weights in the second weight set are weighted and summed with the outputs of the first nodes respectively in the first feature vector to obtain a second result; the second feature vector is determined based on the second result, and the second feature vector is used to represent the outputs of all second nodes.

[0071] Each second node has a set of weights (the second weight set) used to determine the relative importance of the outputs of different first nodes in the first feature vector. The output of the second node can be calculated using the following formula:

[0072]

[0073] Where, n output_k Indicates each n i The output of (first node), m output_i Indicates each m i The output of node (the second node), m ik This indicates that the output of the first node for m i The weight of a node.

[0074] The second hidden layer processes the first feature vector in parallel by multiple second nodes, with each node independently performing weighted summation and feature mapping. This parallel processing mechanism enables the model to extract features from multiple angles and depths, increasing the richness of the feature representation. Through interactions between nodes (e.g., by connecting to the same output node), the second hidden layer achieves information fusion, ensuring that the second feature vector can comprehensively reflect the complexity of the elevator state.

[0075] After obtaining the output of each second node, the first recognition result reflecting the robot's motion state can be calculated using the following formula:

[0076]

[0077] Among them, h output_i Indicates each h iThe output, i.e., the first recognition result, h ik m represents the output weight of each second node. output_k This represents the output of each second node.

[0078] The second recognition result, which serves as the input to the next recognition network of the recognition network, can be calculated using the following formula:

[0079]

[0080] b output_i Indicate each b i The output, i.e., the second recognition result, b ik m represents the output weight of each second node. output_k This represents the output of each second node.

[0081] The activation functions ReLU, sigmoid, and tanhx in Equations 1 to 4 above are defined as follows:

[0082]

[0083]

[0084]

[0085] Figure 3 This is a schematic diagram of the identification model structure of a method for determining the operating state of an elevator according to an embodiment of this application, as shown below. Figure 3 As shown, in some embodiments of this application, the model utilizes a recursive structure to perform deep analysis of accelerometer data to identify the elevator's operating status. Specifically, the figure illustrates the process by which data, starting from acquisition point xt, is transmitted through multiple identification networks Bt, ultimately generating elevator operating status information. Figure 3In this model, the acceleration data at each time step is input into the recognition network B for the next five time steps. For example, xi is input into Bt, Bt+1, Bt+2, Bt+3, and Bt+4. Therefore, each B network receives acceleration data from its five nearest sampling times; for example, Bt receives data from xt-4, xt-3, xt-2, xt-1, and xt, where xt represents the acceleration sensor data at a given time t. This data reflects the acceleration changes in the elevator in which the robot is located. This data is then input into a series of recognition networks Bt, each containing multiple first and second nodes for feature extraction and mapping to generate a recognition result ht representing the robot's motion state. Notably, the acceleration data is not only input into the current recognition network Bt but also into subsequent recognition networks Bt+1 to Bt+4. This reflects the recursive nature of the model, meaning that the current acceleration data is used to predict the future elevator state.

[0086] Each recognition network Bt outputs two results: ht represents the robot's motion state at time t (i.e., the elevator's operating state), where a value closer to 1 indicates that the elevator is running, while a value closer to 0 indicates that the elevator is nearly stationary; bt serves as the input to the next recognition network Bt+1, reflecting the impact of historical recognition results on the model's prediction of future states. The entire model is designed based on deep learning principles, aiming to capture complex patterns in acceleration data to accurately determine the elevator's real-time operating state.

[0087] Figure 4 This is a schematic diagram of the identification network structure of a method for determining the operating state of an elevator according to an embodiment of this application, as shown below. Figure 4As shown in the figure, this diagram further refines the recognition network structure in the elevator operation status determination method, demonstrating how features are extracted from acceleration data and mapped to the elevator status output. The recognition model mainly consists of an input layer, a recognition network B (composed of a first hidden layer and a second hidden layer), and a final output layer. The input layer contains acceleration data xt and data from its four preceding time points xt-4 to xt-1, as well as the output Bt-1 from the previous recognition network. These data together constitute the input data used for analysis by the recognition network. The first hidden layer of recognition network B includes 100 first nodes (n1-n100) used to extract features from the input data and generate a first feature vector. The second hidden layer includes 300 second nodes (m1-m300), which perform more complex feature mapping on the first feature vector to generate a second feature vector. The design of the B network adopts a deep learning architecture, where each node processes data through complex connections. Each node has a specific set of weights used to perform weighted summation of the input data to determine the feature vector. The output layer ultimately converts the output of the second hidden layer into a first recognition result ht of the robot's motion state and a second recognition result bt, which is then passed to the next recognition network. This multi-layer network architecture allows the model to capture deeper features from the acceleration data, thereby improving the accuracy of elevator state recognition. The setting of the number of nodes in the recognition network (e.g., the number of nodes in the second hidden layer is greater than the number of nodes in the first hidden layer) is intended to improve its processing power and recognition accuracy by increasing the complexity of the model.

[0088] Step S206: Determine the elevator's operating status based on the recognition results, wherein the elevator operating status information is used to control the robot's actions.

[0089] In step S206 above, based on the elevator's operating status information, the robot can make corresponding action decisions, such as waiting for the elevator to come to a stop before entering or exiting, or maintaining a safe distance while the elevator is running. Controlling the robot's actions involves planning and executing the robot's behavior, ensuring that the robot can react reasonably according to the environmental conditions and guaranteeing the safe conduct of the inspection task.

[0090] In this application, a recognition model is first used to obtain recognition results to determine the robot's motion state. Then, the elevator's operating state is determined based on the recognition results. Although the robot's motion state can be largely consistent with the elevator's operating state in some cases, there are errors between the obtained robot motion state and the actual elevator operating state in certain situations (such as when the robot falls or other external environmental interference). To reduce erroneous judgments and ensure the robot's safe operation within the elevator even under external interference or sensor malfunctions, the following steps can also be performed:

[0091] Set a time window length, such as 1-2 seconds. The time window length can be determined based on the sampling frequency of the accelerometer and the average motion cycle of the elevator. Collect recognition results within the time window and statistically analyze the distribution and trend of these results to obtain a first result. For example, the trend can be quantified using a value of 0-1. If the fluctuation of the recognition results is large or the state changes frequently (such as frequently jumping from a stationary state to a running state and then back to a stationary state), the first result can be determined as 1. Obtain a historical state queue maintained internally by the robot, which records the elevator's operating state over a period of time. Compare the current recognition result with the states in the historical state queue to determine the difference between the current recognition result and the historical state, obtaining a second result. For example, the difference can be quantified using a value of 0-1. For example, a sudden change from a long-term stationary state to a running state can determine the second result as 1. Based on the first result (time window analysis) and the second result (historical state comparison), determine the recognition result used to determine the elevator's operating state. For example, filter out recognition results with errors.

[0092] It should be noted that, in order to save computing resources, in some embodiments of this application, the identification result used to determine the elevator operating status may be determined solely based on the first result or the second result.

[0093] In some embodiments of this application, the elevator operating status can be determined based on the recognition results through the following steps: obtaining multiple recognition results within a preset time period, wherein each recognition result represents the probability that the robot's motion state belongs to the target's motion state, and the target's motion state includes: moving state and stationary state; comparing the multiple recognition results with preset thresholds respectively to obtain comparison results, wherein the comparison results include the number of recognition results that meet the preset thresholds; and determining the elevator operating status information based on the comparison results.

[0094] Figure 5 This is a schematic diagram illustrating the identification result of a method for determining the operating state of an elevator according to an embodiment of this application. In one specific embodiment, an acceleration sensor collects acceleration data, for example, 9.8121 m / s², at a specific time point from 198012.01 s to 198015.06 s. 2 These data are then input into the recognition network B, which outputs a floating-point number with a value between 0 and 1. This value reflects the elevator's operating status; the closer the value is to 1, the more likely the elevator is running, and the closer it is to 0, the more likely it is stationary. Figure 3The method specifically emphasizes how to determine whether the elevator is currently stationary based on the recognition results. Specifically, when the robot needs to determine the current elevator status, it analyzes the output of the recognition network over the previous two seconds. Since the sensor's sampling frequency is 100Hz, this means there are 200 recognition results within two seconds. If more than 20 (10%) of these 200 results have a value greater than 0.8, the robot determines that the elevator is currently running; otherwise, the elevator is judged to be stationary. This method combines the concepts of time windows and thresholds, making the elevator status determination more accurate and stable. Through statistical analysis of multiple recognition results, the model can more robustly avoid misjudgments caused by short-term data fluctuations, enhancing the reliability and practicality of elevator status determination.

[0095] After determining the elevator's operating status based on the recognition results, the following steps can be performed: if the elevator's operating status information indicates that the elevator is stationary, determine whether the robot meets the preset conditions; if the robot meets the preset conditions, control the robot to move out of the elevator and stop collecting acceleration data.

[0096] Preset conditions refer to a series of safety and operational requirements that a robot must meet before it can be moved out of an elevator. These conditions may include, but are not limited to: the elevator being confirmed to be stationary; the elevator doors being fully open; the robot being positioned appropriately within the elevator doors for safe exit; the robot confirming that there are no obstacles in the surrounding environment; and the robot itself being in a ready state and capable of smooth movement.

[0097] In some embodiments of this application, the robot can obtain information about the current elevator floor and compare it with the floor to be reached in the task instructions. The floor information can be obtained by communicating with the elevator control system or indirectly through a display screen or sound prompts inside the elevator. When the elevator is stationary, the robot compares the current floor with the target floor to confirm whether it is on the predetermined floor. It also determines whether the elevator door is open. After confirming that the floor is correct and the elevator door is fully open, the robot exits the elevator using a moving mechanism.

[0098] To enable robots to utilize accelerometer data, combined with analysis of historical elevator operating status, to predict elevator stops in advance and prepare for pre-action actions, ensuring a rapid and safe exit when the elevator stops, thus improving the robot's inspection efficiency and safety in scenarios such as IDC server rooms, the following steps can also be performed:

[0099] (1) Determine the trend of acceleration data by collecting acceleration data. For example, a sliding time window (e.g., 5 seconds) can be used to analyze the trend of acceleration data, observe the slope of acceleration change and the distribution of acceleration values. By setting a prediction threshold (e.g., the slope of acceleration change is less than a certain threshold and the acceleration value is continuously within a stable range), the elevator can be predicted to stop (i.e., be in a stationary state) in advance.

[0100] (2) Determine the acceleration change pattern using historical elevator operating status. For example, the robot stores acceleration data and recognition results for a recent period of time to form a historical elevator operating status database. Analyze the historical data to identify the acceleration change pattern when the elevator starts, runs and stops. In some embodiments of this application, the prediction threshold in step (1) can also be adjusted according to the determined acceleration change pattern. For example, if historical data shows that the slope of the acceleration reduction of the elevator before stopping is usually within a certain target range, the prediction threshold can be set near the target range (e.g., the prediction threshold belongs to the target range).

[0101] (3) When the acceleration data within a preset time period (e.g., 1-2 seconds) meets the threshold condition for predicted stopping, and the historical data trend analysis also supports the prediction that the elevator is about to stop, a pre-action is triggered. The pre-action may include adjusting the moving mechanism to move towards the elevator door, sending a request to the elevator control system to obtain floor information, and issuing a reminder to passengers that the elevator is about to exit through a voice synthesis device.

[0102] Through steps S202 to S206 above, by analyzing the data from the robot's built-in accelerometer, the operating status of the elevator can be accurately identified. This achieves the goal of accurately capturing subtle changes in acceleration data through deep learning of the recognition network, thereby improving the accuracy of elevator operating status identification and enhancing the robot's safety and autonomy in the elevator environment. It also solves the technical problem that related technologies that use external ultrasonic sensors or cameras on the robot to assist in judging the elevator operating status are easily affected by external environmental interference during the detection process, leading to inaccurate detection results.

[0103] Figure 6 This is a schematic diagram of the overall process of a method for determining the operating status of an elevator according to an embodiment of this application, as shown below. Figure 6 As shown, the overall process of the elevator operation status determination method can include business process control, acceleration data acquisition, identification network analysis, status judgment, and robot behavior control. Specifically:

[0104] (1) Step 1: The business process control machine starts the robot to enter the elevator, for example, when the robot needs to reach different floors by taking the elevator while performing a task.

[0105] (2) Step 2: The accelerometer on the robot body begins to collect data, which reflects the motion state of the robot (and the elevator). For example, the sensor can collect data at a frequency of 100Hz to ensure the real-time and continuous nature of the data.

[0106] (3) Step 3: The collected acceleration data is input into the recognition network B for analysis. In each recognition network, the data first passes through the input layer, and then through the feature extraction and mapping of the first and second hidden layers to generate the recognition result ht of the elevator's operating status and the recognition result bt as the input of the next recognition network.

[0107] (4) Step 4: Determine the elevator's operating status based on the identification results. For example, the value range of ht can be set between 0 and 1. The elevator's running or stationary status can be obtained through this floating-point number.

[0108] (5) Step 5: When determining the elevator status, the system will consider the current business requirements and the elevator network identification results. If the current business requires the robot to exit the elevator, and the identification network determines that the elevator is stationary, then the robot will perform the exit action. Otherwise, if the elevator is determined to be still running or the robot does not need to exit, the process will return to step 2 to continue collecting acceleration data and performing status determination.

[0109] (6) Step 6 (Business Termination Step): Once the robot successfully exits the elevator, the business process will end and the detection of the identification network will also stop to save computing resources. This closed-loop control process ensures the efficient and safe movement of the robot in the elevator environment while minimizing unnecessary resource consumption.

[0110] In this application, the determination of the elevator's operating status does not rely on external equipment or additional sensors, but entirely on the data provided by the robot's own acceleration sensor. The sensor can detect the acceleration changes of the elevator during vertical movement. Through these changes, various states of the elevator, such as stationary, ascending, descending, or stopped, can be inferred. Therefore, it is not necessary to detect the elevator's environment through an external optical detection system, and it is not affected by the obstruction of people inside the elevator. The determination of the elevator's operating status is less affected by the environment.

[0111] Figure 7 This is a structural diagram of an elevator operating state determination device according to an embodiment of this application, as shown below. Figure 7 As shown, the device includes:

[0112] The acquisition module 702 is used to acquire acceleration data collected by the accelerometer when the robot is detected to have entered the elevator, wherein the robot moves with the elevator and the accelerometer is installed in the robot;

[0113] The recognition module 704 is used to analyze the acceleration data using a recognition model to obtain recognition results, wherein the recognition results are used to determine the motion state of the robot;

[0114] The determination module 706 is used to determine the elevator's operating status based on the recognition results, wherein the elevator operating status information is used to control the robot's actions.

[0115] It should be noted that, Figure 7 The elevator operating status determination device shown is used to perform... Figure 2 The method for determining the elevator's operating status is shown below, therefore Figure 2 The relevant explanations in the method for determining the elevator's operating status also apply to... Figure 7 The device for determining the elevator's operating status, as shown, will not be described in detail here.

[0116] This application also provides a robot, including: an accelerometer, a moving mechanism, a memory, and a processor. The accelerometer is used to collect acceleration data, the moving mechanism is used to control the robot to enter and exit an elevator, and the memory is used to store program instructions. The processor is connected to the memory and is used to execute the method for determining the elevator's operating state. For example, the processor executes the following functions by executing the program instructions stored in the memory: when the robot is detected to have entered the elevator, it acquires acceleration data collected by the accelerometer, wherein the robot follows the elevator and the accelerometer is located within the robot; it analyzes the acceleration data using a recognition model to obtain a recognition result, wherein the recognition result is used to determine the robot's motion state; and it determines the elevator's operating state based on the recognition result, wherein the elevator operating state information is used to control the robot's actions.

[0117] In some embodiments of this application, the robot can be an IDC data center inspection service robot, and also includes an ARM industrial control computer. An acceleration sensor is installed at the center of the bottom of the robot, and the ARM industrial control computer is used to run a neural network (i.e., a recognition network) for elevator status detection.

[0118] This application also provides a non-volatile storage medium, which includes a stored computer program, wherein the device containing the non-volatile storage medium executes the steps of the elevator operation state determination method in various embodiments of this application by running the computer program.

[0119] This application also provides a computer program product, including computer instructions, which, when executed by a processor, implement the steps of the elevator operation state determination method in various embodiments of this application.

[0120] This application also provides a computer program that, when executed by a processor, implements the steps of the elevator operation state determination method in various embodiments of this application.

[0121] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0122] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0123] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0124] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0125] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0126] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0127] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for determining the operating status of an elevator, characterized in that, include: When a robot is detected entering an elevator, acceleration data collected by an accelerometer is acquired, wherein the robot moves with the elevator and the accelerometer is installed in the robot; The acceleration data is analyzed using a recognition model to obtain a recognition result, wherein the recognition result is used to determine the motion state of the robot. The recognition model includes an input layer, a recognition network, and an output layer, and the recognition network includes a first hidden layer and a second hidden layer. The elevator's operating status is determined based on the recognition result, wherein the elevator operating status information is used to control the robot's actions; The process involves analyzing the acceleration data using a recognition model to obtain recognition results. This includes: acquiring a preset number of acceleration sequence data prior to the acquisition time of the acceleration data through the input layer, and the historical recognition results corresponding to the previous recognition network of the current recognition network, to obtain input data; extracting features from the input data through the first node in the first hidden layer of the recognition network to obtain a first feature vector, wherein the first hidden layer includes multiple first nodes; mapping the first feature vector to the first feature vector through the second node in the second hidden layer of the recognition network to obtain a second feature vector, wherein the second hidden layer includes multiple second nodes, and the number of second nodes is greater than the number of first nodes; converting the second feature vector into the motion state of the robot through the output layer, and determining the recognition result based on the motion state of the robot.

2. The method according to claim 1, characterized in that, The first feature vector is obtained by extracting features from the input data through the first node in the first hidden layer of the recognition network, including: Obtain the first weight set corresponding to each first node, wherein the first weight set includes the weight corresponding to each acceleration sequence data and the weight corresponding to the historical recognition result; The weights in the first weight set are summed with the corresponding data in the input data to obtain the first result. The first feature vector is determined based on the first result, wherein the first feature vector is used to represent the output of all first nodes.

3. The method according to claim 1, characterized in that, The first feature vector is mapped using a second node in the second hidden layer of the recognition network to obtain a second feature vector, including: Obtain the second weight set corresponding to each second node, wherein the second weight set includes the weights corresponding to the outputs of all first nodes; The weights in the second weight set are weighted and summed with the outputs of the first nodes corresponding to the first feature vectors to obtain the second result. The second feature vector is determined based on the second result, wherein the second feature vector is used to represent the output of all second nodes.

4. The method according to claim 1, characterized in that, The recognition model includes multiple recognition networks, which share the same number and parameters of first and second nodes. In each recognition network, the parameters differ between the multiple first nodes and between the multiple second nodes.

5. The method according to claim 4, characterized in that, The acceleration data is analyzed using a recognition model to obtain recognition results, including: For the target recognition network among the plurality of recognition networks, the acceleration data is analyzed using the target recognition network to obtain a first recognition result and a second recognition result. The first recognition result is used to reflect the motion state of the robot, and the second recognition result is used as the input to the next recognition network of the target recognition network. The target recognition network is any one of the plurality of recognition networks.

6. The method according to claim 1, characterized in that, After determining the elevator's operating status based on the identification result, the method further includes: When the elevator operation status information indicates that the elevator is stationary, determine whether the robot meets the preset conditions; When the robot meets the preset conditions, control the robot to move out of the elevator and stop collecting the acceleration data.

7. A device for determining the operating status of an elevator, characterized in that, include: The acquisition module is used to acquire acceleration data collected by the accelerometer when the robot is detected to enter the elevator, wherein the robot moves with the elevator and the accelerometer is installed in the robot; A recognition module is used to analyze the acceleration data using a recognition model to obtain a recognition result, wherein the recognition result is used to determine the motion state of the robot. The recognition model includes an input layer, a recognition network, and an output layer. The recognition network includes a first hidden layer and a second hidden layer. Analyzing the acceleration data using the recognition model to obtain the recognition result includes: obtaining a preset number of acceleration sequence data prior to the acquisition time of the acceleration data through the input layer, and the historical recognition result corresponding to the previous recognition network of the current recognition network, to obtain input data; extracting features from the input data through the first nodes in the first hidden layer of the recognition network to obtain a first feature vector, wherein the first hidden layer includes multiple first nodes; performing feature mapping on the first feature vector through the second nodes in the second hidden layer of the recognition network to obtain a second feature vector, wherein the second hidden layer includes multiple second nodes, and the number of second nodes is greater than the number of first nodes; converting the second feature vector into the motion state of the robot through the output layer, and determining the recognition result based on the motion state of the robot. The determination module is used to determine the elevator's operating status based on the identification result, wherein the elevator operating status information is used to control the robot's actions.

8. A robot, characterized in that, include: The system includes an accelerometer, a moving mechanism, a memory, and a processor. The accelerometer is used to collect acceleration data, the moving mechanism is used to control the robot to enter and exit the elevator, and the memory is used to store program instructions. The processor is connected to the memory and is used to execute the method for determining the elevator operating state according to any one of claims 1 to 6.

9. A non-volatile storage medium, characterized in that, The non-volatile storage medium includes a stored computer program, wherein the device containing the non-volatile storage medium executes the elevator operating state determination method according to any one of claims 1 to 6 by running the computer program.

10. A computer program product comprising computer instructions, characterized in that, When the computer instructions are executed by the processor, they implement the method for determining the elevator operating status as described in any one of claims 1 to 6.

Images