Intelligent operation site protection method and system based on high-precision positioning, device, and storage medium

Abstract

An intelligent operation site protection method and system based on high-precision positioning, a device, and a storage medium, applied to the technical field of intelligent security. The method comprises: on the basis of UWB positioning, using a multipath suppression method and a dynamic Kalman filtering method to acquire a precise location of a person or device inside or outside an operation site; using a machine learning algorithm to perform multi-modal physical sign information recognition and behavior pattern recognition on the person or device the precise location of which has been acquired; and on the basis of a recognition result, sending corresponding protection warning information. In this way, the safety protection level and safety management efficiency for operation sites can be effectively improved.

Description

Worksite intelligent protection method, system and device based on high-precision positioning and storage medium TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of intelligent security and protection, further relates to the technical field of intelligent security and protection of industrial sites, in particular to a worksite intelligent protection method, system, device and storage medium based on high-precision positioning. BACKGROUND

[0002] With the continuous expansion of industrial production scale and the rapid progress of technology, the safety protection requirements of engineering worksites are increasingly improved, and the traditional protection methods and systems gradually expose many limitations in the application of industrial sites.

[0003] The traditional protection system has low positioning accuracy in industrial sites, and it is difficult to accurately determine the specific positions of personnel and equipment, and cannot timely detect potential dangerous boundary crossing behaviors. At the same time, it cannot effectively distinguish different types of targets, and may misjudge passing animals, natural factors, etc. as dangerous situations and trigger unnecessary alarms, affecting the normal production order, and lacks intelligent judgment ability and cannot accurately identify non-malicious triggering behaviors such as temporary movement of equipment, wind blowing animals, etc., resulting in frequent false alarms and increasing the work burden of safety management personnel. TECHNICAL PROBLEM

[0004] How to improve the positioning accuracy, target distinguishing ability and intelligent judgment ability of the industrial site protection system to improve the safety protection level and safety management efficiency of the worksite and reduce the work burden of safety management personnel? TECHNICAL SOLUTION

[0005] Therefore, the present disclosure provides a worksite intelligent protection method, system, device and storage medium based on high-precision positioning.

[0006] According to a first aspect of the present disclosure, a worksite intelligent protection method based on high-precision positioning is provided, which is applied to an intelligent protection system, an intelligent electronic fence is arranged in the intelligent protection system, and the area surrounded by the intelligent electronic fence is defined as a worksite. The intelligent protection method comprises:

[0007] Based on UWB positioning, the multi-path suppression method and dynamic Kalman filtering method are used to obtain the accurate positioning of personnel or equipment inside and outside the worksite;

[0008] The machine learning algorithm is used to identify the multi-modal sign information and behavior patterns of the personnel or equipment that have obtained accurate positioning;

[0009] According to the identification result, corresponding protection warning information is sent.

[0010] In some implementations of the first aspect, based on the UWB positioning, the precise positioning of personnel or equipment inside and outside the work site is obtained by using a multipath suppression method and a dynamic Kalman filtering method, including:

[0011] Based on the UWB positioning, the multipath interference suppression method is used to suppress the multipath interference of the received signal.

[0012] Through the dynamic Kalman filtering method, the UWB signal after multipath interference suppression is used as a measurement value, and the prior estimate value is fused to improve the positioning accuracy of UWB, so as to obtain the precise positioning of personnel or equipment inside and outside the work site.

[0013] In some implementations of the first aspect, through the dynamic Kalman filtering method, the signal after multipath interference suppression is used as a measurement value, and the prior estimate value is fused to improve the positioning accuracy of UWB, so as to obtain the precise positioning of personnel or equipment inside and outside the work site, including:

[0014] Establish a state space model describing the motion state of personnel or equipment inside and outside the work site;

[0015] Obtain the motion state historical data of personnel or equipment inside and outside the work site and analyze it to determine the state transition matrix and the observation matrix;

[0016] The UWB signal after multipath interference suppression is used as a measurement value, and based on the state estimate at the historical time and the state transition matrix, the prior estimate value is calculated;

[0017] Through the Kalman gain, the measurement value and the prior estimate value are fused to improve the positioning accuracy of UWB, so as to obtain the precise positioning of personnel or equipment inside and outside the work site.

[0018] In some implementations of the first aspect, the machine learning algorithm is used to identify the multi-modal sign information and behavior pattern of the personnel or equipment that has obtained precise positioning, including:

[0019] The machine learning algorithm is used to train the identification model;

[0020] The multi-modal sign information and behavior pattern information of the personnel or equipment that has obtained precise positioning are input into the identification model for identification.

[0021] In some implementations of the first aspect, the identification model is obtained by training as follows:

[0022] Obtain a large amount of multi-modal sign information of personnel or equipment in different states and corresponding behavior pattern labeled data;

[0023] The multi-modal sign information of the personnel or the equipment in different states is taken as a sample, and the corresponding behavior mode label data is taken as a sample label to generate a training set, model training is performed until a recognition model is generated.

[0024] In some implementable manners of the first aspect, according to the recognition result, corresponding protection warning information is issued, including:

[0025] According to the recognition result, it is determined whether the state and behavior mode of the personnel or the equipment inside and outside the work site are normal, and if there is an abnormality, corresponding protection warning information is issued.

[0026] According to the second aspect of the present disclosure, a high-precision positioning-based intelligent protection system for a work site is provided, which applies the method as above, and the system comprises:

[0027] A first processing module is configured to acquire accurate positioning of the personnel or the equipment inside and outside the work site based on UWB positioning by using a multi-path suppression method and a dynamic Kalman filtering method;

[0028] A second processing module is configured to perform multi-modal sign information recognition and behavior mode recognition on the personnel or the equipment with accurate positioning by using a machine learning algorithm;

[0029] A third processing module is configured to issue corresponding protection warning information according to the recognition result.

[0030] According to the third aspect of the present disclosure, an electronic device is provided. The electronic device comprises at least one processor and a memory communicatively connected to the at least one processor. The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method as described above.

[0031] According to the fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, and the computer instructions are used to enable a computer to perform the method as described above.

[0032] According to the fifth aspect of the present disclosure, a computer program product is provided, which comprises a computer program, and the computer program, when executed by a processor, implements the method as described above.

[0033] It should be understood that the content described in the technical solution part is not intended to limit the key or important features of the embodiments of the present disclosure, nor to limit the scope of the present disclosure. Other features of the present disclosure will become apparent through the following description. Advantages

[0034] In the present disclosure, based on UWB positioning, a precise UWB positioning is obtained by combining a multipath suppression method and a dynamic Kalman filtering method, and then a machine learning algorithm is used for multi-modal sign information and behavior pattern recognition, so as to realize real-time monitoring of personnel and equipment, so as to timely discover potential dangers and give accurate early warnings, and effectively improve the safety protection level and safety management efficiency of the work site. BRIEF DESCRIPTION OF DRAWINGS

[0035] The above and other features, advantages, and aspects of embodiments of the present disclosure will become more apparent upon consideration of the following detailed description, taken in conjunction with the accompanying drawings. The following detailed description, given by way of example in conjunction with the accompanying drawings, serves to illustrate the present disclosure, and is not intended to limit the present disclosure in any manner. In the drawings:

[0036] FIG. 1 shows a flowchart of a high-precision positioning-based work site intelligent protection method according to an embodiment of the present disclosure;

[0037] FIG. 2 shows a block diagram of a high-precision positioning-based work site intelligent protection system according to an embodiment of the present disclosure;

[0038] FIG. 3 shows a block diagram of an exemplary electronic device capable of implementing an embodiment of the present disclosure. BEST MODE FOR CARRYING OUT THE INVENTION

[0039] To make the objectives, technical solutions, and advantages of the embodiments of the present disclosure clearer, the technical solutions in the embodiments of the present disclosure will be described clearly and completely below with reference to the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are some, but not all, of the embodiments of the present disclosure. Based on the embodiments in the present disclosure, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present disclosure.

[0040] In addition, the term "and / or" herein merely describes an association relationship of associated objects, and indicates that there can be three relationships, for example, A and / or B can represent the three cases of A alone, A and B together, and B alone. In addition, the character " / " herein generally represents an "or" relationship between the front and rear associated objects.

[0041] To solve the problems mentioned in the background, the present disclosure provides a high-precision positioning-based work site intelligent protection method, system, device, and storage medium.

[0042] Specifically, the intelligent protection method is applied to an intelligent protection system, the intelligent protection system is provided with an intelligent electronic fence, and a region surrounded by the intelligent electronic fence is defined as a work site. The intelligent protection method comprises the following steps: based on UWB positioning, precise positioning of personnel or equipment inside and outside the work site is obtained by using a multipath suppression method and a dynamic Kalman filtering method; a machine learning algorithm is used to identify multi-modal sign information and behavior patterns of the personnel or equipment that have been precisely positioned; and corresponding protection warning information is sent according to the identification result.

[0043] In this way, potential dangers can be discovered in time and precise warnings can be sent, effectively improving the safety protection level and safety management efficiency of the work site.

[0044] The intelligent protection method, system, device and storage medium based on high-precision positioning provided by the present disclosure will be described in detail below in combination with the accompanying drawings and specific embodiments.

[0045] FIG. 1 shows a flowchart of an intelligent protection method for a work site based on high-precision positioning according to an embodiment of the present disclosure. As shown in FIG. 1, the intelligent protection method for a work site based on high-precision positioning 100 can comprise the following steps:

[0046] S110, based on UWB positioning, precise positioning of personnel or equipment inside and outside the work site is obtained by using a multipath suppression method and a dynamic Kalman filtering method.

[0047] Specifically, based on UWB positioning, the multipath interference of the received signal is suppressed by using a multipath suppression method; by using the dynamic Kalman filtering method, the UWB signal after multipath interference suppression is used as a measurement value, which is fused with a priori estimate value, to improve the positioning accuracy of UWB, so as to obtain the precise positioning of personnel or equipment inside and outside the work site.

[0048] It should be noted that in wireless communication, a signal may pass through multiple different paths from the transmitting end to the receiving end, which is the multipath effect. The multipath effect can cause signal fading, time delay spread and increased interference, thereby affecting the positioning accuracy. The purpose of the multipath suppression method is to reduce or eliminate the influence of the multipath effect and improve the quality of the received signal and the positioning accuracy.

[0049] There are various multipath suppression methods, and the more common ones are time reversal technology and a multipath suppression algorithm based on minimum mean square error (MMSE). The time reversal technology is to perform time reversal processing on the received signal and then transmit it back, so that the multipath signals are aligned in time and mutually enhanced, while the noise and interference signals are weakened. The multipath suppression algorithm based on minimum mean square error (MMSE) is to analyze the statistical characteristics of the received signal, estimate the parameters of the multipath channel, and then filter the received signal to remove the multipath interference.

[0050] Taking a multipath interference suppression algorithm based on minimum mean square error (MMSE) as an example, the received signal can be expressed as:

[0051] ;

[0052] where y(t) is the received signal, s(t) is the transmitted signal, h i (t) is the channel response of the i th path, is the time delay of the i th path, and n(t) is noise.

[0053] By analyzing the statistical characteristics of the received signal, the parameters of the multipath channel are estimated; it is assumed that, after a period of measurement and analysis, the estimated value of the channel impulse response h(t) is:

[0054] ;

[0055] where is the estimated amplitude of the j th path, is the estimated time delay of the j th path, and M is the total number of estimated paths.

[0056] An MMSE filter is designed:

[0057] ;

[0058] where is the frequency response of the channel, is the conjugate of , N0 is the noise power spectral density, and E s is the energy of the signal.

[0059] The received signal is filtered by the filter to remove multipath interference and obtain a relatively pure signal.

[0060] According to an embodiment of the present disclosure, the UWB signal after multipath interference suppression is taken as a measurement value to provide the position information of the target at the current time; the prior estimate value contains the prediction of the current state of the target, and considers the motion trend and dynamic characteristics of the target; by fusing the measurement value and the prior estimate value, the advantages of both can be fully utilized to improve the positioning accuracy.

[0061] Further, by using the signal after multipath interference suppression as a measurement value and fusing it with the prior estimate value by means of dynamic Kalman filtering, the positioning accuracy of the UWB is improved to obtain the accurate positioning of personnel or equipment inside and outside the work site, which can include:

[0062] A state space model describing the motion state of personnel or equipment inside and outside the work site is established; the motion state historical data of personnel or equipment inside and outside the work site are acquired and analyzed to determine a state transition matrix and an observation matrix; a UWB signal after multipath interference suppression is used as a measurement value, and based on the state estimation at a historical time and the state transition matrix, a priori estimation value is calculated; the measurement value and the priori estimation value are fused through Kalman gain to improve the positioning accuracy of UWB, so as to obtain accurate positioning of personnel or equipment inside and outside the work site.

[0063] Specifically, the target motion model is established as a linear motion model, the state variable includes position x and velocity v, and the state equation is:

[0064] ;

[0065] wherein T is a sampling time interval, ak is acceleration, and wk is process noise.

[0066] The measurement equation is:

[0067] ;

[0068] wherein z k is a measurement value (i.e. a position measured by the UWB signal after multipath interference suppression), H =

[0010] , v k is a measurement noise.

[0069] Suppose the process noise covariance ;

[0070] The measurement noise covariance ;wherein, (supposed process noise variance); (supposed measurement noise variance).

[0071] According to the state estimation at the last time and the state transition matrix, the a priori estimation value is calculated.

[0072] The a priori state estimation ;the a priori error covariance .

[0073] When the measurement value is greatly affected by noise and interference, the a priori estimation value can provide certain stability to avoid large fluctuations in the positioning result, and at the same time, with the continuous updating of the measurement value, the a priori estimation value will also be continuously adjusted, so that the positioning result is more accurate and stable.

[0074] Then, the updating step is performed to calculate the Kalman gain:

[0075] ;

[0076] Update state estimation:

[0077]

[0078] Update error covariance:

[0079]

[0080] In this way, the positioning accuracy of the UWB is improved, and the accurate positioning of personnel or equipment inside and outside the work site is obtained. It can be understood that the environment of the work site may change, such as the movement of personnel and equipment, the appearance of obstacles, etc. The dynamic Kalman filtering method can adjust the positioning result in real time according to the change of the environment, and adapt to the needs of the dynamic environment. By fusing the measured value and the prior estimate value, the adaptability and reliability of the positioning system can be better adapted to environmental changes and improved.

[0081] According to the embodiments of the present disclosure, the received signal is subjected to interference suppression by using the multipath suppression method, which can reduce the positioning error caused by the multipath effect and improve the accuracy and stability of the UWB positioning. By fusing the signal subjected to multipath interference suppression with the prior estimate value through the dynamic Kalman filtering method, the positioning accuracy is further improved, providing a reliable basis for subsequent identification of physical information and judgment of behavior patterns.

[0082] S120, using a machine learning algorithm to identify the multi-modal physical information and behavior patterns of the personnel or equipment that have been accurately positioned; according to the identification result, corresponding protection warning information is issued.

[0083] Specifically, a recognition model is trained using a machine learning algorithm; the multi-modal physical information and behavior pattern information of the personnel or equipment that have been accurately positioned are input into the recognition model for identification.

[0084] The recognition model can be obtained by training as follows:

[0085] Obtain a large amount of multi-modal physical information of personnel or equipment in different states and corresponding behavior pattern annotation data;

[0086] Using a machine learning algorithm, the multi-modal physical information of personnel or equipment in different states is used as a sample, and the corresponding behavior pattern annotation data is used as a sample label to generate a training set, and the model is trained until a recognition model is generated.

[0087] According to the embodiments of the present disclosure, the model is trained by obtaining a large amount of multi-modal physical information and behavior pattern annotation data in different states, which can make the recognition model more accurately adapt to various actual situations, and the generated recognition model can quickly and accurately judge the state of personnel and equipment, improving the reliability and practicality of the intelligent protection system. ​​

[0088] To facilitate explanation, an example is provided below:

[0089] Suppose there are 50 workers in the work site, each wearing a smart bracelet that can collect heart rate, respiratory rate, and body temperature; at the same time, there are 10 devices in the work site, each equipped with a temperature sensor and a vibration sensor.

[0090] Collect multi-modal sign information of 50 workers and 10 devices, assume a total of 1000 data samples, of which 500 are normal working state, 200 are fatigue state, 100 are sick state, 150 are normal operation state, and 50 are fault state.

[0091] Use convolutional neural network (CNN) in deep learning algorithm to extract features and classify image type sign information (such as body temperature thermal imaging collected by smart bracelet), and use long short-term memory network (LSTM) for time series data such as heart rate and respiratory rate.

[0092] Take multi-modal sign information of personnel or equipment in different states as samples, and corresponding behavior mode label data as sample labels to generate training set, train model, and assume that after 100 training cycles, the accuracy of the model reaches 90%, meeting the requirements, and generating an identification model.

[0093] Use the identification model to identify the multi-modal sign information and behavior mode of the personnel or equipment that has been accurately located, and determine whether the state and behavior mode of the personnel or equipment in the work site are normal according to the identification results. For example, if a worker's heart rate is detected to be abnormally fast (assuming heart rate exceeds 120 beats per minute) and is close to a sensitive area (assuming distance from the boundary of the intelligent electronic fence is less than 3 meters), the system will immediately issue an audible and visual alarm and send a short message to the manager, which includes: personnel information, location, and the fact that the worker's heart rate exceeds 120 beats per minute and is less than 3 meters from the boundary of the intelligent electronic fence; if a device's temperature is too high (assuming it exceeds 80 degrees Celsius) and its vibration amplitude is abnormally large (assuming it exceeds 5 millimeters), the system will automatically send a warning message to the maintenance personnel, which includes: device information, location, and the fact that the device's temperature exceeds 80 degrees Celsius and its vibration amplitude exceeds 5 millimeters, so that the maintenance personnel can check and maintain.

[0094] Further, the model training code can be as follows:

[0095] # Generate data samples

[0096] def generate_data_samples(num_samples):

[0097] data_samples = []

[0098] statuses = ["normal", "fatigue", "sick", "normal", "fault"]

[0099] for _ in range(num_samples):

[0100] person_status = random.choice(statuses[:3])

[0101] device_status = random.choice(statuses[3:])

[0102] heart_rate, breathing_rate, body_temperature = generate_wristband_data(person_status)

[0103] device_temperature, device_vibration = generate_device_data(device_status)

[0104] data_samples.append({

[0105] "person": {

[0106] "heart_rate": heart_rate,

[0107] "breathing_rate": breathing_rate,

[0108] "body_temperature": body_temperature,

[0109] "status": person_status

[0110] },

[0111] "device": {

[0112] "temperature": device_temperature,

[0113] "vibration": device_vibration,

[0114] "status": device_status

[0115] }

[0116] })

[0117] return data_samples

[0118] # Data preprocessing

[0119] def preprocess_data(data_samples):

[0120] person_features = []

[0121] person_labels = []

[0122] device_features = []

[0123] device_labels = []

[0124] for sample in data_samples:

[0125] person_data = [sample["person"]["heart_rate"], sample["person"]["breathing_rate"], sample["person"]["body_temperature"]]

[0126] person_features.append(person_data)

[0127] person_labels.append(sample["person"]["status"])

[0128] device_data = [sample["device"]["temperature"], sample["device"]["vibration"]]

[0129] device_features.append(device_data)

[0130] device_labels.append(sample["device"]["status"])

[0131] return np.array(person_features), np.array(person_labels), np.array(device_features), np.array(device_labels)

[0132] # Deep learning model

[0133] class SimpleDeepLearningModel:

[0134] def __init__(self, input_dim, output_dim):

[0135] self.weights = np.random.randn(input_dim, output_dim)

[0136] self.bias = np.zeros(output_dim)

[0137] def forward(self, x):

[0138] return np.dot(x, self.weights) + self.bias

[0139] def train(self, x, y, learning_rate):

[0140] predictions = self.forward(x)

[0141] error = predictions - y

[0142] gradient = np.dot(xT, error) / len(x)

[0143] self.weights -= learning_rate * gradient

[0144] self.bias -= learning_rate * np.mean(error, axis=0)

[0145] # Model Training

[0146] def train_model(person_features, person_labels, device_features, device_labels):

[0147] person_model = SimpleDeepLearningModel(3, 3) # 3 input features, 3 person states

[0148] device_model = SimpleDeepLearningModel(2, 2) # 2 input features, 2 device states

[0149] learning_rate = 0.01

[0150] for epoch in range(100):

[0151] person_model.train(person_features, np.eye(3)[np.array([["normal", "fatigue", "sick"].index(label) for label in person_labels])], learning_rate)

[0152] device_model.train(device_features, np.eye(2)[np.array([["normal", "fault"].index(label) for label in device_labels])], learning_rate)

[0153] person_predictions = [["normal", "fatigue", "sick"][np.argmax(person_model.forward(np.array([f])))] for f in person_features]

[0154] device_predictions = [["normal", "fault"][np.argmax(device_model.forward(np.array([f])))] for f in device_features]

[0155] person_accuracy = sum([1 if p == l else 0 for p, l in zip(person_predictions, person_labels)]) / len(person_labels)

[0156] device_accuracy = sum([1 if p == l else 0 for p, l in zip(device_predictions, device_labels)]) / len(device_labels)

[0157] print(f"Epoch {epoch + 1}, Person Accuracy: {person_accuracy}, Device Accuracy: {device_accuracy}")

[0158] return person_model, device_model

[0159] # Model prediction

[0160] def predict_status(person_data,device_data, person_model, device_model):

[0161] person_status=["normal","fatigue", "sick"][np.argmax(person_model.forward(np.array(person_data)))]

[0162] device_status=["normal", "fault"][np.argmax(device_model.forward(np.array(device_data)))]

[0163] return person_status, device_status

[0164] # Send alerts and notifications

[0165] def send_alerts(person_status,device_status,person_distance_to_fence, device_temperature, device_vibration):

[0166] if person_status!= "normal" and person_distance_to_fence < 3:

[0167] print("Issues an audible and visual alarm and sends an SMS notification to administrators")

[0168] if device_status!= "normal" and (device_temperature > 80 or device_vibration > 5):

[0169] The system automatically sends a warning SMS to maintenance personnel, notifying them to perform inspection and maintenance.

[0170] # Main Program

[0171] if __name__ == "__main__":

[0172] # Generate data samples

[0173] data_samples = generate_data_samples(1000)

[0174] # Data preprocessing

[0175] person_features, person_labels, device_features, device_labels = preprocess_data(data_samples)

[0176] # Train model

[0177] person_model, device_model = train_model(person_features, person_labels, device_features, device_labels)

[0178] print("Model training is completed").

[0179] It can be understood that the model training process can also be based on other neural networks suitable for processing time series data such as recurrent neural network (RNN) and gated recurrent unit (GRU); at the same time, it can also be combined with the method of ensemble learning such as random forest and gradient boosting tree to improve the accuracy and generalization ability of the model; in addition, data augmentation techniques such as random cropping, rotation, and flipping can be used to expand the training data set to avoid overfitting; in the model evaluation stage, accuracy, recall, F1 value and other indicators can be used to comprehensively evaluate the model, and the parameters and structure of the model can be adjusted according to the actual demand to achieve the best performance, so that the algorithm scheme of the recognition model provided in the embodiments of the present disclosure is more diverse, and the present disclosure does not make specific limitations here.

[0180] According to the embodiments of the present disclosure, the recognition model is trained by a machine learning algorithm to automatically learn and extract features of multi-modal sign information and behavior patterns, improve the accuracy and efficiency of recognition, and input the personnel or device information accurately positioned into the recognition model for recognition, so that abnormal situations can be discovered in time and accurately.

[0181] S130, according to the recognition result, corresponding protection warning information is issued.

[0182] Specifically, according to the recognition result, it is judged whether the personnel or equipment state and behavior mode in the work site is normal, and if there is an abnormality, corresponding protection warning information is sent out. More specifically:

[0183] The personnel or equipment state and behavior mode abnormality in the work site includes but is not limited to: personnel vital signs, state, behavior abnormalities; personnel are too close to the electronic fence; abnormal personnel stay in the key work area for too long; equipment temperature, amplitude, running state abnormalities; abnormal equipment in the key work area, etc.

[0184] For personnel or equipment in the work site, personnel or equipment state is mainly monitored and identified for better scheduling management.

[0185] The personnel or equipment state and behavior mode abnormality outside the work site includes but is not limited to: personnel are too close to the electronic fence, personnel behavior abnormalities, etc.; equipment are too close to the electronic fence, equipment running state abnormalities, etc.

[0186] For personnel or equipment outside the work site, it is mainly monitored whether there are illegal behaviors such as approaching and destroying the electronic fence, approaching and taking pictures in violation of regulations, etc.

[0187] According to the embodiments of the present disclosure, according to the recognition result, it is judged whether the personnel and equipment state and behavior mode is normal, and warning information is sent out in abnormal cases, which can timely remind relevant personnel to take measures to avoid accidents, can reduce the safety risk of the work site, and protect the safety of personnel and equipment.

[0188] According to the embodiments of the present disclosure, the following technical effects are achieved:

[0189] By using the multipath suppression method, the positioning accuracy and stability of UWB positioning are effectively improved by reducing the positioning error caused by multipath effect. On this basis, combined with the dynamic Kalman filtering method, the signal after multipath interference suppression is fused with the prior estimate value, and the positioning accuracy is further improved. High-precision positioning information becomes a reliable basis for subsequent vital sign information recognition and behavior mode judgment, ensuring the accuracy and reliability of subsequent analysis. Further, by training the identification model through the machine learning algorithm, the identification model and the accurate positioning information are closely coordinated to effectively improve the accuracy and efficiency of model identification, so that the system can timely and accurately find abnormal situations and effectively avoid accidents, thereby reducing the safety risk of the work site. From the combination of multipath suppression method and dynamic Kalman filtering method to improve positioning accuracy, to the accurate identification of the identification model, to the abnormal warning mechanism, a series of links are interrelated and synergistic, forming a complete and efficient safety protection system, and comprehensively improving the safety and management efficiency of the work site.

[0190] It should be noted that, for the foregoing method embodiments, for the sake of simple description, they are all described as a combination of a series of actions, but those skilled in the art should know that the present disclosure is not limited by the order of the described actions, because according to the present disclosure, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should know that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily necessary for the present disclosure.

[0191] The above is the introduction of the method embodiment, and the scheme of the present disclosure will be further described through the system embodiment.

[0192] FIG. 2 shows a block diagram of a high-precision positioning-based work site intelligent protection system according to an embodiment of the present disclosure; as shown in FIG. 2, the high-precision positioning-based work site intelligent protection system 200 can include:

[0193] The first processing module 210 is configured to obtain accurate positioning of personnel or equipment inside and outside the work site based on UWB positioning using a multipath suppression method and a dynamic Kalman filtering method.

[0194] The second processing module 220 is configured to use a machine learning algorithm to identify multi-modal sign information and behavior patterns of personnel or equipment that have been accurately positioned.

[0195] The third processing module 230 is configured to issue corresponding protection warning information according to the identification result.

[0196] It can be understood that each module / unit in the high-precision positioning-based work site intelligent protection system 200 shown in FIG. 2 has the function of implementing each step in the high-precision positioning-based work site intelligent protection method 100 provided by the embodiments of the present disclosure, and can achieve its corresponding technical effects. The specific working process of the described module can be referred to the corresponding process in the foregoing method embodiments. For the convenience and brevity of description, it will not be described here.

[0197] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0198] FIG. 3 shows a block diagram of an exemplary electronic device capable of implementing an embodiment of the present disclosure.

[0199] As shown in FIG. 3, electronic device 300 is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. Electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not intended to limit the implementations of the present disclosure described and / or claimed in this document.

[0200] Electronic device 300 includes a computing unit 301 that can perform various appropriate actions and processes according to a computer program stored in ROM 302 or a computer program loaded into RAM 303 from storage unit 308. Various programs and data required for operation of electronic device 300 can also be stored in RAM 303. Computing unit 301, ROM 302, and RAM 303 are connected to each other by bus 304. I / O interface 305 is also connected to bus 304.

[0201] Various components in electronic device 300 are connected to I / O interface 305, including an input unit 306, such as a keyboard, a mouse, etc., an output unit 307, such as various types of displays, speakers, etc., a storage unit 308, such as a magnetic disk, an optical disk, etc., and a communication unit 309, such as a network card, a modem, a wireless communication transceiver, etc. Communication unit 309 allows electronic device 300 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.

[0202] Computing unit 301 can be various general and / or special purpose processing components with processing and computing capabilities. Some examples of computing unit 301 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. Computing unit 301 performs various methods and processes described above, such as method 100. For example, in some embodiments, method 100 can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as storage unit 308. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 300 via ROM 302 and / or communication unit 309. When the computer program is loaded into RAM 303 and executed by computing unit 301, one or more steps of method 100 described above can be performed. Alternatively, in other embodiments, computing unit 301 can be configured to perform method 100 by any other appropriate means, such as by means of firmware.

[0203] Various implementations of the systems and techniques described above can be realized in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a programmable logic device (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0204] Program code for carrying out methods of the present disclosure can be written in any combination of one or more programming languages. The program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the program code, when executed by the processor or controller, produces the functions / operations specified in the flowcharts and / or the block diagrams. The program code can be executed entirely on a machine, partially on a machine, partially on a machine as a stand-alone software package, partially on a machine and partially on a remote machine or entirely on a remote machine or server.

[0205] In the context of the present disclosure, a machine-readable medium can be a tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include but is not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium will include one or more lines of electrical connections, portable computer disks, hard disk drives, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or Flash memory), optical fibers, portable compact disc read-only memories (CD-ROMs), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0206] To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0207] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0208] The computer system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server is generally established by computer programs running on the respective computers and having a client-server relationship to each other. The servers can be cloud servers, servers of a distributed system, or servers combined with a blockchain.

[0209] It should be understood that the various forms of flow shown above can be re-ordered, added to, or deleted from without departing from the scope of the present disclosure. For example, the steps recited in the present disclosure can be performed in parallel, in series, or in a different order, without departing from the desired results of the technology disclosed in the present disclosure, and are not limited herein.

[0210] The specific embodiments described above are not intended to be limiting, and persons skilled in the art will appreciate that various modifications, combinations, sub-combinations and alternatives can be made to the specific embodiments without departing from the spirit and principles of the disclosure. Any further modifications, equivalents and / or alternatives thereof are also included within the scope of the present disclosure. Industrial Applicability

[0211] It can be understood that the work site intelligent protection method 100 based on high-precision positioning is applied to an intelligent protection system, the intelligent protection system is provided with an intelligent electronic fence, an area surrounded by the intelligent electronic fence is defined as a work site; a plurality of sensors and warning devices can be provided on the intelligent electronic fence, and a monitoring camera and a communication device can be further provided; a UWB base station, a monitoring camera, a communication device, a body information collector and the like can be provided in the work site.

[0212] The traditional protection system has poor linkage with other security systems, and cannot quickly work with other security subsystems such as video monitoring and access control systems when an abnormal situation occurs, making it difficult to achieve comprehensive and efficient security protection. The intelligent protection system provided in the embodiments of the present disclosure can realize efficient linkage and cooperative work with other security subsystems by applying the work site intelligent protection method 100 based on high-precision positioning, and the specific process can be as follows:

[0213] When the sensor on the intelligent electronic fence detects an abnormal situation, the sensor sends a signal to the intelligent protection system, the system obtains the accurate positioning of personnel or equipment inside and outside the work site based on UWB positioning, uses the multipath suppression method and the dynamic Kalman filtering method, determines the position information of the personnel or equipment at the abnormal position, and at the same time, sends an instruction to the monitoring camera inside the work site, the monitoring camera adjusts the angle and focal length according to the UWB accurate positioning, aims at the abnormal area, and uses a machine learning algorithm to identify the multi-modal body information and behavior pattern of the personnel or equipment that have been accurately positioned. If it is identified that the personnel have malicious behavior, the intelligent protection system sends corresponding protection warning information in time through the communication device according to the identification result, and notifies the management personnel to take corresponding measures.

[0214] For example, when it is detected that a certain person behaves abnormally near a specific dangerous area, the monitoring camera quickly focuses on the position, and at the same time, judges whether the person has malicious behavior according to the multi-modal body information identification result, if so, immediately sends an alarm and notifies the management personnel to handle it.

[0215] When the intelligent protection system identifies an emergency situation in the work site, such as a person being injured or a device malfunctioning, it can send an instruction to the access control system. If the person is injured and located near an exit, the UWB positioning system in the intelligent protection system continuously tracks the position of the person, and the access control system linked with the intelligent protection system automatically opens the nearest exit for rescue personnel to quickly enter, at the same time, closes or restricts the exits of some areas that may cause the spread of danger to prevent unrelated personnel from entering the dangerous area.

[0216] In addition, the intelligent protection system can also be linked with the fire-fighting equipment in the fire-fighting system. When a fire occurs, the positioning system in the intelligent protection system continuously tracks the position changes of the personnel, ensures that the fire-fighting personnel can accurately find the trapped personnel, and simultaneously judges the direction of fire spread according to the moving track of the personnel, to provide support for fire-fighting decision-making, such as automatically opening the fire-fighting spraying device, opening the fireproof door, etc., to improve the efficiency and safety of fire rescue.

Claims

1. A high-precision positioning-based intelligent protection method for a work site, the intelligent protection method being applied to an intelligent protection system, wherein an intelligent electronic fence is arranged in the intelligent protection system, and a region enclosed by the intelligent electronic fence is defined as a work site, and the intelligent protection method is characterized in that, The intelligent protection method comprises: Based on UWB positioning, using multipath suppression method and dynamic Kalman filtering method, the accurate positioning of personnel or equipment inside and outside the work site is obtained; Using machine learning algorithm, the multi-modal sign information and behavior mode of the personnel or equipment with accurate positioning are identified; According to the identification result, the corresponding protection warning information is sent out.

2. The method of claim 1, wherein, The UWB positioning based on the multipath suppression method and the dynamic Kalman filtering method is used to obtain the accurate positioning of personnel or equipment inside and outside the work site, comprising: Based on UWB positioning, using multipath suppression method to suppress multipath interference of received signal; Through dynamic Kalman filtering method, the UWB signal after multipath interference suppression is used as measurement value, and the prior estimate value is fused to improve the positioning accuracy of UWB, so as to obtain the accurate positioning of personnel or equipment inside and outside the work site.

3. The method of claim 2, wherein, The UWB signal after multipath interference suppression is used as measurement value, and the prior estimate value is fused to improve the positioning accuracy of UWB, so as to obtain the accurate positioning of personnel or equipment inside and outside the work site, comprising: Establishing a state space model describing the motion state of personnel or equipment inside and outside the work site; Obtaining the motion state historical data of personnel or equipment inside and outside the work site and analyzing to determine the state transition matrix and observation matrix; Using the UWB signal after multipath interference suppression as measurement value, and based on the state estimation at historical time and the state transition matrix, the prior estimate value is calculated; Through Kalman gain, the measurement value and the prior estimate value are fused to improve the positioning accuracy of UWB, so as to obtain the accurate positioning of personnel or equipment inside and outside the work site.

4. The method of claim 1, wherein, The multi-modal sign information and behavior mode of the personnel or equipment with accurate positioning are identified by using machine learning algorithm, comprising: Using machine learning algorithm to train identification model; The multi-modal sign information and behavior mode information of the personnel or equipment with accurate positioning are input into the identification model for identification.

5. The method of claim 4, wherein, The identification model is obtained by training as follows: Obtaining a large amount of multi-modal sign information of personnel or equipment in different states and corresponding behavior mode annotation data; Using machine learning algorithm, the multi-modal sign information of personnel or equipment in different states is used as sample, and the corresponding behavior mode annotation data is used as sample label to generate training set for model training until the identification model is generated.

6. The method of claim 1, wherein, According to the identification result, the corresponding protection warning information is sent out, comprising: According to the identification result, it is judged whether the state and behavior mode of personnel or equipment inside and outside the work site are normal, and if there is an abnormality, the corresponding protection warning information is sent out.

7. A high-precision positioning-based intelligent protection system for a work site, applying the intelligent protection method according to any one of claims 1-6, characterized in that, The system comprises: The first processing module is used for obtaining the accurate positioning of personnel or equipment inside and outside the work site based on UWB positioning, using multipath suppression method and dynamic Kalman filtering method; The second processing module is used for identifying the multi-modal sign information and behavior mode of the personnel or equipment with accurate positioning by using machine learning algorithm; The third processing module is used for sending out the corresponding protection warning information according to the identification result.

8. An electronic device, comprising: The device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-6.

9. A non-transitory computer readable storage medium having stored thereon computer instructions, characterized in that, the computer instructions are for causing the computer to perform the method of any one of claims 1-6.

10. A computer program product comprising a computer program which, when executed by a processor, implements the method of any one of claims 1-6.

Images