Navigation control system and method for inspection robot

Abstract

The invention relates to the technical field of inspection robots, in particular to an inspection robot navigation control system and method.In the system, a main navigation control module is used for conducting navigation control on an inspection robot based on a main navigation algorithm, and a historical data storage module is used for cyclically storing historical operation data of the inspection robot within a preset time period; the operation state diagnosis module is used for acquiring real-time operation data of the inspection robot and performing operation state diagnosis according to the real-time operation data to obtain an operation state, and the emergency navigation control module is used for obtaining a virtual path according to historical operation data when the operation state is an abnormal state and sending the virtual path to the inspection robot. And the virtual path is used for replacing a main navigation algorithm to carry out navigation control on the inspection robot. According to the invention, the robot can still continue to work when encountering sudden abnormal conditions, high equipment shutdown loss caused by shutdown waiting for manual intervention is avoided, and the problem that an inspection robot in the prior art can only stop for waiting in a fault scene is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of inspection robots, and in particular to a navigation control system and method for an inspection robot. Background Art

[0002] Inspection robots are automated devices that integrate multi-sensor fusion, autonomous navigation, and intelligent analysis technologies. They are designed to replace or assist humans in completing equipment status monitoring and safety inspection tasks in high-risk, repetitive industrial scenarios. In modern factory workshops, compared to traditional manual inspection methods, robots significantly reduce labor intensity and safety risks, and maintain stable performance in complex working conditions (such as high temperature, high pressure, and toxic and hazardous environments). As a result, they are widely used in the power, petroleum, chemical, and intelligent manufacturing industries.

[0003] However, the dynamic complexity of industrial scenarios places stringent demands on robot reliability. Crowded equipment within a factory can block wireless communication signals, leading to transmission interruptions. Mechanical vibration and dusty environments can easily cause sensor accuracy drift. Oily floor surfaces and temperature fluctuations can cause the mobile chassis to slip and lose control. If these unexpected failures are not addressed promptly, they can directly impact production continuity.

[0004] Existing technologies generally employ a conservative "failure-as-stop" strategy for these types of failures, suspending operations upon detecting an anomaly and awaiting manual intervention. However, in continuous production scenarios, every minute of equipment downtime can result in tens of thousands of yuan in economic losses. Combined with manual response delays and the risk of secondary failures, the conflict between timeliness and cost-effectiveness of traditional solutions is becoming increasingly prominent. Therefore, a control solution for inspection robots that can maintain operation even in failure scenarios is needed. Summary of the Invention

[0005] Therefore, the present invention provides a navigation control system and method for an inspection robot, so as to solve the problem in the prior art that the inspection robot can only stop and wait in a fault scenario.

[0006] The present invention provides a navigation and control system for an inspection robot, comprising:

[0007] The main navigation control module is used to control the inspection robot's navigation based on the main navigation algorithm;

[0008] A historical data storage module is used to cyclically store the historical operation data of the inspection robot within a preset time period;

[0009] The operation status diagnosis module is used to obtain the real-time operation data of the inspection robot and perform operation status diagnosis based on the real-time operation data to obtain the operation status;

[0010] The emergency navigation control module is used to obtain a virtual path based on historical operation data when the operating state is abnormal, and use the virtual path to replace the main navigation algorithm to navigate and control the inspection robot.

[0011] In a preferred embodiment, when the operation state is abnormal, a virtual path is obtained based on historical operation data, and the virtual path is used to replace the main navigation algorithm to control the inspection robot's navigation, including:

[0012] When the operation status is abnormal, the historical operation data is analyzed to obtain a virtual path;

[0013] Based on historical operation data, the available data of the inspection robot in abnormal state is predicted to obtain predicted data;

[0014] Control the inspection robot's movement based on the virtual path, and collect the available data of the inspection robot in abnormal state to obtain real-time data;

[0015] Based on the difference between predicted data and real-time data, the inspection robot's navigation is corrected.

[0016] In a preferred embodiment, the historical operation data includes historical path coordinates and anchor point positions; when the operation state is abnormal, the historical operation data is analyzed to obtain a virtual path, further comprising:

[0017] Divide the historical path coordinates based on the anchor point positions to obtain multiple sets of periodic path coordinates;

[0018] Take the average value of multiple sets of periodic path coordinates to obtain the initial path coordinates;

[0019] The path formed by the initial path coordinates is smoothed to obtain a virtual path.

[0020] In a preferred embodiment, based on historical operation data, the available data of the inspection robot in an abnormal state is predicted to obtain the predicted data, including:

[0021] Based on the anchor point position, the data collection period is obtained;

[0022] Divide historical operation data based on data collection cycle to obtain multiple groups of periodic data;

[0023] Make predictions based on multiple sets of periodic data to obtain complete prediction data;

[0024] From the complete prediction data, data of the same type as the data that can be obtained by the inspection robot in an abnormal state is filtered out to obtain the prediction data.

[0025] In a preferred embodiment, prediction is performed based on multiple sets of periodic data to obtain complete prediction data, including:

[0026] Evaluate the similarity of multiple sets of periodic data and obtain similar eigenvalues;

[0027] Fit multiple sets of periodic data to obtain fitted prediction data;

[0028] Input multiple sets of periodic data into a preset prediction model to obtain model prediction data;

[0029] According to similar feature values, the fitted prediction data and the model prediction data are fused to obtain the complete prediction data.

[0030] In a preferred embodiment, the navigation of the inspection robot is corrected according to the difference between the predicted data and the real-time data, including:

[0031] Calculate the difference between the predicted data and the real-time data;

[0032] The navigation speed of the inspection robot is corrected according to the difference value, wherein the difference value is inversely proportional to the navigation speed.

[0033] In a preferred embodiment, calculating the difference between the predicted data and the real-time data includes:

[0034] Get the preset priority of each forecast data;

[0035] The difference between the predicted data and the real-time data is calculated, and the difference is weighted and summed with the preset priority as the weight to obtain the difference value.

[0036] In a preferred embodiment, when the operation state is abnormal, a virtual path is obtained based on historical operation data, and the virtual path is used to replace the main navigation algorithm to control the inspection robot's navigation, further comprising:

[0037] Identify risk areas based on historical operating data;

[0038] A virtual path is used instead of the main navigation algorithm to control the inspection robot's navigation, and the inspection robot is navigated to avoid risk areas when it runs into them.

[0039] In a preferred embodiment, identifying risk areas based on historical operating data includes:

[0040] Based on historical operating data, the tire friction coefficient, wheel speed PID adjustment record, IMU angular velocity change rate and wheel speed encoder speed difference are obtained;

[0041] The risk area is obtained based on the position coordinates corresponding to the tire friction coefficient, wheel speed PID adjustment record, IMU angular velocity change rate, and wheel speed encoder speed difference.

[0042] The present invention also provides a navigation control method for an inspection robot, comprising:

[0043] Navigation control of the inspection robot based on the main navigation algorithm;

[0044] Circularly store the historical operation data of the inspection robot within a preset time period;

[0045] Obtain the real-time operation data of the inspection robot, and perform operation status diagnosis based on the real-time operation data to obtain the operation status;

[0046] When the operating state is abnormal, a virtual path is obtained according to the historical operating data, and the virtual path is used to replace the main navigation algorithm to navigate and control the inspection robot.

[0047] The beneficial effects of adopting the above embodiment are:

[0048] The present invention provides a patrol robot navigation control system and method, in which a main navigation control module is used to perform navigation control on the patrol robot based on a main navigation algorithm, a historical data storage module is used to cyclically store historical operation data of the patrol robot within a preset time period, an operation status diagnosis module is used to obtain real-time operation data of the patrol robot, and perform operation status diagnosis based on the real-time operation data to obtain the operation status, and an emergency navigation control module is used to obtain a virtual path based on the historical operation data when the operation status is abnormal, and use the virtual path to replace the main navigation algorithm to perform navigation control on the patrol robot. The present invention uses the main navigation control module to navigate normally based on the main navigation algorithm, ensuring that the robot performs inspection tasks efficiently and stably under normal conditions, and uses the operation status diagnosis module to timely and accurately detect abnormal conditions of the robot, ensuring the reliability and safety of the entire navigation control process, and uses the historical data storage module to cyclically store historical operation data within a preset time period. When the inspection robot is in an abnormal state, the emergency navigation control module can use these historical operation data to obtain a virtual path to replace the main navigation algorithm for navigation control. This allows the robot to continue working when encountering sudden abnormal conditions such as wireless interruption, sensor failure, road slippage, etc., avoiding high equipment downtime losses caused by downtime waiting for manual intervention, greatly improving the continuity and efficiency of the inspection work, and solving the problem in the prior art that the inspection robot can only stop and wait in a fault scenario. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is a system architecture diagram of the inspection robot navigation control system provided by the present invention;

[0050] Figure 2 A diagram showing the steps of the inspection robot navigation control method provided by the present invention;

[0051] Figure 3 for Figure 2 Specific step diagram of step S204;

[0052] Figure 4 for Figure 3 Specific step diagram of step S302 in FIG. DETAILED DESCRIPTION

[0053] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0054] Combine Figure 1 As shown, a specific embodiment of the present invention discloses a navigation control system for an inspection robot, comprising:

[0055] The main navigation control module 110 is used to perform navigation control on the inspection robot based on the main navigation algorithm;

[0056] The historical data storage module 120 is used to cyclically store the historical operation data of the inspection robot within a preset time period;

[0057] The operation status diagnosis module 130 is used to obtain the real-time operation data of the inspection robot and perform operation status diagnosis based on the real-time operation data to obtain the operation status;

[0058] The emergency navigation control module 140 is used to obtain a virtual path based on historical operation data when the operation state is abnormal, and use the virtual path to replace the main navigation algorithm to perform navigation control on the inspection robot.

[0059] Combine Figure 2 As shown, the present invention also provides a navigation control method for an inspection robot, comprising the following steps:

[0060] S201, performing navigation control on the inspection robot based on the main navigation algorithm;

[0061] S202, cyclically storing historical operation data of the inspection robot within a preset time period;

[0062] S203, obtaining real-time operating data of the inspection robot, and performing operating status diagnosis based on the real-time operating data to obtain the operating status;

[0063] S204: When the operation state is abnormal, a virtual path is obtained according to historical operation data, and the virtual path is used to replace the main navigation algorithm to perform navigation control on the inspection robot.

[0064] In the above process, the primary navigation algorithm refers to any existing algorithm capable of navigating and controlling the inspection robot under normal conditions, such as LiDAR navigation control methods, camera-based navigation, and UWB (ultra-wideband) positioning. The specific duration of the preset time period can be flexibly set based on actual conditions and is generally 72 hours. Historical operation data and real-time operation data refer to data collected or generated by the inspection robot during operation, such as the robot's position coordinates, movement speed, acceleration, attitude angle, detection values of various sensors (including but not limited to temperature sensors, humidity sensors, gas concentration sensors, and visual sensors), motor current and voltage data, joint motion angles, and path deviation information fed back by the navigation system. It is understood that the operating state includes both the state of the inspection robot itself (such as sensor failure) and the state of the inspection robot's environment (such as the presence of obstacles on the road). Abnormal state refers to any of the above-mentioned anomalies that can affect the navigation control of the primary navigation algorithm, such as electromagnetic interference, dust obstructing vision, fire smoke affecting landmark recognition, damaged navigation-related sensors, and wireless network interruption.

[0065] Understandably, the working environment of an industrial plant floor presents significant characteristics. For one thing, the tasks performed by inspection robots are highly repetitive. Typically, the robots follow pre-set inspection routes and procedures, performing status checks and data collection on various equipment within the plant floor at fixed intervals. This highly repetitive operating pattern provides an excellent foundation for data accumulation and path planning. Furthermore, the environmental conditions within an industrial plant floor are relatively stable. Compared to complex outdoor environments or specialized industrial scenarios, environmental conditions within a plant floor tend to remain relatively stable over time. Factors such as temperature, humidity, and layout of the plant floor typically do not fluctuate drastically over short periods of time. This ensures that the historical operating data accumulated by the inspection robots over long periods of time effectively reflects their movement and operating characteristics under normal conditions.

[0066] Based on these features, the present invention fully leverages the value of historical operation data. If an inspection robot encounters an emergency during a mission, such as a wireless outage, sensor failure, or slippery road conditions, rendering the primary navigation algorithm ineffective, the system can quickly invoke a stable and highly reliable virtual path generated based on historical operation data to take over navigation for the inspection robot in the event of an emergency.

[0067] It is understandable that in practice, the specific status diagnosis method is different depending on the type of historical operation data or actual operation data. However, no matter which diagnosis method is used, it can be implemented using existing technology, so this article will not explain it in detail.

[0068] Further, combined Figure 3 As shown, in a preferred embodiment, the above step S204, when the operating state is abnormal, obtains a virtual path based on historical operating data, and uses the virtual path to replace the main navigation algorithm to navigate and control the inspection robot, specifically including:

[0069] S301, when the operation state is abnormal, analyzing historical operation data to obtain a virtual path;

[0070] S302. Predicting the data that can be obtained by the inspection robot in an abnormal state based on historical operation data to obtain predicted data;

[0071] S303, controlling the inspection robot to move based on the virtual path, and simultaneously collecting available data of the inspection robot in an abnormal state to obtain real-time data;

[0072] S304: Perform navigation correction on the inspection robot based on the difference between the predicted data and the real-time data.

[0073] In the above process, the "acquired data" refers to data that the inspection robot can still normally collect under abnormal conditions. For example, when the point cloud radar is damaged, the data collected by other sensors in the inspection robot that have not failed. Based on the navigation control based on the virtual path, the above process further explores the potential of historical operation data. It uses historical operation data to predict predicted data, and compares the predicted data with the real-time data as accurate ideal state data to achieve feedback correction of navigation. This dynamic adjustment mechanism enables the robot to closely adapt to changes in the actual operating environment, promptly correct possible deviations, ensure the accuracy and effectiveness of navigation, greatly enhance the inspection robot's adaptability and autonomous error correction capabilities in complex and unexpected environments, and effectively ensure the smooth completion of inspection tasks. It is understandable that precisely because the inspection work in the factory workshop is highly repetitive and the environmental changes are small, the predicted data in this embodiment can be used as a benchmark for comparison with real-time data.

[0074] Specifically, in a preferred embodiment, the historical operation data includes historical path coordinates and anchor point positions; the above step S301, when the operation state is abnormal, analyzing the historical operation data to obtain a virtual path, specifically includes:

[0075] Divide the historical path coordinates based on the anchor point positions to obtain multiple sets of periodic path coordinates;

[0076] Take the average value of multiple sets of periodic path coordinates to obtain the initial path coordinates;

[0077] The path formed by the initial path coordinates is smoothed to obtain a virtual path.

[0078] In the above process, the anchor point location is a manually designated location for the inspection robot to perform periodic identification, such as a location on the production line, the anchor robot's standby position, and so on. In this embodiment, the period between two runs of the inspection robot to the anchor point location constitutes an operation cycle, and the inspection robot performs repetitive work within each operation cycle. By setting the anchor point location, this embodiment accurately captures the periodic patterns in the robot's operation. Since industrial plant and workshop inspection work is often repetitive, this approach can well adapt to this characteristic and lay the foundation for generating a reliable virtual path. Averaging multiple sets of periodic path coordinates to obtain initial path coordinates effectively reduces the potential error introduced by single-run data, making the resulting initial path coordinates more representative and stable, and improving the accuracy of the virtual path. Smoothing the path constructed from the initial path coordinates to obtain the virtual path further optimizes the path, avoiding path abrupt changes or discontinuities caused by data fluctuations. This makes the virtual path smoother and more natural, more consistent with the robot's kinematic characteristics, and helps the robot continue its inspection mission smoothly and efficiently according to the virtual path even in abnormal conditions.

[0079] Further, combined Figure 4 As shown, the above step S302 predicts the data that can be obtained by the inspection robot in an abnormal state based on the historical operation data to obtain the predicted data, which specifically includes:

[0080] S401, obtaining a data collection period based on the anchor point position;

[0081] S402, dividing the historical operation data based on the data collection period to obtain multiple groups of periodic data;

[0082] S403, performing prediction based on multiple sets of periodic data to obtain complete prediction data;

[0083] S404: Filter out data of the same type as the data obtainable by the inspection robot in an abnormal state from the complete prediction data to obtain prediction data.

[0084] The above process also utilizes the setting of the anchor point position and the periodicity of the inspection robot's operation, and can accurately grasp the data collection rhythm of the inspection robot during normal operation. This rhythm reflects the working rules of the robot under stable working conditions and can effectively mine the periodic characteristics and laws in the data. These laws are the inherent logic of the robot's operation, which helps to more accurately grasp the robot's operating status and data change trends at different stages to obtain more accurate prediction data. On this basis, multiple sets of periodic data are predicted to obtain complete prediction data, which fully integrates the information in the historical data and uses the periodicity and correlation of the data to make reasonable inferences, providing a more comprehensive reference for situations under abnormal conditions. Most importantly, because the failed sensors in practice may be uncertain, the types of data that need to be predicted are also uncertain. Therefore, this embodiment can effectively improve versatility by first performing a complete prediction and then screening.

[0085] Specifically, in a preferred embodiment, the above step S403, performing prediction based on multiple sets of periodic data to obtain complete prediction data, includes:

[0086] Evaluate the similarity of multiple sets of periodic data and obtain similar eigenvalues;

[0087] Fit multiple sets of periodic data to obtain fitted prediction data;

[0088] Input multiple sets of periodic data into a preset prediction model to obtain model prediction data;

[0089] According to similar feature values, the fitted prediction data and the model prediction data are fused to obtain the complete prediction data.

[0090] Fitting refers to the method of using functions to approximate the relationship between data. The advantage of fitting prediction is that it can intuitively reflect the data trend, the calculation is relatively simple, and the interpretability is strong, but its generalization ability is limited and it is difficult to capture complex relationships. Model prediction is just the opposite. It can better capture the complex relationship between data, but it may suffer from overfitting. Therefore, this embodiment adopts a redundant verification method to achieve accurate prediction. First, the commonalities between different periods are analyzed by similar feature values. If the similarity is low, it means that the data regularity of different periods is poor. At this time, it is necessary to obtain more accurate results through model prediction. If the similarity is high, it means that the data regularity of different periods is good. At this time, it is more appropriate to use simple fitting prediction.

[0091] It is understandable that in the above process, fitting prediction (such as linear fitting, polynomial fitting, etc.) and model prediction (such as prediction through a neural network model) are both existing technologies that can be understood by those skilled in the art, so they will not be explained in detail in this article.

[0092] In addition, the above steps: according to similar feature values, the fitting prediction data and the model prediction data are integrated to obtain the complete prediction data. The specific implementation method can also be flexibly designed. For example, when the similar feature value exceeds a certain threshold, the fitting prediction data is used as the complete prediction data; for example, according to the similar feature value, the fitting prediction data and the model prediction data are integrated using dynamically adjusted weights.

[0093] Furthermore, in a preferred embodiment, the above step S304, performing navigation correction on the inspection robot based on the difference between the predicted data and the real-time data, specifically includes:

[0094] Calculate the difference between the predicted data and the real-time data;

[0095] The navigation speed of the inspection robot is corrected according to the difference value, wherein the difference value is inversely proportional to the navigation speed.

[0096] The above process corrects the navigation speed. Obviously, when the difference value is larger, the running speed of the inspection robot should be lower to ensure safety.

[0097] Specifically, in a preferred embodiment, the above step of calculating the difference between the predicted data and the real-time data specifically includes:

[0098] Get the preset priority of each forecast data;

[0099] The difference between the predicted data and the real-time data is calculated, and the difference is weighted and summed with the preset priority as the weight to obtain the difference value.

[0100] This embodiment, by setting a preset priority for each type of prediction data, fully considers the varying importance of different prediction data in the inspection robot's navigation correction process. Different prediction data may have varying degrees of impact on the robot's operating status and navigation decisions. Assigning them appropriate priorities can more accurately reflect actual needs. A weighted calculation method can highlight the impact of important prediction data and weaken the role of less important data, thereby more accurately measuring the degree of deviation between actual operation and expected conditions. Making navigation corrections based on such precise difference values enables the inspection robot to more precisely adjust its navigation strategy, improving navigation accuracy and stability.

[0101] For example, the four types of data collected by IMU, wheel speed encoder, lidar, and temperature sensor are configured with priorities of 0.4, 0.3, 0.2, and 0.1, respectively.

[0102] Furthermore, in a preferred embodiment, the above step S204, when the operating state is abnormal, obtains a virtual path based on historical operating data, and uses the virtual path instead of the main navigation algorithm to navigate and control the inspection robot, further comprising:

[0103] Identify risk areas based on historical operating data;

[0104] A virtual path is used instead of the main navigation algorithm to control the inspection robot's navigation, and the inspection robot is navigated to avoid risk areas when it runs into them.

[0105] The navigation avoidance in the above process refers to any avoidance behavior including deceleration, detour, etc. This embodiment further explores the value of historical operation data, uses historical operation data to identify risk areas, and takes corresponding measures to improve safety.

[0106] Furthermore, in a preferred embodiment, the above step of identifying risk areas based on historical operating data specifically includes:

[0107] Based on historical operating data, the tire friction coefficient, wheel speed PID adjustment record, IMU angular velocity change rate and wheel speed encoder speed difference are obtained;

[0108] The risk area is obtained based on the position coordinates corresponding to the tire friction coefficient, wheel speed PID adjustment record, IMU angular velocity change rate, and wheel speed encoder speed difference.

[0109] This embodiment mainly identifies areas at risk of ground slippage (such as areas prone to dust, water stains, and oil films). It uses the comparison of the tire friction coefficient with a pre-stored coefficient library, the wheel speed PID adjustment records, the IMU angular velocity change rate, and the wheel speed encoder speed difference to identify areas prone to slippage, and perform deceleration control to avoid accidents.

[0110] The present invention provides a patrol robot navigation control system and method, in which a main navigation control module is used to perform navigation control on the patrol robot based on a main navigation algorithm, a historical data storage module is used to cyclically store historical operation data of the patrol robot within a preset time period, an operation status diagnosis module is used to obtain real-time operation data of the patrol robot, and perform operation status diagnosis based on the real-time operation data to obtain the operation status, and an emergency navigation control module is used to obtain a virtual path based on the historical operation data when the operation status is abnormal, and use the virtual path to replace the main navigation algorithm to perform navigation control on the patrol robot. The present invention uses the main navigation control module to navigate normally based on the main navigation algorithm, ensuring that the robot performs inspection tasks efficiently and stably under normal conditions, and uses the operation status diagnosis module to timely and accurately detect abnormal conditions of the robot, ensuring the reliability and safety of the entire navigation control process, and uses the historical data storage module to cyclically store historical operation data within a preset time period. When the inspection robot is in an abnormal state, the emergency navigation control module can use these historical operation data to obtain a virtual path to replace the main navigation algorithm for navigation control. This allows the robot to continue working when encountering sudden abnormal conditions such as wireless interruption, sensor failure, road slippage, etc., avoiding high equipment downtime losses caused by downtime waiting for manual intervention, greatly improving the continuity and efficiency of the inspection work, and solving the problem in the prior art that the inspection robot can only stop and wait in a fault scenario.

[0111] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same and similar parts between the various embodiments can be referenced to each other.

[0112] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A navigation control system for an inspection robot, characterized in that: include: The main navigation control module is used to control the inspection robot's navigation based on the main navigation algorithm; A historical data storage module is used to cyclically store the historical operation data of the inspection robot within a preset time period; The operation status diagnosis module is used to obtain the real-time operation data of the inspection robot and perform operation status diagnosis based on the real-time operation data to obtain the operation status; The emergency navigation control module is used to obtain a virtual path based on historical operation data when the operating state is abnormal, and use the virtual path to replace the main navigation algorithm to navigate and control the inspection robot.

2. The inspection robot navigation control system according to claim 1, characterized in that: When the operation status is abnormal, a virtual path is obtained based on historical operation data, and the virtual path is used to replace the main navigation algorithm to navigate and control the inspection robot, including: When the operation status is abnormal, the historical operation data is analyzed to obtain a virtual path; Based on historical operation data, the available data of the inspection robot in abnormal state is predicted to obtain predicted data; Control the inspection robot's movement based on the virtual path, and collect the available data of the inspection robot in abnormal state to obtain real-time data; Based on the difference between predicted data and real-time data, the inspection robot's navigation is corrected.

3. The inspection robot navigation control system according to claim 2, characterized in that: Historical operation data includes historical path coordinates and anchor point locations. When the operation status is abnormal, the virtual path is obtained by analyzing the historical operation data, which also includes: Divide the historical path coordinates based on the anchor point positions to obtain multiple sets of periodic path coordinates; Take the average value of multiple sets of periodic path coordinates to obtain the initial path coordinates; The path formed by the initial path coordinates is smoothed to obtain a virtual path.

4. The inspection robot navigation control system according to claim 3, characterized in that: Based on historical operating data, the available data of the inspection robot in abnormal status is predicted to obtain the predicted data, including: Based on the anchor point position, the data collection period is obtained; Divide historical operation data based on data collection cycle to obtain multiple groups of periodic data; Make predictions based on multiple sets of periodic data to obtain complete prediction data; From the complete prediction data, data of the same type as the data that can be obtained by the inspection robot in an abnormal state is filtered out to obtain the prediction data.

5. The inspection robot navigation control system according to claim 4, characterized in that: Based on multiple sets of periodic data, we can obtain complete forecast data, including: Evaluate the similarity of multiple sets of periodic data and obtain similar eigenvalues; Fit multiple sets of periodic data to obtain fitted prediction data; Input multiple sets of periodic data into a preset prediction model to obtain model prediction data; According to similar feature values, the fitted prediction data and the model prediction data are fused to obtain the complete prediction data.

6. The inspection robot navigation control system according to claim 2, characterized in that: Based on the difference between the predicted data and the real-time data, the inspection robot makes navigation corrections, including: Calculate the difference between the predicted data and the real-time data; The navigation speed of the inspection robot is corrected according to the difference value, wherein the difference value is inversely proportional to the navigation speed.

7. The inspection robot navigation control system according to claim 6, characterized in that: Calculate the difference between the forecast data and the real-time data, including: Get the preset priority of each forecast data; The difference between the predicted data and the real-time data is calculated, and the difference is weighted and summed with the preset priority as the weight to obtain the difference value.

8. The inspection robot navigation control system according to claim 2, characterized in that: When the operation state is abnormal, a virtual path is obtained based on historical operation data, and the virtual path is used to replace the main navigation algorithm to navigate and control the inspection robot, which also includes: Identify risk areas based on historical operating data; A virtual path is used instead of the main navigation algorithm to control the inspection robot's navigation, and the inspection robot is navigated to avoid risk areas when it runs into them.

9. The inspection robot navigation control system according to claim 8, characterized in that: Based on historical operating data, risk areas are identified, including: Based on historical operating data, the tire friction coefficient, wheel speed PID adjustment record, IMU angular velocity change rate and wheel speed encoder speed difference are obtained; The risk area is obtained based on the position coordinates corresponding to the tire friction coefficient, wheel speed PID adjustment record, IMU angular velocity change rate, and wheel speed encoder speed difference.

10. A navigation control method for an inspection robot, characterized in that: include: Navigation control of the inspection robot based on the main navigation algorithm; Circularly store the historical operation data of the inspection robot within a preset time period; Obtain the real-time operation data of the inspection robot, and perform operation status diagnosis based on the real-time operation data to obtain the operation status; When the operating state is abnormal, a virtual path is obtained according to the historical operating data, and the virtual path is used to replace the main navigation algorithm to navigate and control the inspection robot.

Images