Smart park management system

Abstract

The invention discloses a smart park management system, and relates to the field of data processing. The system comprises an inspection robot, a management server, a park management terminal and a settled enterprise terminal. The inspection robot inspects in the park according to a set inspection path and sends information including inspection images, places and time to the management server; the management server determines an abnormal event of the inspection site according to the inspection image, and sends abnormal event prompt information to the park management terminal and the settled enterprise terminal; when the satellite positioning signal intensity is smaller than a threshold value, the inspection robot estimates the current position according to self motion and sensor observation data to obtain a first estimated position, and the motion state is adjusted when the inspection robot deviates from an inspection path. The park is inspected through the inspection robot, the problems that the park inspection efficiency is low and the dependence on manpower is large are solved, the management server is used for sending the abnormal prompt information to the park management terminal and the settled enterprise terminal, and rapid transmission and accurate processing of the information are achieved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a smart park management system. Background Art

[0002] Smart parks use new-generation information technologies such as the Internet of Things, big data, and artificial intelligence to fully perceive and integrate park infrastructure, resources, personnel, and enterprises. They deploy sensors and smart devices to monitor infrastructure, integrate resources, and reduce operating costs; manage people intelligently to provide convenient services; build service platforms to help companies develop; and achieve departmental collaboration. It aims to create an efficient, convenient, green, and safe park environment, promote industrial upgrading and sustainable development, and enhance the competitiveness and attractiveness of the park.

[0003] In order to ensure the normal operation of the smart park, it is necessary to conduct regular inspections and management of the park. The traditional management method of smart parks mainly relies on a combination of manual inspections and regular equipment maintenance. Staff members follow fixed inspection routes and time intervals, hold paper inspection forms, and check the infrastructure in the park, such as buildings, roads, water and electricity facilities, one by one, and record the operating status of the equipment, environmental parameters, etc. For key equipment, regular shutdown maintenance and inspection are also required. At the same time, information communication and feedback are carried out through telephone or intercom. Once a problem is found, it is reported in a timely manner and arranged for maintenance personnel to handle it.

[0004] However, in these traditional park management methods, there are problems such as high reliance on manpower and low inspection efficiency. Summary of the invention

[0005] In view of the above-mentioned technical problems and defects, the purpose of the present invention is to provide a smart park management system that can alleviate the problems of low park inspection efficiency and high reliance on manpower.

[0006] To achieve the above-mentioned objectives, the present invention provides a smart park management system, comprising a patrol robot, a management server, a park management terminal and a resident enterprise terminal; the patrol robot is used to conduct patrols in the smart park according to a set patrol path, and send the collected patrol information to the management server, the patrol information including patrol images, patrol locations and patrol times; the management server is used to determine, based on the patrol images, that there is an abnormal event at the patrol location, and send abnormal event prompt information to the park management terminal and the resident enterprise terminal; the patrol robot is also used to estimate the position at the current moment to obtain a first estimated position based on its own motion model data and sensor observation data when the satellite positioning signal strength is less than a set signal strength threshold, and adjust its own motion state when it is determined that the patrol robot deviates from the patrol path based on the first estimated position; the motion model data includes the position information, moving speed and moving direction of the patrol robot at the previous moment, and the sensor observation data includes the park terrain feature information collected by the sensor.

[0007] The present invention improves inspection efficiency and reduces dependence on manpower by introducing inspection robots to replace traditional manual inspections. The inspection robot can automatically perform inspections according to the set inspection path, without the need for manual hand-held paper inspection forms to be checked one by one, and can work continuously without being limited by time and physical strength. In addition, by sending the collected inspection information to the management server, rapid transmission and accurate processing of information are achieved. The management server determines abnormal events based on the inspection images, and promptly sends abnormal event prompt information to the park management terminal and the resident enterprise terminal, greatly improving the timeliness of problem handling. At the same time, when the satellite positioning signal strength is less than the set threshold, the robot estimates the position using its own motion model data and sensor observation data to ensure accurate inspections even in complex environments, alleviating the problem of inaccurate positioning due to signal problems, causing inspection interruptions or inaccuracies.

[0008] In some embodiments, the inspection robot is specifically used to: obtain a group of particles with the same weight, which represent the possible position and posture of the inspection robot; propagate the particles to a new position according to the motion model data to obtain a priori estimated values ​​of the particles to predict the current possible position of the inspection robot; based on the sensor measurement data, calculate the likelihood of the state represented by each particle through a preset measurement model to update the weight of the particle; perform a resampling operation according to the weight of the particle, discard the particles with lower weights, and copy the particles with higher weights to make the particle set more concentrated on the possible area of ​​the actual state of the inspection robot; perform weighted averaging on the resampled particles or select the particle with the largest weight to obtain an estimated result of the previous position, and determine the estimated result as the first estimated position.

[0009] The technical solution of the above embodiment is adopted to provide an effective position estimation method for the inspection robot when the satellite positioning signal is poor. By adopting the particle filtering technology, a group of particles with the same weights and motion model data and sensor observation data are used to first propagate the particles to the new position to obtain the prior estimate value to predict the current position, and then the likelihood is calculated based on the sensor measurement data to update the particle weight, perform resampling operations and determine the first estimated position. This method makes full use of the robot's motion history and sensor information, improves the accuracy and reliability of position estimation, ensures that the robot can better position itself in a complex environment, avoids interruption of inspection work or deviation from the inspection path due to weak satellite positioning signals, and ensures the continuity and stability of the inspection task.

[0010] In some embodiments, the inspection robot is also used to: determine a second estimated position of the inspection robot based on terrain feature information around the current position and stored campus map information; determine an algorithm estimated position of the inspection robot based on the first estimated position and the second estimated position; and determine whether the inspection robot has deviated from the inspection path based on the algorithm estimated position.

[0011] By adopting the technical solution of the above-mentioned embodiment, the inspection robot has a more powerful position estimation and path monitoring capability. It combines the terrain feature information around the current position and the stored park map information to determine the second estimated position, and then combines the first estimated position and the second estimated position to obtain the algorithm estimated position, so as to judge whether it deviates from the inspection path. This not only utilizes the local terrain information, but also combines the macro information of the overall park map, providing multi-dimensional data support for the robot's position estimation, making the robot more accurate in positioning and path tracking, and helping the inspection robot to complete the inspection task in the smart park more efficiently and accurately. At the same time, it also enhances the system's control over the robot's position and provides more accurate data basis for park management.

[0012] In some embodiments, the inspection robot is specifically used to determine the one with higher confidence between the first estimated position and the second estimated position as the algorithm estimated position when the difference between the first estimated position and the second estimated position is greater than a set threshold.

[0013] Using the technical solution of the above embodiment, when the difference between the first estimated position and the second estimated position is greater than the set threshold, the one with higher confidence is determined as the algorithm estimated position. The advantage of this approach is that it takes into account the reliability differences of different position estimation methods in different scenarios. By selecting the estimation result with higher confidence, it avoids the positioning deviation caused by the error of a single position estimation method, improves the credibility and accuracy of the algorithm estimated position, thereby ensuring that the position estimation result of the inspection robot is closer to the actual situation, ensuring the navigation accuracy of the robot during the inspection process, and further improving the stability and reliability of the inspection system in a complex campus environment, providing more reliable location information for subsequent inspections and abnormal event handling operations.

[0014] In some embodiments, the terrain feature information includes laser scanning information, the park map information includes three-dimensional model data of multiple inspection points in the smart park, and the inspection robot is specifically used to compare the laser scanning information with the three-dimensional model data to determine the target inspection point from the multiple inspection points, and determine the position of the target inspection point as the second estimated position.

[0015] By using the technical solution of the above embodiment, the laser scanning information in the terrain feature information is compared with the three-dimensional model data of multiple inspection points in the park map information, and the target inspection point is determined and its position is used as the second estimated position. This technical solution plays an important role in the smart park management system. The laser scanning information can accurately reflect the details of the current environment, while the three-dimensional model data provides global park information. The combination of the two enables the inspection robot to more accurately locate itself in a complex park environment. By utilizing different types of data, this method enables the robot to fully utilize the local details and global information of the environment during the positioning process, significantly improving the accuracy of the second estimated position, helping the robot to better perform inspection tasks, while improving the positioning accuracy and inspection efficiency of the entire inspection system.

[0016] In some embodiments, the system also includes an environmental sensor, which is used to detect environmental information of the smart park and send the environmental information to the management server. The management server is used to send abnormal event prompt information to the park management terminal when the environmental information exceeds a normal range.

[0017] By adopting the technical solution of the above embodiment, the addition of environmental sensors greatly enriches the functions of the smart park management system. It can detect various environmental information of the smart park and send it to the management server. When the environmental information exceeds the normal range, the management server can send abnormal event prompt information to the park management terminal. This not only enables park managers to be aware of environmental anomalies in a timely manner and take corresponding measures to ensure the environmental safety and stability of the park, but also provides more dimensional monitoring and protection for the operation of the park, which is conducive to real-time monitoring and maintenance of the environment, reducing the potential risks to the normal operation of the park due to environmental anomalies, and improving the safety and sustainable operation capabilities of the park.

[0018] In some embodiments, the park management terminal is used to send abnormal event processing completion information to the management server after the abnormal event is processed.

[0019] By adopting the technical solution of the above embodiment, the park management terminal sends abnormal event processing completion information to the management server after the abnormal event is processed, realizing the feedback mechanism of abnormal event processing. This helps the management server to grasp the processing progress and status of abnormal events in a timely manner, update the information of abnormal events in the system, and make the information flow of the entire system smoother and more complete. At the same time, the transmission of this information can provide a basis for subsequent management decisions, facilitate the management server to perform subsequent information processing and coordination, help optimize the entire abnormal event processing process, and improve the system's management efficiency and responsiveness to abnormal events.

[0020] In some embodiments, the management server is used to generate a processing result report of the abnormal event according to the abnormal event processing completion information, and send the processing result report to the resident enterprise terminal.

[0021] By adopting the technical solution of the above-mentioned embodiment, the management server generates a processing result report based on the abnormal event processing completion information and sends it to the resident enterprise terminal. This can strengthen the information communication between the management server and the resident enterprise terminal. The resident enterprise can understand the handling of abnormal events in the park through the processing result report, provide a reference for the operation and decision-making of the enterprise, and enable the enterprise to better grasp the management quality and operation status of the park. This helps to enhance the confidence of the resident enterprises in the park management and promote the cooperation between the enterprise and the park management. At the same time, it is also convenient for the resident enterprises to adjust their own operation plans according to the results of the abnormal event processing, promote the sharing and coordination of information among all parties in the smart park, and provide better guarantees for the sustainable and stable development of the enterprise.

[0022] In some embodiments, the resident enterprise terminal is used to send an enterprise evaluation of the processing result of the abnormal event to the management server.

[0023] By adopting the technical solution of the above embodiment, the settled enterprises can express their satisfaction and opinions on the handling of abnormal events, and provide valuable feedback information for the management server. The management server can find the deficiencies in its own management and services based on the evaluation of the enterprise, and then adjust and optimize the abnormal event handling mechanism to improve the service quality. Through this evaluation feedback, the continuous improvement of the system is achieved, the interaction between the settled enterprises and the park management is enhanced, which helps to build a better park service system, create a better business environment for enterprises, and improve the overall service level of the smart park and the satisfaction of the settled enterprises.

[0024] In some embodiments, the management server is also used to send an abnormal event processing inspection instruction to the inspection robot based on the abnormal event processing completion information. The inspection robot is used to move to the location where the abnormal event occurred according to the abnormal event processing inspection instruction, and send the captured processing result inspection image to the management server. The management server is used to determine the processing result management evaluation of the abnormal event based on the processing result inspection image.

[0025] Using the technical solution of the above embodiment, the management server sends an abnormal event processing inspection instruction to the inspection robot. The inspection robot moves to the location where the abnormal event occurs according to the instruction and takes an inspection image of the processing result and sends it to the management server. The management server determines the management evaluation of the abnormal event processing result based on the image. This process forms a closed-loop verification mechanism for the abnormal event processing results, which helps to ensure the quality of abnormal event processing and avoid incomplete processing or omissions. In this way, the management server can more accurately evaluate the processing effect of abnormal events, promptly discover and solve potential problems, ensure the refinement of park management, and provide strong support for the stable operation and security of the park.

[0026] One or more technical solutions provided by the present invention have at least the following technical effects or advantages: 1. The present invention realizes accurate estimation of the inspection position of the inspection robot and effective path monitoring through a variety of innovative technical means. On the one hand, when the satellite positioning signal strength is insufficient, the inspection robot uses the particle filter algorithm to estimate the position according to its own motion model data and sensor observation data, and finally obtains the first estimated position through particle propagation, likelihood calculation, weight update and resampling operations; on the other hand, the second estimated position is determined by combining the terrain feature information around the current position (such as laser scanning information) and the stored park map information (including the three-dimensional model data of multiple inspection points), and the final algorithm estimated position is determined according to the difference and confidence between the two. This comprehensive position estimation method fully considers the historical state of the robot's movement and the current environmental information, not only ensuring the accuracy of the position estimation when the satellite positioning signal is poor, but also utilizing the advantages of data from different sources to achieve multi-dimensional position evaluation and positioning, avoiding the robot from deviating from the inspection path, ensuring the high-precision execution of the inspection task, greatly improving the navigation and positioning capabilities of the inspection robot in a complex park environment, and providing reliable position information guarantee for the inspection work of the entire park.

[0027] 2. The system demonstrates high efficiency and systematicness in monitoring and handling abnormal events. During the inspection process, the inspection robot collects inspection information and sends it to the management server. The management server can identify abnormal events based on the inspection images, and send abnormal event prompt information to the park management terminal and the resident enterprise terminal, realizing the timely discovery of abnormal events and the rapid transmission of information. At the same time, when the abnormal event is handled, the park management terminal will feedback the processing completion information to the management server. The management server can generate a processing result report and send it to the resident enterprise terminal. The resident enterprise terminal can also evaluate the processing results, forming a complete abnormal event processing feedback loop. This process ensures the efficient operation of abnormal events from discovery to processing and subsequent evaluation, so that all parties can grasp the abnormal situation in a timely manner and take countermeasures, which helps to improve the emergency response capabilities of the park and ensure the normal operation and management of the park. At the same time, it also strengthens the information communication and collaboration between the park management and the resident enterprises, providing a good mechanism guarantee for improving the park management level and service quality.

[0028] 3. Environmental sensors are responsible for detecting various environmental information of the smart park. Once the environmental information exceeds the normal range, the management server will send abnormal event prompt information to the park management terminal. This realizes real-time monitoring of the park environment, allowing managers to promptly know the abnormal conditions of the environment, such as abnormal changes in indicators such as temperature, humidity, and air quality. Through this environmental monitoring and abnormal prompt mechanism, the park can respond quickly and initiate corresponding treatment measures to avoid environmental anomalies from causing harm to personnel and equipment in the park, ensuring the environmental safety and stable operation of the park. At the same time, it also provides a good environmental foundation for the long-term sustainable development of the park, reduces the operational risks that may be caused by environmental problems, and improves the safety and stability of the park. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The drawings herein are incorporated into and constitute a part of the specification, showing embodiments consistent with the present invention, and together with the specification, are used to explain the principles of the present invention. Obviously, the drawings described below are only some embodiments of the present invention, and for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work. In the drawings: Figure 1 It is a schematic diagram of the architecture of a smart park management system according to an embodiment of the present invention; Figure 2 It is a connection relationship diagram of a smart park management system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0030] The terms used in the following embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to be limiting of the present invention. As used in the specification of the present invention, the singular expressions "a", "a", "above", "the" and "this" are intended to also include plural expressions, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in the present invention refers to any or all possible combinations comprising one or more of the listed items.

[0031] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood as implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present invention, unless otherwise specified, the meaning of "plurality" is two or more.

[0032] It should also be noted that, unless otherwise clearly specified and limited, in the embodiments of the present invention, the terms such as "setting" and "connection" should be understood in a broad sense. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two components; it can be a wired communication connection or a wireless communication connection. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances. The embodiments of the present invention are described in detail below.

[0033] The embodiment of the present invention provides a smart park management system, which introduces inspection robots to replace traditional manual inspections, greatly improving inspection efficiency and reducing dependence on manpower. The inspection robot can automatically perform inspections according to the set inspection path, without the need for manual hand-held paper inspection forms to be checked one by one, and can work continuously without being restricted by time and physical strength. In addition, by sending the collected inspection information to the management server, rapid transmission and accurate processing of information are achieved. The management server determines abnormal events based on the inspection images, and promptly sends abnormal event prompt information to the park management terminal and the resident enterprise terminal, greatly improving the timeliness of problem handling. At the same time, when the satellite positioning signal strength is less than the set threshold, the robot estimates the position using its own motion model data and sensor observation data to ensure accurate inspections even in complex environments, alleviating the problem of inaccurate positioning due to signal problems, causing inspection interruptions or inaccuracies.

[0034] like Figure 1 and Figure 2 As shown, the smart park management system of this embodiment includes an inspection robot 1, a management server 2, a park management terminal 3 and a resident enterprise terminal 4. Among them, the park management terminal 3 and the resident enterprise terminal can include but are not limited to mobile phones, computers, tablets, smart wearable devices, smart cars and other devices.

[0035] The inspection robot 1 is used to conduct inspections in the smart park according to a set inspection path, and send the collected inspection information to the management server 2, where the inspection information includes inspection images, inspection locations and inspection times.

[0036] Specifically, the inspection robot 1 conducts patrol inspections in an orderly manner in the smart park according to a pre-set inspection route. During the inspection process, the robot obtains inspection images through the equipped image acquisition equipment. These images can intuitively reflect the actual conditions of various locations in the park, such as the appearance of the equipment, whether there are any abnormal phenomena, etc. At the same time, the robot uses the positioning system to determine the inspection location and accurately records the inspection time. Subsequently, the robot sends the inspection information including the inspection image, inspection location and inspection time to the management server 2. In this way, the management server 2 can comprehensively analyze and judge the status of the park based on these detailed information, promptly discover potential problems and take corresponding measures to ensure the normal operation of the smart park.

[0037] The management server 2 is used to determine whether an abnormal event exists at the inspection location based on the inspection image, and send abnormal event prompt information to the park management terminal 3 and the resident enterprise terminal 4.

[0038] Specifically, when the inspection robot 1 sends the inspection information such as the inspection image, inspection location and inspection time, the management server 2 will analyze and process the inspection image. The management server 2 uses image recognition technology and other means to compare the image features under normal conditions to determine whether there is an abnormal event at the current inspection location. Once an abnormality is detected, such as equipment damage, environmental changes, safety hazards, etc., the management server 2 will immediately send abnormal event prompt information to the park management terminal 3 and the resident enterprise terminal 4. In this way, the park managers and resident enterprises can understand the problems in the park in a timely manner so as to take countermeasures quickly to ensure the normal operation of the park and the production and operation activities of the enterprises.

[0039] In some embodiments, the management server 2 identifies the inspection image by calling the abnormal event detection model to determine whether an abnormal event exists at the inspection location. The abnormal event detection model is usually trained in the following way: First, collect a large amount of inspection image data. This data should cover images of the smart park in various states, including normal states and various abnormal states. Abnormal states can include equipment failure, environmental damage, safety hazards, etc. For each collected image, manual annotation is performed to clearly indicate whether there is an abnormal event in the image and the specific type of abnormality.

[0040] Then, the labeled image data is divided into training set, validation set and test set. The training set is used to train the abnormal event detection model. The model continuously adjusts its parameters to improve the recognition accuracy of abnormal events by learning the features of the image and the corresponding relationship with abnormal events. During the training process, the validation set is used to monitor the performance of the model and adjust the hyperparameters to optimize the model.

[0041] Finally, the trained model is evaluated using the test set to ensure that the model can accurately detect abnormal events on new, unseen data. If the performance of the model does not meet the requirements, the model structure can be further adjusted, the amount of data can be increased, or the training method can be improved, and retraining can be performed until the model reaches good performance indicators.

[0042] Furthermore, the inspection robot 1 is also used to estimate the position at the current moment to obtain a first estimated position based on its own motion model data and sensor observation data when the satellite positioning signal strength is less than a set signal strength threshold, and adjust its own motion state when it is determined that the inspection robot 1 deviates from the inspection path based on the first estimated position; the motion model data includes the position information, moving speed and moving direction of the inspection robot 1 at the previous moment, and the sensor observation data includes the terrain feature information of the park collected by the sensor.

[0043] In this embodiment, the inspection robot 1 has the ability of autonomous positioning and path adjustment. When the satellite positioning signal strength is less than the set threshold, in order to ensure the accurate determination of its own position, the robot will estimate the position based on its own motion model data and sensor observation data. The motion model data contains the position information, moving speed and moving direction of the previous moment, which enables the robot to make position inferences based on the previous motion state. At the same time, the sensor observation data includes the terrain feature information of the park collected by the sensor, such as identifying the terrain features such as buildings and roads in the park through sensors such as laser radar. By combining this information, the robot estimates the first estimated position at the current moment. When it is determined that the robot has deviated from the preset inspection path according to this estimated position, it will adjust its own motion state in time, such as adjusting the direction of travel, speed, etc., to ensure that it can return to the correct inspection path and continue to efficiently complete the inspection task. This embodiment effectively solves the problems of low efficiency and heavy reliance on manpower in the traditional smart park inspection method by introducing an intelligent inspection robot 1, combining the management server 2, the park management terminal 3 and the resident enterprise terminal 4. The inspection robot 1 can automatically inspect along the preset path, collect inspection images, locations, time and other information in real time through the sensors and cameras carried, and send them to the management server 2. The server analyzes these data through image recognition technology, quickly identifies abnormal events, and promptly sends prompt information to the park management terminal 3 and the resident enterprise terminal 4, thereby realizing the rapid transmission and processing of information. In addition, in response to the problem of weak satellite positioning signals, the inspection robot 1 can use its own motion model data and sensor observation data to estimate the position, ensure accurate positioning even in environments with poor positioning signals, and automatically adjust the motion state to stay on the preset inspection path. This intelligent inspection method not only improves inspection efficiency and reduces the consumption of human resources, but also improves the accuracy and timeliness of information processing through automation and intelligent means, enhances the park's safety management and emergency response capabilities, and provides strong technical support for the efficient, convenient, green and safe operation of smart parks.

[0044] In some embodiments, the inspection robot 1 is specifically used to: obtain a group of particles with the same weight, which represent the possible position and posture of the inspection robot 1; propagate the particles to a new position according to the motion model data to obtain a priori estimated values ​​of the particles to predict the current possible position of the inspection robot 1; based on the sensor measurement data, calculate the likelihood of the state represented by each particle through a preset measurement model to update the weight of the particle; perform a resampling operation according to the weight of the particle, discard the particles with lower weights, and copy the particles with higher weights to make the particle set more concentrated on the possible area of ​​the actual state of the inspection robot 1; perform weighted averaging on the resampled particles or select the particle with the largest weight to obtain an estimated result of the previous position, and determine the estimated result as the first estimated position.

[0045] Based on the above technical solution, the inspection robot 1 of this embodiment uses a particle filter algorithm to estimate the position when the satellite positioning signal strength is less than a threshold.

[0046] The particle filter algorithm has a strong nonlinear processing capability, which can effectively deal with the nonlinear problem of the inspection robot 1 moving in the complex environment of the park and more accurately describe its real motion state. Secondly, it does not need to make strict assumptions about the linearity and Gaussianity of the system, and can adapt to various complex actual situations, such as the impact of different terrains, different lighting, etc. on sensor observations, which improves the robustness of position estimation. Furthermore, by continuously updating the weights of the particles and performing resampling operations, the particle set can be more concentrated on the possible area of ​​the robot's real state, so that when the satellite positioning signal strength is insufficient, the robot's position can still be estimated with high accuracy, ensuring that the inspection robot 1 can maintain good positioning and path tracking performance even in the complex and changeable park environment, thereby improving the reliability and stability of the entire inspection system.

[0047] The following is a detailed process of implementing the particle filter algorithm to obtain the first estimated position in this embodiment: 1) Initialization: Generate particle set: At the beginning of the algorithm, the inspection robot 1 obtains a set of particles with the same weight. These particles are randomly distributed in the state space, and each particle represents the possible position and posture of the robot. Usually, the number of particles is set according to the actual situation. Generally speaking, the more particles there are, the higher the estimation accuracy is, but the amount of calculation will also increase accordingly.

[0048] 2) Prediction stage: Particle propagation: propagate particles to new positions according to the motion model data to obtain the prior estimate of the particles. Specifically, the position information, moving speed and moving direction of the inspection robot 1 at the previous moment are used to update each particle through the motion equation. For example, assuming that the robot moves on a two-dimensional plane, for the i-th particle (x i ,y i ,θ i ), its position update formula can be: x i =x i +v×Δt×cos(θ i ), y i =y i +v×Δt×sin(θ i ), where x is the horizontal coordinate, y is the vertical coordinate, v is the moving speed, Δt is the time interval, θ i is the direction angle. In this way, the prior estimated position of each particle at the current moment is obtained, thereby predicting the possible current position of the robot.

[0049] 3) Update phase: Calculate likelihood: Based on the sensor measurement data, the likelihood of the state represented by each particle is calculated through a preset measurement model, where the likelihood is used to indicate the probability that the particle is consistent with the actual observation data. Sensor observation data includes information about the park's terrain features collected by sensors, such as the distance to surrounding obstacles scanned by lidar, and landmark images obtained by visual sensors. These observation data are compared with the observation values ​​predicted based on the particle position, and the likelihood is calculated using a certain probability distribution function.

[0050] For example, if the actual distance of an obstacle observed by the LiDAR in front of the robot is d obs , and the distance of the obstacle predicted by the position of the i-th particle is d pred,i , then the likelihood can be calculated according to the Gaussian distribution, and the calculation formula is as follows: Among them, p(d obs |x i ,y i ,θ i ) is the likelihood, where σ is the standard deviation of the measurement noise.

[0051] Update weight: Update the weight of the particle according to the calculated likelihood.

[0052] Usually the formula w i =w i ×p(d obs |x i ,y i,θ i ), the particle weight w at the previous moment i Multiply it by the currently calculated likelihood to get the updated weight.

[0053] Furthermore, this embodiment takes into account that in a complex campus environment, the environment may be dynamic, for example, the movement of pedestrians and vehicles may affect sensor observation. It is possible to consider incorporating the dynamic information of the environment into the likelihood calculation. Assume that the dynamic information of the environment can be obtained through the sensor, such as the speed v of the object env , and predict the impact of this dynamic on the observation. The calculation formula of likelihood can be optimized as: Among them, v env Δt takes into account the impact of environmental dynamic factors on distance observation.

[0054] 3) Resampling stage: Evaluate particle validity: Perform resampling operations based on particle weights. First, evaluate the validity of each particle. Particles with lower weights indicate that they match the actual observations poorly and contribute less to the estimated results, while particles with higher weights are more likely to be close to the actual state of the robot.

[0055] Perform resampling: Through certain resampling algorithms, such as system resampling, stratified resampling, etc., particles with lower weights are discarded and particles with higher weights are copied, so that the particle set is more concentrated in the possible area of ​​the robot's real state. For example, in system resampling, a cumulative distribution function is generated according to the particle weight, and then multiple sampling points are randomly generated in the interval [0,1]. According to the interval where the sampling point is located, the corresponding particles are selected for copying or retention, thereby adjusting the distribution of the particle set.

[0056] 4) Estimation phase: Position estimation: Perform weighted averaging on the resampled particles or select the particle with the largest weight to obtain the estimated result of the current position. If the weighted averaging method is used, the estimated position (x est ,y est ,θ est ) can be obtained by the formula Calculated, where N is the number of particles. You can also directly select the particle with the largest weight as the estimated position. Finally, the estimated result is determined as the first estimated position.

[0057] In some embodiments, the inspection robot 1 is also used to: determine a second estimated position of the inspection robot 1 based on terrain feature information around the current position and stored campus map information; determine an algorithm estimated position of the inspection robot 1 based on the first estimated position and the second estimated position; and determine whether the inspection robot 1 deviates from the inspection path based on the algorithm estimated position.

[0058] Specifically, first, the inspection robot 1 uses a map matching algorithm to determine the second estimated position based on the terrain feature information around the current position and the stored campus map information. Specifically, the inspection robot 1 will collect terrain feature information around the current position, such as the outline of surrounding buildings, the direction of the road, the distribution of obstacles, etc., through its own sensors, such as lidar, visual sensors, etc. At the same time, it will access the campus map information stored in its system, and match and compare the collected terrain features with the information in the map. Using the map matching algorithm, find the position on the map that best matches the currently perceived terrain features to determine the second estimated position.

[0059] Next, the algorithmic estimated position of the inspection robot 1 is determined based on the first estimated position and the second estimated position. The first estimated position here is calculated by a particle filter algorithm in combination with motion model data and sensor observation data, while the second estimated position is obtained by a map matching algorithm. In order to obtain a more accurate algorithmic estimated position, the first estimated position and the second estimated position can be fused by weighted averaging, Kalman filter fusion or other fusion strategies. For example, different weights are assigned to the two algorithms based on their reliability and performance in different environments, and then the weighted average is calculated as the final algorithmic estimated position to fully utilize the advantages of the two algorithms and obtain a more accurate position estimate.

[0060] Finally, the position estimated by the algorithm is used to determine whether the inspection robot 1 has deviated from the inspection path. Before starting the inspection task, the inspection robot 1 will be set with an inspection path, which may be composed of a series of preset coordinate points. Once the algorithm-estimated position is obtained, the robot will compare the position with the preset inspection path and calculate the deviation between the two. If the deviation exceeds a certain threshold, it is considered that the robot has deviated from the inspection path. At this point, the robot can take corresponding adjustment measures based on the relative relationship between the algorithm-estimated position and the inspection path, such as adjusting its own movement direction and speed, so that it can return to the inspection path to ensure the smooth progress of the inspection task.

[0061] In this way, the inspection robot 1 comprehensively utilizes different positioning algorithms and information sources, which can not only maintain a relatively accurate position estimation in various complex environments, but also effectively monitor whether it deviates from the inspection path, thereby improving the reliability and efficiency of the inspection and ensuring the high-quality completion of the smart park inspection tasks.

[0062] In some embodiments, the inspection robot 1 is specifically used to determine the one with higher confidence between the first estimated position and the second estimated position as the algorithm estimated position when the difference between the first estimated position and the second estimated position is greater than a set threshold.

[0063] Among them, when the difference between the two estimated positions is greater than the set threshold, it means that the two position estimation methods have produced a large disagreement. In order to determine the final algorithm estimated position, the inspection robot 1 needs to evaluate the confidence of the first estimated position and the second estimated position. For the first estimated position, its confidence can be judged based on the distribution of particles in the particle filter algorithm, the concentration of weights, and the reliability of the motion model and sensor data. For example, if the particles are concentrated in a smaller area and the weight distribution is relatively uniform, it may mean that the confidence of the first estimated position is higher. For the second estimated position, its confidence can be evaluated based on the similarity score of the match in the map matching algorithm, the consistency of the search results, and the accuracy and detail of the park map. For example, if the terrain features match the map information very well and the search process is stable, its confidence is correspondingly high.

[0064] Ultimately, the robot will determine the position with a higher confidence level between the first estimated position and the second estimated position as the algorithm estimated position to ensure that the estimated position finally adopted is more reliable, thereby providing a more accurate position reference for subsequent inspection operations, ensuring that the inspection robot 1 can successfully complete the inspection task along the preset path, and also helping to promptly discover and deal with various abnormal situations in the park.

[0065] In some embodiments, the terrain feature information includes laser scanning information, the park map information includes three-dimensional model data of multiple inspection points in the smart park, and the inspection robot 1 is specifically used to compare the laser scanning information with the three-dimensional model data to determine the target inspection point from the multiple inspection points, and determine the position of the target inspection point as the second estimated position.

[0066] The laser scanning information is obtained by the laser scanning device carried by the robot, which can accurately reflect the distance and shape of the robot's surrounding environment. The park map information contains the 3D model data of multiple inspection points in the smart park. These 3D model data are detailed and accurate spatial modeling of each inspection point in the park, providing a complete location reference for the robot.

[0067] During specific operations, the inspection robot 1 will carefully compare the acquired laser scanning information with the stored three-dimensional model data. It will match and compare the data such as the distance, angle and shape of the surrounding environment obtained by the laser scanning with the corresponding information of each inspection point in the three-dimensional model data. In this process, the robot will look for the part that best matches the current laser scanning information, and determine the target inspection point from multiple inspection points by comparing and matching each feature one by one. Once the target inspection point is found, the robot will use the position of the target inspection point as the second estimated position. Such operations can help the inspection robot 1 use the existing park map resources in a complex park environment to more accurately determine its own position through the comparison of laser scanning information, provide a reliable position basis for subsequent inspection tasks and position correction, and thus ensure the efficient and accurate operation of the entire inspection system.

[0068] In some embodiments, the system also includes an environmental sensor 5, which is used to detect environmental information of the smart park and send the environmental information to the management server 2. The management server 2 is used to send abnormal event prompt information to the park management terminal 3 when the environmental information exceeds a normal range.

[0069] Environmental sensors are deployed at key locations in the smart park and are specifically used to detect the park's environmental information, which can cover multiple aspects, such as temperature, humidity, air quality, light intensity, noise level, concentration of harmful gases, etc. The environmental sensor 5 will continuously monitor these environmental parameters in the park and send the collected information to the management server 2 in real time.

[0070] As the information processing and control center of the entire system, the management server 2 will analyze and judge the received environmental information. It pre-sets the normal range of various environmental information, and when receiving data from the environmental sensor 5, it will compare it with the normal range. Once it is found that the environmental information exceeds the normal range, such as the temperature is too high or too low, the concentration of harmful gases exceeds the standard, the noise is too loud, etc., the management server 2 will immediately take action and send abnormal event prompt information to the park management terminal 3. In this way, the park management personnel can receive notifications in time through the park management terminal 3, and grasp the abnormal conditions of the park environment at the first time, so as to quickly take corresponding measures, such as starting the ventilation system, adjusting the temperature control equipment, conducting environmental management or evacuating personnel, etc., to ensure the safety and comfort of the environment in the park, and to ensure the normal operation of the park and the health and safety of personnel.

[0071] In some embodiments, the park management terminal 3 is used to send abnormal event processing completion information to the management server 2 after the abnormal event is processed.

[0072] Among them, the park management terminal 3 is an important interface for managers to interact with the system. When the management server 2 sends abnormal event prompt information to the park management terminal 3, the manager will handle the corresponding abnormal event according to the prompt information. After completing the processing operation of the abnormal event, such as repairing equipment failures, eliminating environmental hazards, solving safety problems, etc., the park management terminal 3 will play its feedback function.

[0073] The park management terminal 3 will send abnormal event processing completion information to the management server 2, which is intended to inform the management server 2 that the abnormal event has been properly handled. This information transmission is very important, as it allows the management server 2 to update the status of the abnormal event and mark the abnormal event as processed. It can also adjust related system settings or trigger subsequent task arrangements based on the processing results. For example, if the abnormal event is a device failure, after the management server 2 receives the processing completion information, it can update the maintenance record of the device and re-evaluate the operating status of the device; if it is an environmental abnormality, the management server 2 can restart the monitoring and analysis process of the environmental data to ensure that the operation and management of the park return to normal, and can continuously ensure the safe, efficient and stable operation of the park.

[0074] In some embodiments, the management server 2 is used to generate a processing result report of the abnormal event according to the abnormal event processing completion information, and send the processing result report to the resident enterprise terminal 4.

[0075] Among them, the management server 2 undertakes important information processing and distribution tasks. When it receives the abnormal event processing completion information from the park management terminal 3, it will start to organize and analyze the entire abnormal event processing process. First, it will generate a complete abnormal event processing result report based on various information from the occurrence to the processing of the abnormal event, including the type of abnormal event, the time of discovery, the area or equipment involved, the specific measures for processing, and the final processing results.

[0076] This report will contain detailed data and information. For example, if the abnormal event is a device failure, the report will detail the name of the faulty device, its location, the specific manifestation of the fault, the maintenance steps taken by the maintenance personnel, and whether the repair was successful in the end. Then, the management server 2 will send this processing result report to the resident enterprise terminal 4. The purpose of this is to enable the resident enterprises to promptly understand the abnormal events that occurred in the park and their handling, so that the enterprises can know the stability and reliability of the park's operations, so that the enterprises can adjust their own operating plans or production arrangements based on this information. At the same time, it also helps enterprises to evaluate the service quality and management level of the park, enhance information communication and trust between enterprises and the park, and jointly promote the good operation and development of the smart park.

[0077] In some embodiments, the resident enterprise terminal 4 is used to send an enterprise evaluation of the processing result of the abnormal event to the management server 2.

[0078] Specifically, when the resident enterprise receives the abnormal event handling result report from the management server 2, the enterprise can send the enterprise evaluation of the handling result of the abnormal event to the management server 2 through the resident enterprise terminal 4 according to its own needs and satisfaction with the abnormal event handling result.

[0079] This evaluation can cover multiple aspects, such as the enterprise can evaluate the timeliness of the processing, that is, whether the time spent from the discovery of the abnormal event to the completion of the processing is within the acceptable range of the enterprise; the effect of the processing can also be evaluated, such as whether the abnormal event is completely resolved and whether it will have a subsequent impact on the normal operation of the enterprise; the communication and coordination involved in the processing process, the service attitude of the personnel, etc. can also be evaluated. The enterprise sends these evaluation information to the management server 2, which allows the management server 2 to understand the needs of the enterprise and its views on the handling of abnormal events, and helps the management server 2 to improve its own services and management processes according to the feedback of the enterprise, continuously optimize the abnormal event handling mechanism, improve the overall service quality, create a better business environment for the settled enterprises, and thus promote the sustainable development of the smart park and improve the satisfaction of the settled enterprises. In some embodiments, the management server 2 is also used to send an abnormal event processing inspection instruction to the inspection robot 1 according to the abnormal event processing completion information, and the inspection robot 1 is used to move to the place where the abnormal event occurs according to the abnormal event processing inspection instruction, and send the captured processing result inspection image to the management server 2, and the management server 2 is used to determine the processing result management evaluation of the abnormal event according to the processing result inspection image.

[0080] Among them, the management server 2 plays a multi-faceted coordination and supervision function. When it receives the abnormal event processing completion information, it will send an abnormal event processing inspection instruction to the inspection robot 1. This is to further check and confirm the processed abnormal events to ensure that the processing results meet the expected standards.

[0081] After receiving the instruction, the inspection robot 1 will move to the location where the abnormal event occurred according to the instruction information. During this process, the inspection robot 1 will use its own movement and positioning functions to accurately reach the designated location. After arriving, the inspection robot 1 will use its equipped image acquisition device to take a picture of the processed situation, thereby obtaining a processing result inspection image.

[0082] Subsequently, the inspection robot 1 sends the captured inspection images of the processing results to the management server 2. After receiving these images, the management server 2 will evaluate the processing results of the abnormal event based on the information in the images. It will carefully analyze the images to see if there are any residual problems. For example, if the abnormal event is equipment damage, it will check whether the equipment has been completely repaired and whether the surrounding environment has returned to normal, so as to determine the management evaluation of the processing results of the abnormal event. This management evaluation can help the park management understand the actual situation of the processing results, discover possible problems in a timely manner, and then take corresponding improvement measures to ensure the safety and stable operation of the park. At the same time, it also provides references and lessons for the subsequent abnormal event processing.

[0083] In this embodiment, the inspection robot 1 is equipped with an image acquisition device for obtaining inspection images during the inspection process. The device can clearly and accurately record the images of each inspection location in the park, providing an important basis for the subsequent abnormal event judgment of the management server. At the same time, it is also equipped with a satellite positioning system for positioning to ensure that its own position is accurately grasped when the signal is good.

[0084] In order to deal with the situation where the satellite positioning signal strength is less than the set threshold, the robot is also equipped with a series of sensors, including laser radar, visual sensors, etc. These sensors can collect information on the terrain characteristics of the park. For example, laser radar can obtain accurate distance and terrain contour information, and visual sensors can obtain visual features of the environment, providing strong support for position estimation based on motion model data and sensor observation data. In addition, the robot has moving parts, such as motors, transmission devices, etc., which can achieve movement in different directions and speeds, so as to move smoothly in the smart park according to the set inspection path.

[0085] The inspection robot 1 may also include a data storage unit for storing park map information, such as three-dimensional model data of multiple inspection points in the smart park, to help the robot determine the second estimated position based on the terrain feature information around the current position and the stored park map information. In addition, the robot is equipped with a communication module, which can send the collected inspection information to the management server, and receive information such as abnormal event processing inspection instructions from the management server, to ensure information exchange and coordinated operation between the robot and other parts of the system, such as the management server, the park management terminal and the resident enterprise terminal, and provide strong hardware support for efficient inspection and abnormal event management of the smart park.

[0086] The embodiment of the present invention provides a smart park management system, which includes multiple components and multiple functions, and is intended to achieve efficient and intelligent park management and inspection. The system mainly includes an inspection robot 1, a management server 2, a park management terminal 3, a resident enterprise terminal 4, and an environmental sensor 5. The inspection robot 1 will carry out inspection work in the smart park according to the set inspection path, and send inspection information such as inspection images, inspection locations, and inspection time to the management server 2. The management server 2 determines whether there are abnormal events at the inspection location based on the inspection images, and sends abnormal event prompt information to the park management terminal 3 and the resident enterprise terminal 4.

[0087] When the satellite positioning signal strength is insufficient, the inspection robot 1 can use its own motion model data (including the position information, moving speed and moving direction of the last moment) and sensor observation data (including the terrain feature information of the park) to estimate the current position. The specific method is as follows: first obtain a group of particles with the same weight and representing its own possible position and posture, propagate the particles to the new position according to the motion model data to obtain a priori estimated values ​​to predict the current possible position; then calculate the likelihood of the particle state through the preset measurement model according to the sensor measurement data, and update the particle weight; then perform a resampling operation according to the weight, discard low-weight particles and copy high-weight particles, so that the particle set is closer to the real state area; finally, weighted average the resampled particles or select the particle with the largest weight to obtain the first estimated position.

[0088] The inspection robot 1 will also determine a second estimated position based on the terrain feature information of the current position and the stored park map information, where the terrain feature information includes laser scanning information, and the park map information contains three-dimensional model data of multiple inspection points. The target inspection point is determined by comparing the laser scanning information with the three-dimensional model data and its position is used as the second estimated position; the algorithm estimated position can be determined by combining the first and second estimated positions. When the difference between the two is greater than a threshold, the one with a higher confidence level is selected as the algorithm estimated position, and then it is determined whether it deviates from the inspection path.

[0089] The environmental sensor 5 can detect the park environment information and send it to the management server 2. When the environmental information exceeds the normal range, the management server 2 will send abnormal event prompt information to the park management terminal 3. When the abnormal event is processed, the park management terminal 3 will send the processing completion information to the management server 2. The management server 2 will generate an abnormal event processing result report and send it to the resident enterprise terminal 4. The resident enterprise terminal 4 can send the processing result enterprise evaluation. In addition, the management server 2 will send an abnormal event processing inspection instruction to the inspection robot 1 based on the processing completion information. The inspection robot 1 will go to the place where the abnormal event occurred to take the processing result inspection image and send it to the management server 2. The management server 2 determines the management evaluation of the abnormal event processing result. Through the coordinated operation of these components and functions, the system realizes a comprehensive management process from inspection, abnormal event detection, location estimation to abnormal processing and subsequent evaluation, which improves the intelligence and efficiency of smart park management.

[0090] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A smart park management system, characterized in that: Including inspection robots, management servers, park management terminals and enterprise terminals; The inspection robot is used to perform inspections in the smart park according to a set inspection path, and send the collected inspection information to the management server, wherein the inspection information includes inspection images, inspection locations and inspection times; The management server is used to determine that an abnormal event exists at the inspection location according to the inspection image, and send abnormal event prompt information to the park management terminal and the resident enterprise terminal; The inspection robot is also used to estimate the current position to obtain a first estimated position based on its own motion model data and sensor observation data when the satellite positioning signal strength is less than a set signal strength threshold, and to adjust its own motion state when it is determined that the inspection robot deviates from the inspection path based on the first estimated position; the motion model data includes the position information, moving speed and moving direction of the inspection robot at the previous moment, and the sensor observation data includes the terrain feature information of the park collected by the sensor.

2. The system according to claim 1, characterized in that The inspection robot is specifically used for: Obtain a group of particles with the same weight, wherein the particles represent possible positions and postures of the inspection robot; Propagate the particle to a new position according to the motion model data to obtain a priori estimated value of the particle to predict the possible current position of the inspection robot; Based on the sensor measurement data, the likelihood of the state represented by each particle is calculated by a preset measurement model to update the weight of the particle; Resampling is performed according to the weights of the particles, the particles with lower weights are discarded, and the particles with higher weights are copied, so that the particle set is more concentrated in the possible area of ​​the real state of the inspection robot; The resampled particles are weighted averaged or the particle with the largest weight is selected to obtain an estimation result of the previous position, and the estimation result is determined as the first estimated position.

3. The system according to claim 1 or 2, characterized in that: The inspection robot is also used for: Determine a second estimated position of the inspection robot based on terrain feature information around the current position and stored park map information; Determining an algorithm-estimated position of the inspection robot according to the first estimated position and the second estimated position; It is determined whether the inspection robot deviates from the inspection path according to the position estimated by the algorithm.

4. The system according to claim 3, characterized in that The inspection robot is specifically used to determine the one with higher confidence between the first estimated position and the second estimated position as the algorithm estimated position when the difference between the first estimated position and the second estimated position is greater than a set threshold.

5. The system according to claim 3, characterized in that The terrain feature information includes laser scanning information, and the park map information includes three-dimensional model data of multiple inspection points in the smart park. The inspection robot is specifically used to compare the laser scanning information with the three-dimensional model data to determine the target inspection point from the multiple inspection points, and determine the position of the target inspection point as the second estimated position.

6. The system according to claim 1, characterized in that It also includes an environmental sensor, which is used to detect environmental information of the smart park and send the environmental information to the management server. The management server is used to send abnormal event prompt information to the park management terminal when the environmental information exceeds a normal range.

7. The system according to claim 1, characterized in that The park management terminal is used to send abnormal event processing completion information to the management server after the abnormal event is processed.

8. The system according to claim 7, characterized in that The management server is used to generate a processing result report of the abnormal event according to the abnormal event processing completion information, and send the processing result report to the resident enterprise terminal.

9. The system according to claim 8, characterized in that The resident enterprise terminal is used to send an enterprise evaluation of the processing result of the abnormal event to the management server.

10. The system according to claim 7, characterized in that The management server is also used to send an abnormal event processing inspection instruction to the inspection robot based on the abnormal event processing completion information. The inspection robot is used to move to the location where the abnormal event occurred according to the abnormal event processing inspection instruction, and send the captured processing result inspection image to the management server. The management server is used to determine the processing result management evaluation of the abnormal event based on the processing result inspection image.

Images