Safety patrol method of sensing robot for smart parking lot

Abstract

The invention discloses a safety patrol method of a sensing robot for an intelligent parking lot, and relates to the technical field of intelligent sensing, and the method comprises the steps: collecting parking lot layout information, constructing a parking lot map, and presetting a fixed patrol route of the sensing robot; the sensing robot executes safety patrol based on a fixed patrol route, senses the environment of the parking lot in real time, and obtains environment information of the parking lot; and carrying out synchronous path dynamic optimization and abnormity identification based on the parking lot environment information, obtaining an abnormity identification result, evaluating a safety risk, and carrying out corresponding safety early warning according to a safety evaluation level. The technical problems that in the prior art, manual patrol efficiency of a traditional parking lot is low, and environment changes and safety events cannot be responded rapidly in real time are solved, and the technical effects that intelligent and automatic safety patrol is achieved through the sensing robot, and the safety and management efficiency of the parking lot are improved are achieved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent sensing technology, and in particular to a safety patrol method for a sensing robot used in a smart parking lot. Background Art

[0002] In modern cities, parking has become an increasingly serious traffic problem, especially in large shopping malls, office buildings, residential areas and other places. The demand for parking continues to grow, making parking lot management more complicated. Traditional parking lot management methods mainly rely on manual patrols and monitoring systems. This method not only has high labor costs and low efficiency, but also makes it difficult to respond to sudden safety issues in real time, and cannot effectively ensure the safety and order of parking lots. Summary of the invention

[0003] The present application provides a security patrol method for a smart parking lot using a perception robot, which is used to solve the technical problems in the prior art of low efficiency of manual patrols in traditional parking lots and the inability to respond quickly to environmental changes and security incidents in real time.

[0004] The present application provides a security patrol method for a perception robot used in a smart parking lot, the method comprising: collecting parking lot layout information and constructing a parking lot map; presetting a fixed patrol route for the perception robot based on the parking lot map; having the perception robot perform security patrols based on the fixed patrol route, perceive the parking lot environment in real time, and obtain parking lot environment information, the parking lot environment information including parking space status information, obstacle location information, and personnel activity information; performing synchronous path dynamic optimization and anomaly identification based on the parking lot environment information to obtain anomaly identification results; evaluating safety risks based on the anomaly identification results, and issuing corresponding safety warnings based on the safety assessment level.

[0005] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0006] The security patrol method for a perception robot used in a smart parking lot provided in the present application relates to the field of intelligent perception technology. By constructing a parking lot map and presetting a patrol route, the robot performs patrols based on the route, obtains environmental information such as parking space status, obstacle locations, and personnel activities in real time, and dynamically optimizes paths and abnormalities. Traditional parking lot manual patrols are inefficient and cannot respond to environmental changes and security incidents in real time. Rapid response identification, safety risks are assessed and corresponding safety warnings are generated. The technical problems of low efficiency of traditional parking lot manual patrols and inability to respond to environmental changes and security incidents in real time in the prior art are solved, and the technical effect of realizing intelligent and automated security patrols through perception robots and improving the safety and management efficiency of parking lots is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0008] Figure 1 A schematic diagram of a safety patrol method for a smart parking lot using a perception robot provided in an embodiment of the present application;

[0009] Figure 2 A schematic flowchart of a fixed patrol route for a perception robot preset in a safety patrol method for a perception robot for a smart parking lot provided in an embodiment of the present application. DETAILED DESCRIPTION

[0010] The present application provides a security patrol method for a smart parking lot using a perception robot, which is used to solve the technical problems in the prior art of low efficiency of manual patrols in traditional parking lots and the inability to respond quickly to environmental changes and security incidents in real time.

[0011] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0012] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices.

[0013] Embodiment 1, as Figure 1 As shown, the present application provides a safety patrol method for a perception robot in a smart parking lot, the method comprising:

[0014] P10: Collect parking lot layout information and build a parking lot map.

[0015] Furthermore, step P10 of the embodiment of the present application further includes:

[0016] P11: Obtain a parking space distribution design diagram of a target parking lot; P12: Use distributed sensing equipment to obtain obstacle distribution information of the target parking lot; P13: Based on the parking space distribution design diagram and obstacle distribution information, generate a digital map of the target parking lot through simultaneous positioning and mapping.

[0017] It should be understood that the process of collecting parking lot layout information and building a parking lot map is divided into several key links to ensure that the patrol robot can effectively perform its tasks, namely, information collection, processing and map generation.

[0018] First, obtain the parking space distribution design diagram of the target parking lot. The parking space distribution design diagram is the basic layout file of the parking lot, usually generated from the parking lot design drawings or CAD files. It details the location, shape and relative relationship of each parking space in the parking lot. By obtaining the parking space distribution design diagram, the system can clearly understand the distribution of parking spaces and provide a basis for subsequent patrol route planning.

[0019] Next, the distribution information of obstacles in the parking lot is obtained by using distributed sensing equipment. These sensing devices include lidar, ultrasonic sensors, infrared sensors, etc., which are widely deployed in different locations of the parking lot to monitor and record obstacles in the environment in real time. These obstacles can be fixed, such as pillars, walls, etc., or dynamic, such as parked vehicles, temporarily stacked items, etc. Distributed sensing equipment ensures that the perception robot can obtain obstacle information in the parking lot in a timely manner through high-precision real-time perception, so as to facilitate subsequent path planning and obstacle avoidance operations.

[0020] Finally, based on the parking space distribution design and obstacle distribution information, a digital map of the target parking lot is generated through simultaneous localization and mapping (SLAM) technology. SLAM technology combines the obstacle data and parking space design acquired by the sensor, and constructs an accurate digital map of the parking lot by combining real-time positioning with environmental data. This map not only shows the structural information of the parking lot (such as parking space location and layout), but also includes the spatial distribution of obstacles, which can provide a full range of environmental perception views for the perception robot, thereby supporting its patrol, path planning and abnormality recognition in the parking lot.

[0021] Through the close combination of the above steps, the perception robot can accurately understand the layout of the parking lot, dynamically perceive environmental changes, and provide accurate data support for subsequent security patrols, anomaly identification and risk assessment.

[0022] P20: According to the parking lot map, a fixed patrol route of the perception robot is preset.

[0023] Further, such as Figure 2 As shown, step P20 of the embodiment of the present application also includes:

[0024] P21: Based on the parking lot map, perform regional division and regional risk assessment to generate regional risk assessment results; P22: Generate regional priority ranking with reference to the regional risk assessment results; P23: Based on the regional priority ranking and combined with the parking lot design route, generate a fixed patrol route for the perception robot, and the fixed patrol route has reserved space for dynamic adjustment.

[0025] Optionally, in order to ensure patrol efficiency and safety, a fixed patrol route for the perception robot is preset according to the constructed parking lot map.

[0026] First, the parking lot is divided into several areas based on the constructed parking lot map. This step divides the parking lot into several different areas according to the structural characteristics of the parking lot. For example, the parking lot can be divided into entrance area, exit area, parking area, walkway area, etc. according to factors such as parking space distribution, traffic flow, and obstacle concentration. Each area will face different security risks according to its function and importance.

[0027] After the division is completed, a sub-regional risk assessment is conducted. By analyzing the characteristics of each area, such as the density of parking spaces, the frequency of vehicle and personnel activities, potential obstacles, etc., the security risk of each area is assessed. For example, areas close to entrances and exits may have higher traffic density and personnel activities, and the risk is relatively high; while areas far from the main passages may be relatively low. The results of the sub-regional risk assessment provide a basis for the subsequent optimization of patrol routes.

[0028] Based on the results of the above regional risk assessment, each area is prioritized. Areas with higher risk assessment results will be given higher priority, which means that the perception robot needs to stay in these areas longer or patrol more frequently. This process helps ensure that the robot's patrol tasks can be concentrated in high-risk areas to ensure that the safety management of the parking lot is maximized.

[0029] After completing the regional priority sorting, the fixed patrol route of the perception robot will be generated in combination with the design route of the parking lot, the layout of parking spaces, and other environmental factors. This route is not only a fixed patrol path based on the existing parking lot layout and risk assessment results, but also needs to leave room for dynamic adjustment. Considering the changes in obstacles, personnel mobility, and the occurrence of emergencies in the parking lot, the patrol route should have a certain degree of flexibility and be able to be adjusted according to real-time perception information. In this way, even if the parking lot environment changes, the perception robot can still perform patrol tasks efficiently.

[0030] Through the above steps, not only the parking lot patrol route is preset, but also the flexibility and dynamic adaptability of the route are ensured, which provides a guarantee for the perception robot to complete the patrol task efficiently and safely.

[0031] P30: The perception robot performs security patrols based on the fixed patrol route, perceives the parking lot environment in real time, and obtains parking lot environment information, wherein the parking lot environment information includes parking space status information, obstacle location information, and personnel activity information.

[0032] Furthermore, step P30 of the embodiment of the present application further includes:

[0033] P31: The perception robot is equipped with a multi-sensor module, which includes an image sensor device, a distance sensor device and an infrared sensor device; P32: The parking space occupancy status information is obtained through the image sensor device; P33: The distance sensor device is used to perceive the position and dynamic changes of obstacles and obtain the obstacle position information; P34: The infrared sensor device is used to monitor the activities of people in the parking lot and obtain information on people's activities.

[0034] Specifically, the perception robot performs security patrols based on preset fixed patrol routes, and perceives the parking lot environment in real time to obtain key environmental information of the parking lot, including parking space status, obstacle location information, and personnel activity information. In order to achieve these goals, the perception robot needs to be equipped with a variety of sensing technologies to be able to obtain and process various types of information in the parking lot in real time and accurately.

[0035] First, the perception robot is equipped with a multi-sensor module, which integrates multiple types of sensors, which are used to obtain different environmental information in the parking lot. The multi-sensor module includes three main sensor devices: image sensor device, distance sensor device and infrared sensor device. Each sensor device undertakes a specific environmental perception task to ensure that the robot can obtain parking lot environmental data comprehensively and accurately.

[0036] The image sensor device is mainly used to collect visual information in the parking lot. Through a high-resolution camera, the robot can capture the status of each parking space in the parking lot in real time and obtain the occupancy information of the parking space. By analyzing the captured image data, the image sensor device can determine whether the parking space is occupied by a vehicle and whether the occupancy is abnormal (for example, there are obstacles in the parking space, or parking is not standardized, etc.), thereby providing important data for subsequent path planning and anomaly detection.

[0037] The distance sensing devices (such as laser radar, ultrasonic sensors, etc.) are used to accurately sense the location of obstacles in the parking lot and their dynamic changes. These sensors can obtain the distance information of obstacles in real time by transmitting and receiving signals, and form detailed three-dimensional data of the environment through the sensor array. This not only helps the robot identify fixed obstacles, but also tracks the location information of dynamic obstacles (such as vehicles entering or temporarily stacked items), providing data support for the robot's obstacle avoidance and dynamic adjustment of patrol paths.

[0038] In addition, the infrared sensor device is used to monitor the activities of people in the parking lot. Through the infrared sensor, the robot can detect the heat source of human activities and determine the location of people and their dynamic behavior. The infrared sensor device is sensitive to temperature changes, so it can still effectively monitor the activities of people even in low light or night environments. This is crucial for identifying potential safety risks (such as people staying, malicious behavior, etc.), and can provide real-time feedback on personnel activities to support safety assessment and early warning.

[0039] Through the collaborative work of the above-mentioned multiple sensing devices, the perception robot can comprehensively and accurately perceive the parking lot environment, obtain parking space status, obstacle location information and personnel activity information in real time, provide data support for subsequent dynamic path optimization and anomaly identification, and ensure the efficiency and accuracy of parking lot safety patrols.

[0040] P40: Perform synchronous path dynamic optimization and anomaly identification based on the parking lot environment information to obtain anomaly identification results.

[0041] Furthermore, step P40 of the embodiment of the present application also includes:

[0042] P41: updating the parking lot map in real time according to the parking lot environment information; P42: dynamically adjusting the fixed patrol route based on the updated parking lot map in combination with the path planning algorithm to generate an optimized patrol route.

[0043] It should be understood that based on the parking lot environment information obtained in real time, synchronous path dynamic optimization and anomaly identification are performed to ensure that the robot can adapt to environmental changes and effectively complete patrol tasks.

[0044] First, the perception robot updates the original parking map in real time based on the collected parking environment information (including parking space status, obstacle location, and personnel activities). This update process relies on the real-time data continuously collected by the perception robot and integrated into the map. By continuously capturing changes in the parking lot through sensors (for example, new obstacles appear or parking spaces are occupied), the map update enables the robot to always obtain the latest status of the current parking environment. The core of real-time map updates is to be able to adjust the robot's patrol behavior according to the dynamic changes of the environment, ensuring that the robot can maintain accurate path planning in a constantly changing environment.

[0045] After the parking lot map is updated in real time, the next step is to dynamically adjust the fixed patrol route based on the updated parking lot map and the path planning algorithm. Traditional patrol routes are often static, but in actual applications, the parking lot environment often changes (for example, the appearance of obstacles or changes in parking spaces), so the patrol route also needs to be dynamically adjusted accordingly. The path planning algorithm will optimize the patrol route in real time based on the latest map information, combined with factors such as the priority of the target area and the complexity of the environment.

[0046] Optimizing patrol routes is not just about avoiding obstacles, but also about considering multiple factors such as the safety risk assessment results of the parking lot, the activities of vehicles and people, etc. For example, if the safety risk of a certain area increases (such as a parking area suddenly becomes more crowded or there is abnormal human activity), the path planning algorithm will prioritize adjusting the patrol route so that the robot can patrol the area more frequently to ensure safety.

[0047] Through this dynamic adjustment, the robot can perform more efficient and safer patrol tasks in the complex and ever-changing parking environment. At the same time, the generation of this optimized patrol route can also greatly improve the robot's patrol efficiency and abnormality recognition ability, thereby achieving better safety warning effects.

[0048] Furthermore, step P40 of the embodiment of the present application also includes:

[0049] P43: Based on a machine learning algorithm, train and generate an abnormality recognition model, wherein the abnormality recognition model includes a dynamic abnormality recognition unit and a static abnormality recognition unit; P44: Use the dynamic abnormality recognition unit and the static abnormality recognition unit to synchronously recognize the parking lot environment information and generate an abnormality recognition result.

[0050] In a possible embodiment of the present application, by integrating a machine learning algorithm, the perception robot can perform a more intelligent analysis of parking environment information and identify potential abnormal situations in real time.

[0051] To further improve the accuracy and real-time performance of anomaly recognition, the robot uses machine learning algorithms to train and generate a comprehensive anomaly recognition model. The goal of this model is to automatically identify possible anomalies by analyzing real-time data of the parking environment. The anomaly recognition model consists of two main units: a dynamic anomaly recognition unit and a static anomaly recognition unit. The dynamic anomaly recognition unit is mainly responsible for identifying real-time anomalies, such as obstacles that suddenly appear in the parking lot, illegal parking of vehicles, or abnormal activities of people; the static anomaly recognition unit focuses on identifying long-term potential problems in the parking lot, such as long-term unused parking spaces, long-term accumulated obstacles, etc. By combining these two types of anomaly recognition, the robot can comprehensively monitor the parking lot environment and ensure that safety hazards are discovered in a timely manner.

[0052] The perception robot simultaneously analyzes the environmental information of the parking lot by activating the dynamic anomaly recognition unit and the static anomaly recognition unit. The dynamic anomaly recognition unit processes real-time sensor data, such as the movement of obstacles, illegal parking of vehicles, and abnormal behavior of personnel; while the static anomaly recognition unit analyzes the long-term changes in the parking lot and identifies possible long-term hidden dangers or equipment damage. Through comprehensive analysis of these data, the robot can generate anomaly recognition results in real time and take corresponding actions based on the identified anomalies, such as adjusting patrol routes or issuing safety warnings, to ensure the safety and stable operation of the parking lot.

[0053] Through this dynamic optimization and anomaly recognition mechanism, the perception robot can efficiently perform security patrol tasks in complex and changing parking environments, improving the accuracy, timeliness and safety of patrols.

[0054] Furthermore, step P43 of the embodiment of the present application also includes:

[0055] P43-1: Extract the historical abnormal event recognition records of the target parking lot, perform feature identification, and then divide and generate a training data set and a verification data set; P43-2: Based on the training data set and the verification data set, supervised training is performed in combination with a machine learning algorithm to generate the abnormality recognition model.

[0056] Specifically, a machine learning algorithm is used to train and generate an anomaly recognition model that can recognize various abnormal situations in the parking lot.

[0057] Before training the anomaly recognition model, it is necessary to first collect and organize the historical anomaly event recognition records of the target parking lot. These records may include past safety incidents, illegal parking, improper behavior of people in the parking lot, the presence of obstacles or dangerous objects, etc. By feature-marking these historical events, key features can be extracted from them, such as the time and location of the event, the behavior patterns of the objects or people involved, the type of anomaly, etc. These features provide input data for the machine learning algorithm.

[0058] Then, all the collected historical abnormal event records will be divided into training data sets and validation data sets. The training data set is used to train the anomaly recognition model to help the model learn how to identify different types of abnormal behaviors. The validation data set is used to evaluate the accuracy and generalization ability of the model during the model training process to ensure that the model can still make accurate predictions when facing new and unseen abnormal events.

[0059] After preparing the training data set and the validation data set, the next task is to use the supervised learning machine learning algorithm to train the anomaly recognition model. In this process, the feature data and labels (i.e. the category of abnormal events) in the training data set will be used as input. The machine learning algorithm will learn how to identify abnormal events according to changes in the parking environment by continuously adjusting the model parameters.

[0060] Through this training method, the model will be able to gradually learn how to associate dynamic changes in the parking environment (such as obstacles, parking space status, etc.) with labeled abnormal events. After training, the model will be able to accurately identify potential anomalies in the parking lot and make corresponding judgments based on real-time environmental data. The validation data set is used to check the effectiveness of the model to ensure that its recognition ability does not overfit the training data and has good generalization ability in practical applications.

[0061] Through the combination of these two steps, the perception robot is able to obtain an accurate anomaly recognition model, enabling it to identify abnormal situations in the parking lot in real time and improve the reliability and timeliness of security patrols.

[0062] P50: Assess security risks based on the abnormal identification results and issue corresponding security warnings based on the security assessment level.

[0063] Furthermore, step P50 of the embodiment of the present application also includes:

[0064] P51: Based on the abnormality identification results, safety risk classification is performed to generate abnormality risk levels, which include low risk, medium risk and high risk; P52: In the case of low risk, the abnormality is recorded and the log is uploaded to the parking lot management system; P53: In the case of medium risk, a voice warning is issued by the perception robot or an early warning signal is sent to the management center; P54: In the case of high risk, an emergency alarm is triggered and real-time video and abnormal data are uploaded to the management system.

[0065] It should be understood that the security risks are assessed based on the anomaly identification results, and corresponding security warning measures are taken according to different risk levels.

[0066] First, the perception robot classifies the safety risks in the parking lot based on the abnormality recognition results. By analyzing the severity, frequency, location of the abnormal events and other factors, and performing weighted average calculations, the abnormality risk level is generated, which is usually divided into three types: low risk, medium risk and high risk. Low risk usually refers to some small, non-emergency abnormal events, such as minor illegal parking or unobvious obstacles; medium risk events may include more obvious safety hazards, such as frequent illegal parking or potential channel blockages; high risk events are usually emergency situations that endanger the safety of the parking lot, such as fire, theft or serious obstacles. Through this classification, the robot can take corresponding response measures according to different risk levels.

[0067] When an abnormality is identified in a low-risk situation, the robot will record the abnormality and upload the relevant log information to the parking lot management system. At this time, although the abnormality may not pose a direct threat to overall safety, it still needs to be recorded for subsequent analysis and management. The uploaded log includes the time, location, specific content of the abnormality, and the robot's preliminary assessment results of the abnormal event. Managers can view these logs in the system and decide whether further processing is required.

[0068] In medium-risk situations, the degree of abnormality is high enough to cause concern. At this point, the perception robot will take appropriate warning measures to alert the personnel or managers in the parking lot. Specific measures include reminding people around to pay attention to safety through voice warnings, or sending abnormal information to the parking lot management center through early warning signals. These early warning signals may be in the form of text, voice, email or SMS, ensuring that managers can receive notification of abnormal situations in a timely manner and respond quickly.

[0069] In high-risk situations, the robot will immediately activate the emergency alarm mechanism. At this time, the robot will not only trigger a loud alarm to alert people around, but also upload real-time video and related abnormal data (such as sensor data, image data, etc.) to the parking lot management system. This information will help managers understand the current crisis situation in a timely manner and support them to make faster decisions and responses. High-risk events may involve security threats such as fires, failures or other emergencies that may lead to serious consequences, so they need to be responded to and handled quickly.

[0070] Through these step-by-step measures, the safety risks of parking lots can be effectively assessed, and corresponding response strategies can be taken according to different risk levels to ensure the safety of parking lots.

[0071] In summary, the embodiments of the present application have at least the following technical effects:

[0072] This application collects parking lot layout information and builds a parking lot map, presets a fixed patrol route, and the perception robot performs patrols based on the route, and perceives environmental information such as parking space status, obstacle locations, and personnel activities in real time. Through synchronous path dynamic optimization and anomaly recognition, the anomaly recognition results are obtained, and based on this, the safety risk is evaluated and different levels of safety warnings are generated.

[0073] The technical effect of realizing intelligent and automated security patrols through perception robots and improving the safety and management efficiency of parking lots has been achieved.

[0074] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0075] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0076] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A security patrol method for a smart parking lot using a perception robot, characterized in that: The method comprises: Collect parking lot layout information and build parking lot maps; According to the parking lot map, a fixed patrol route of the perception robot is preset; The perception robot performs security patrols based on the fixed patrol route, perceives the parking lot environment in real time, and obtains parking lot environment information, wherein the parking lot environment information includes parking space status information, obstacle location information, and personnel activity information; Perform synchronous path dynamic optimization and abnormality identification based on the parking lot environment information to obtain abnormality identification results; The security risk is assessed based on the abnormal identification result, and corresponding security warnings are issued according to the security assessment level.

2. The safety patrol method of a perception robot for a smart parking lot according to claim 1, characterized in that: Collect parking lot layout information and build parking lot maps, including: Obtain the parking space distribution design diagram of the target parking lot; Use distributed sensing equipment to obtain obstacle distribution information in the target parking lot; Based on the parking space distribution design diagram and obstacle distribution information, a digital map of the target parking lot is generated through simultaneous positioning and mapping.

3. The safety patrol method of a perception robot for a smart parking lot according to claim 1, characterized in that: According to the parking lot map, a fixed patrol route of the perception robot is preset, including: Based on the parking lot map, perform regional division and regional risk assessment to generate regional risk assessment results; Generate regional priority rankings with reference to the regional risk assessment results; According to the regional priority sorting and in combination with the parking lot design route, a fixed patrol route for the perception robot is generated, and the fixed patrol route has space reserved for dynamic adjustment.

4. The security patrol method of a perception robot for a smart parking lot according to claim 1, characterized in that: The perception robot performs safety patrols based on the fixed patrol route, perceives the parking lot environment in real time, and obtains parking lot environment information, including: The perception robot is equipped with a multi-sensor module, which includes an image sensor, a distance sensor and an infrared sensor; Acquiring parking space occupancy status information through the image sensing device; Using the distance sensor device to sense the position and dynamic changes of obstacles and obtain obstacle position information; The infrared sensor device is used to monitor the activities of people in the parking lot and obtain the information of the activities of people.

5. The safety patrol method of a perception robot for a smart parking lot according to claim 1, characterized in that: Based on the parking lot environment information, synchronous path dynamic optimization and abnormality identification are performed, including: According to the parking lot environment information, the parking lot map is updated in real time; Based on the updated parking lot map, the fixed patrol route is dynamically adjusted in combination with a path planning algorithm to generate an optimized patrol route.

6. The security patrol method of a perception robot for a smart parking lot according to claim 5, characterized in that: Based on the parking lot environment information, synchronous path dynamic optimization and abnormality identification are performed, including: Based on a machine learning algorithm, training generates an anomaly recognition model, wherein the anomaly recognition model includes a dynamic anomaly recognition unit and a static anomaly recognition unit; The dynamic anomaly recognition unit and the static anomaly recognition unit are used to synchronously recognize the parking lot environment information to generate an anomaly recognition result.

7. The safety patrol method of a perception robot for a smart parking lot according to claim 6, characterized in that: Based on machine learning algorithms, training and generating anomaly recognition models include: Extract the historical abnormal event recognition records of the target parking lot, perform feature identification, and divide them into training data set and verification data set; Based on the training data set and the validation data set, supervised training is performed in combination with a machine learning algorithm to generate the anomaly recognition model.

8. The safety patrol method of a perception robot for a smart parking lot according to claim 1, characterized in that: Assessing security risks based on the abnormal identification results, including: According to the abnormality identification result, security risk classification is performed to generate abnormality risk levels, wherein the abnormality risk levels include low risk, medium risk and high risk; In low-risk situations, record the exception and upload the log to the parking management system; In medium-risk situations, the perception robot will issue a voice warning or send an early warning signal to the management center; In high-risk situations, emergency alarms are triggered and real-time video and abnormal data are uploaded to the management system.

Images