Industrial robot obstacle avoidance control method based on sensor data analysis

Abstract

The invention relates to the technical field of robot obstacle avoidance, and particularly discloses an industrial robot obstacle avoidance control method based on sensor data analysis, and the method comprises the steps: capturing image data in a current advancing route in real time through a visual sensor in an industrial robot; the risk coefficients of different time points of the current advancing route are calculated, whether the industrial robot can continue to move on the current advancing route or not can be judged, and when the intelligent analysis module judges that the industrial robot cannot continue to move on the current advancing route, the risk of the current advancing route is continuously monitored, so that the safety of the industrial robot is improved. According to the obstacle avoidance method, the risk coefficient variable quantity of the current advancing route within a period of time is further analyzed, the position change of the obstacle within a period of time can be judged according to the data, the obstacle avoidance mode of the robot can be optimized through the obstacle avoidance method, and then the robot is helped to more accurately decide whether the advancing route needs to be changed or not.

Description

Technical Field

[0001] The present invention relates to the technical field of robot obstacle avoidance, and specifically to an industrial robot obstacle avoidance control method based on sensor data analysis. Background Art

[0002] An industrial robot is a mechanical device widely used in the field of industrial automation. During operation, the robot moves along a pre-set route, replacing humans to perform some monotonous, frequent, and repetitive long-term operations.

[0003] Since the movement of the robot is based on a pre-set route, in order to ensure the normal use of the robot, an obstacle avoidance system is usually set in the robot. The common obstacle avoidance method of the obstacle avoidance system is based on multi-sensor fusion. Data information of obstacles is collected through multiple types of sensors, and data from different sensors are fused to obtain more comprehensive and accurate obstacle information. Finally, by analyzing the obstacle information, it is decided whether to adjust the route, thereby realizing the obstacle avoidance function.

[0004] The common obstacle avoidance method in the prior art is to analyze the obstacle information to decide whether to adjust the route. However, since there will be a slowdown in movement efficiency when adjusting to other alternative routes and there may also be situations where the road is blocked by obstacles, and due to the diversity of obstacles, some obstacles may stay temporarily on the robot's travel route. Therefore, it is necessary to optimize the obstacle avoidance method of the robot to ensure that the robot selects a suitable obstacle avoidance route and improves work efficiency. Summary of the Invention

[0005] The purpose of the present invention is to provide an industrial robot obstacle avoidance control method based on sensor data analysis, and solve the following technical problems:

[0006] How to optimize the obstacle avoidance method of the robot.

[0007] The purpose of the present invention can be achieved by the following technical solutions:

[0008] An industrial robot obstacle avoidance control method based on sensor data analysis, the method includes:

[0009] S1: Real-time capture image data in the current travel route through a vision sensor in the industrial robot;

[0010] S2: Perform image quality analysis on the collected image data through an intelligent analysis module in the industrial robot to determine whether the quality of the current image data is qualified. If so, proceed to step S3; otherwise, reshoot;

[0011] S3: When it is determined that the quality of the current image data is qualified, the intelligent analysis module extracts the required features in the image data and calculates the risk coefficients at different time points of the current travel route in combination with this data;

[0012] S4: The intelligent analysis module analyzes in combination with the risk coefficient of the current travel route and makes a judgment on whether the industrial robot can continue to move on the current travel route. If so, it continues to move and continuously monitors. Otherwise, it proceeds to step S5;

[0013] S5: When the intelligent analysis module determines that it cannot continue to move on the current travel route, the robot stops moving and further analyzes the change amount of the risk coefficient of the current travel route over a period of time by continuously monitoring the risk of the current travel route;

[0014] S6: The intelligent analysis module analyzes in combination with the change amount of the risk change coefficient of the current travel route and makes a decision on the next action of the robot.

[0015] Further, the process of performing image quality analysis on the collected image data in S2 includes:

[0016] By the formula The quality impact coefficient of the a-th captured image is calculated ;

[0017] where a is any piece of image data captured by the visual sensor, is the number of noise points in the a-th piece of image data, is the preset number of noise points, is the number of spots in the a-th piece of image data, is the preset number of spots, is the damaged area of the image in the a-th piece of image data, is the preset damaged area of the image, is the error impact coefficient, set by empirical fitting, The number of pixel points in the a-th piece of image data, is the preset number of pixel points, is the standard value of, is the clarity of the a-th piece of image data, is the preset clarity, is the standard value of, is a defined function. If , then let , otherwise, let .

[0018] Further, the process of performing image quality analysis on the collected image data in S2 further includes:

[0019] By comparing the quality influence coefficient of the a-th captured image with a preset image quality influence coefficient threshold ;

[0020] If , it is determined that the quality of the current image data is poor, and the image data is continuously captured;

[0021] If , it is determined that the quality of the current image data is high, and the intelligent analysis module extracts the required features from the image data.

[0022] Further, the calculation process in S3 includes:

[0023] By using the formula calculate the risk coefficient at the i-th detection of the current travel route ;

[0024] where i is any risk coefficient detection, is the distance between the robot and the obstacle at the i-th detection, is the moving speed of the robot at the i-th detection, is the data processing time of the robot, is the warning distance between the robot and the obstacle, is the number of obstacles at the i-th detection, s is any obstacle, is the size of the s-th obstacle at the i-th detection, is the standard value of, is the position influence coefficient of the s-th obstacle at the i-th detection, which is set by empirical fitting.

[0025] Further, the analysis process in S4 includes:

[0026] By comparing the risk coefficient at the i-th time point of the current travel route with a preset risk coefficient threshold interval ;

[0027] If , it is determined that the risk coefficient at this time point of the current travel route is low, and the robot can continue to move on the current travel route;

[0028] If , it is determined that the risk coefficient at this time point of the current travel route is high, and the intelligent analysis module issues an instruction to the robot to stop moving and continuously monitors the risk coefficient of the current travel route.

[0029] Further, the process of analyzing the change amount of the risk coefficient of the current travel route in S5 includes:

[0030] By continuously monitoring the risk coefficient of the current travel route by the robot, a change curve of the risk coefficient of the current travel route is established ;

[0031] And through the formula Calculate the change amount of the risk coefficient of the current travel route over a period of time ;

[0032] Wherein, n is the total number of detections of the risk coefficient during the continuous monitoring after the robot stops moving, is the start time point during the continuous monitoring after the robot stops moving, is for all the average value of, is the end time during the continuous monitoring after the robot stops moving, is the proportionality coefficient, which is set by empirical fitting.

[0033] Further, the process of analyzing the change amount of the risk coefficient of the current travel route in S5 further includes:

[0034] By comparing the change amount of the risk coefficient of the current travel route over a period of time with the preset change amount threshold for comparison;

[0035] If , it is determined that the obstacle risk coefficient in the current travel route is gradually decreasing;

[0036] If , it is determined that the obstacle risk coefficient in the current travel route is gradually increasing or remains unchanged.

[0037] Further, the decision-making process in S6 includes:

[0038] When it is determined that the obstacle risk coefficient in the current travel route is gradually decreasing, it indicates that the obstacle is moving away from the robot, and the robot can be controlled to continue moving through the intelligent analysis module;

[0039] When it is determined that the obstacle risk coefficient in the current travel route is gradually increasing or remains unchanged, it indicates that the obstacle has not moved or is moving towards the robot, and the travel route is adjusted through the intelligent analysis module.

[0040] Advantages of the present invention:

[0041] (1) By calculating the risk coefficients at different time points of the current travel route, the present invention can determine whether the industrial robot can continue to move on the current travel route. When the intelligent analysis module determines that it cannot continue to move on the current travel route, by continuously monitoring the risk of the current travel route, further analysis is carried out on the change amount of the risk coefficient of the current travel route within a period of time. According to this data, the position change of the obstacle within a period of time can be judged. Through this obstacle avoidance method, the obstacle avoidance method of the robot can be optimized, thereby helping the robot make a more accurate decision on whether to change the travel route.

[0042] (2) By comparing the quality influence coefficient of the a-th captured image with the preset image quality influence coefficient threshold in the present invention, the quality of the currently captured image can be analyzed. According to the quality of the image, it can be decided whether to continue image capture or extract the required features in the image data through the intelligent analysis module, so that the intelligent analysis module extracts features based on high-quality image data, thereby improving the accuracy and reliability of feature extraction, and further providing accurate data support for the subsequent analysis of obstacles.

[0043] (3) By comparing the risk coefficient at the i-th time point of the current travel route with the preset risk coefficient threshold range in the present invention, through this comparison method, an accurate judgment can be made on the level of the risk coefficient at this time point of the current travel route, that is, it can be decided whether to continue to move on the current route according to the judgment result. And when it is judged that it cannot continue to move, by continuously monitoring the risk coefficient of the current travel route for further analysis. By setting like this, the optimization of the obstacle avoidance method of the robot can be realized to ensure that the robot selects a suitable obstacle avoidance route and improves work efficiency.

[0044] (4) By comparing the change amount of the risk coefficient of the current travel route within a period of time with the preset change amount threshold in the present invention, through this comparison method, a judgment can be made on the change of the obstacle risk coefficient in the current travel route, thereby providing accurate data for the subsequent decision-making of the robot, ensuring that the robot selects a suitable obstacle avoidance route and improving work efficiency.

[0045] (5) By judging the change of the risk coefficient of obstacles in the current traveling route, the present invention can analyze the behavior of obstacles during the monitoring process. When it is judged that the obstacle is moving away from the robot, it means that the obstacle stays temporarily on the traveling route of the robot, indicating that the obstacle will leave the traveling route of the robot subsequently. According to this analysis result, it can help the robot avoid blindly adjusting the traveling route, which may slow down the moving efficiency, and can also avoid the situation that the alternative route is blocked by the obstacle after adjusting the traveling route, resulting in immobility, thereby realizing the optimization of the obstacle avoidance method of the robot. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The present invention will be further described below with reference to the drawings.

[0047] Figure 1 It is a flowchart of an obstacle avoidance control method for an industrial robot based on sensor data analysis in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0048] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0049] Please refer to Figure 1 As shown, in one embodiment, the present application provides an obstacle avoidance control method for an industrial robot based on sensor data analysis. The method includes:

[0050] S1: Real-time capture image data in the current traveling route through a vision sensor in the industrial robot;

[0051] S2: Analyze the image quality of the collected image data through an intelligent analysis module in the industrial robot, judge whether the quality of the current image data is qualified. If it is, proceed to step S3; otherwise, reshoot;

[0052] S3: When it is judged that the quality of the current image data is qualified, the intelligent analysis module extracts the required features in the image data and calculates the risk coefficient at different time points of the current traveling route in combination with the data;

[0053] S4: Analyze through the intelligent analysis module in combination with the risk coefficient of the current traveling route, and judge whether the industrial robot can continue to move on the current traveling route. If it is, continue to move and continuously monitor; otherwise, proceed to step S5;

[0054] S5: When the intelligent analysis module determines that it is unable to continue moving on the current travel route, the robot stops moving and further analyzes the change amount of the risk coefficient on the current travel route for a period of time by continuously monitoring the risk of the current travel route;

[0055] S6: Analyze through the intelligent analysis module in combination with the change amount of the risk change coefficient of the current travel route, and make a decision on the next action of the robot;

[0056] Through the above technical solution, this embodiment provides an obstacle avoidance control method for an industrial robot based on sensor data analysis. First, the visual sensor in the industrial robot is used to capture the image data in the current travel route in real time. Then, the intelligent analysis module in the industrial robot performs image quality analysis on the collected image data to determine whether the quality of the current image data is qualified. When it is determined that the quality of the current image data is qualified, the intelligent analysis module extracts the required features in the image data and calculates the risk coefficients at different time points of the current travel route in combination with this data. Then, the intelligent analysis module analyzes in combination with the risk coefficient of the current travel route and determines whether the industrial robot can continue to move on the current travel route. When the intelligent analysis module determines that it is unable to continue moving on the current travel route, the robot stops moving and further analyzes the change amount of the risk coefficient on the current travel route for a period of time by continuously monitoring the risk of the current travel route. Finally, the intelligent analysis module analyzes in combination with the change amount of the risk change coefficient of the current travel route and makes a decision on the next action of the robot;

[0057] Through the above technical solution, this embodiment can calculate the risk coefficients at different time points of the current travel route, determine whether the industrial robot can continue to move on the current travel route, and when the intelligent analysis module determines that it is unable to continue moving on the current travel route, further analyze the change amount of the risk coefficient on the current travel route for a period of time by continuously monitoring the risk of the current travel route. According to this data, the position change of the obstacle within a period of time can be judged. Through this obstacle avoidance method, the obstacle avoidance method of the robot can be optimized, thereby helping the robot make a more accurate decision on whether to change the travel route.

[0058] The process of performing image quality analysis on the collected image data in S2 includes:

[0059] Obtain the quality influence coefficient of the a-th captured image through the formula ; ;

[0060] where a is any piece of image data captured by the visual sensor, is the number of noise points in the a-th image data, is the preset number of noise points, is the number of spots in the a-th image data, is the preset number of spots, is the damaged area of the a-th image data, is the preset damaged area of the image, is the error influence coefficient, set by empirical fitting, the number of pixel points in the a-th image data, is the preset number of pixel points, is the standard value of, and the above standard value is selected and set according to the allowable error in the empirical data, is the clarity of the a-th image data, is the preset clarity, is the standard value of, and the above standard value is selected and set according to the allowable error in the empirical data, is a defined function. If , then let , otherwise, let ;

[0061] Through the above technical solution, this example provides the quality influence coefficient of the a-th captured image, which can be obtained by the formula Obviously, when the number of noise points and spots in the a-th image data is more, the damaged area is larger, the number of pixel points is less, and the clarity is worse, then the quality influence coefficient of the captured image is larger, indicating that the quality of the current image is poor. When performing feature extraction based on this image, there will be errors and the image capture needs to be redone. On the contrary, when the number of noise points and spots in the a-th image data is less, the damaged area is smaller, the number of pixel points is more, and the clarity is better, then the quality influence coefficient of the captured image is smaller, indicating that the quality of the current image is higher. When performing feature extraction based on this image, the error is smaller;

[0062] Through this calculation method, the quality of the image can be analyzed according to the calculation result, so as to ensure that the subsequent robot can perform feature extraction based on high-quality images, providing accurate data support for the subsequent analysis of obstacles.

[0063] The process of performing image quality analysis on the collected image data in S2 further includes:

[0064] By comparing the quality influence coefficient of the a-th captured image with the preset image quality influence coefficient threshold ;

[0065] If , it is determined that the quality of the current image data is poor, and the image data is continuously captured;

[0066] If , it is determined that the quality of the current image data is high, and the required features in the image data are extracted through the intelligent analysis module;

[0067] Through the above technical solution, in this example, by comparing the quality influence coefficient of the a-th captured image with the preset image quality influence coefficient threshold , the quality of the currently captured image can be analyzed. According to the quality of the image, it can be decided whether to continue image capture or extract the required features in the image data through the intelligent analysis module, so that the intelligent analysis module extracts features based on high-quality image data, thereby improving the accuracy and reliability of feature extraction, and further providing accurate data support for subsequent obstacle analysis.

[0068] The calculation process in S3 includes:

[0069] The risk coefficient at the i-th detection of the current travel route is calculated through the formula ;

[0070] where i is any risk coefficient detection, is the distance between the robot and the obstacle at the i-th detection, is the moving speed of the robot at the i-th detection, is the data processing time of the robot, is the warning distance between the robot and the obstacle, is the number of obstacles at the i-th detection, s is any obstacle, is the size of the s-th obstacle at the i-th detection, is 's standard value, and the above standard value is selected and set according to the allowable error in empirical data, is the position influence coefficient of the s-th obstacle at the i-th detection, which is set by empirical fitting;

[0071] Through the above technical solution, this example provides the risk coefficient at the i-th detection of the current travel route, which can be calculated through the formula . Obviously, when the distance between the robot and the obstacle at the i-th detection is closer, the moving speed of the robot at the i-th detection is faster, the data processing speed of the robot is slower, and the size and position influence coefficient of the s-th obstacle at the i-th detection are larger, then the risk coefficient The greater it is, it indicates that at this time point, the obstacle has a greater impact on the robot, and there may be a situation where the robot collides with the obstacle if it continues to move. On the contrary, when the distance between the robot and the obstacle at the i-th detection is farther, the moving speed of the robot at the i-th detection is slower, the data processing speed of the robot is faster, and the size and position influence coefficient of the s-th obstacle at the i-th detection is smaller, then the risk coefficient at the i-th detection of the current travel route is smaller, which indicates that at this time point, the obstacle has a smaller impact on the robot, and there will be no situation where the robot collides with the obstacle when it continues to move. And when there is no obstacle, the risk coefficient at the i-th detection of the current travel route is 0, which means that there is no risk of collision when the robot moves on the current route, and it can continue to move;

[0072] Through this calculation method, the calculation result can reflect the risk situation of the robot continuing to move on the current route, thereby providing data support for subsequent analysis of whether the travel route needs to be adjusted. The diversified data can optimize the obstacle avoidance method of the robot, so as to ensure that the robot selects a suitable obstacle avoidance route and improves work efficiency.

[0073] The analysis process in S4 includes:

[0074] By comparing the risk coefficient at the i-th time point of the current travel route with the preset risk coefficient threshold range for comparison;

[0075] If , it is determined that the risk coefficient at this time point of the current travel route is low, and the robot can continue to move on the current travel route;

[0076] If , it is determined that the risk coefficient at this time point of the current travel route is high, and the intelligent analysis module issues an instruction to the robot to stop moving and continuously monitors the risk coefficient of the current travel route;

[0077] Through the above technical solution, in this example, by comparing the risk coefficient at the i-th time point of the current travel route with the preset risk coefficient threshold range for comparison, through this comparison method, an accurate judgment can be made on the high or low of the risk coefficient at this time point of the current travel route, and it can be decided whether to continue moving on the current route according to the judgment result. And when it is judged that it is impossible to continue moving, by continuously monitoring the risk coefficient of the current travel route for further analysis, through such settings, the optimization of the obstacle avoidance method of the robot can be realized to ensure that the robot selects a suitable obstacle avoidance route and improves work efficiency.

[0078] The process of analyzing the change amount of the risk coefficient of the current travel route in S5 includes:

[0079] By continuously monitoring the risk coefficient of the current travel route by the robot, a change curve of the risk coefficient of the current travel route is established ;

[0080] And through the formula Calculate the change amount of the risk coefficient of the current travel route within a period of time ;

[0081] Among them, n is the total number of detections of the risk coefficient during the continuous monitoring after the robot stops moving, is the starting time point during the continuous monitoring after the robot stops moving, is for all average value, is the end time during the continuous monitoring after the robot stops moving, is the proportionality coefficient, which is set by empirical fitting;

[0082] Through the above technical solution, this example provides the change amount of the risk coefficient of the current travel route within a period of time , which can be calculated through the formula By calculating the change amount of the risk coefficient of the current travel route within a period of time , it can reflect the change of the risk coefficient of the current travel route within a period of time. According to the analysis of this data, it can be judged whether the obstacle in the current travel route stays on the travel route of the robot temporarily, that is, whether the obstacle will leave the travel route of the robot subsequently, so as to avoid the situation of blindly adjusting the travel route and causing the movement efficiency to slow down, and it can also avoid the situation that the alternative route is blocked by the obstacle after adjusting the travel route and resulting in immobility.

[0083] The process of analyzing the change amount of the risk coefficient of the current travel route in S5 further includes:

[0084] By comparing the change amount of the risk coefficient of the current travel route within a period of time with the preset change amount threshold ;

[0085] If , it is judged that the risk coefficient of the obstacle in the current travel route is gradually decreasing;

[0086] If , it is judged that the risk coefficient of the obstacle in the current travel route is gradually increasing or has not changed;

[0087] Through the above technical solution, in this embodiment, the change amount of the risk coefficient of the current travel route within a period of time is compared with a preset change amount threshold By this comparison method, it is possible to judge the change in the risk coefficient of the obstacle in the current travel route, so as to provide accurate data for subsequent decisions of the robot, ensure that the robot selects a suitable obstacle avoidance route, and improve work efficiency.

[0088] The decision-making process in S6 includes:

[0089] When it is judged that the risk coefficient of the obstacle in the current travel route is gradually decreasing, it means that the obstacle is moving away from the robot, and the robot can be controlled to continue moving through the intelligent analysis module;

[0090] When it is judged that the risk coefficient of the obstacle in the current travel route is gradually increasing or remains unchanged, it means that the obstacle has not moved or is moving towards the robot, and the travel route is adjusted through the intelligent analysis module;

[0091] Through the above technical solution, this embodiment provides a decision-making process for the robot. When it is judged that the risk coefficient of the obstacle in the current travel route is gradually decreasing, it means that the obstacle is moving away from the robot, and the robot can be controlled to continue moving through the intelligent analysis module. When it is judged that the risk coefficient of the obstacle in the current travel route is gradually increasing or remains unchanged, it means that the obstacle has not moved or is moving towards the robot, and the travel route is adjusted through the intelligent analysis module;

[0092] By setting like this, by judging the change in the risk coefficient of the obstacle in the current travel route, the behavior of the obstacle during the monitoring process can be analyzed. When it is judged that the obstacle is moving away from the robot, it means that the obstacle is temporarily staying on the robot's travel route, indicating that the obstacle will leave the robot's travel route subsequently. According to this analysis result, it can help the robot avoid the situation where the moving efficiency slows down due to blindly adjusting the travel route, and can also avoid the situation where the alternative route is blocked by the obstacle after adjusting the travel route and the robot cannot move, thereby realizing the optimization of the obstacle avoidance method for the robot.

[0093] The above has described a specific embodiment of the present invention in detail, but the content described is only a preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. All equivalent changes and improvements made according to the scope of the application of the present invention should still fall within the scope covered by the patent of the present invention.

Claims

1. An industrial robot obstacle avoidance control method based on sensor data analysis, characterized in that: The method comprises: S1: The visual sensor in the industrial robot captures the image data of the current route in real time; S2: Perform image quality analysis on the collected image data through the intelligent analysis module in the industrial robot to determine whether the quality of the current image data is qualified. If yes, proceed to step S3; otherwise, retake the image; S3: When the quality of the current image data is determined to be qualified, the intelligent analysis module extracts the required features from the image data and calculates the risk coefficients of the current route at different time points based on the data; S4: Analyze the risk factor of the current route through the intelligent analysis module, and determine whether the industrial robot can continue to move on the current route. If yes, continue to move and monitor continuously; otherwise, proceed to step S5; S5: When the intelligent analysis module determines that the robot cannot continue to move on the current route, the robot stops moving and continuously monitors the risk of the current route, and further analyzes the change in the risk factor of the current route over a period of time; S6: The intelligent analysis module analyzes the change in the risk change coefficient of the current route and makes a decision on the robot's next action.

2. The industrial robot obstacle avoidance control method based on sensor data analysis according to claim 1, characterized in that: The process of performing image quality analysis on the collected image data in S2 includes: By formula Calculate the quality impact coefficient of the a-th captured image ; Among them, a is any image data captured by the visual sensor, is the number of noise points in the a-th image data, is the preset number of noise points, is the number of spots in the a-th image data, is the preset number of spots, is the damaged area of ​​the image in the a-th image data, is the preset damaged area of ​​the image, is the error influence coefficient, which is set according to empirical fitting. The number of pixels in the a-th image data, is the preset number of pixels, for The standard value of is the clarity of the a-th image data, To preset the definition, for The standard value of To define a function, if , then let , otherwise, let .

3. The industrial robot obstacle avoidance control method based on sensor data analysis according to claim 2 is characterized in that: The process of performing image quality analysis on the collected image data in S2 further includes: By taking the quality influence coefficient of the a-th captured image The preset image quality impact coefficient threshold Make a comparison; like , determine that the current image data quality is poor, and continue to capture the image data; like , judge that the current image data is of high quality, and extract the required features in the image data through the intelligent analysis module.

4. The industrial robot obstacle avoidance control method based on sensor data analysis according to claim 3 is characterized in that: The calculation process in S3 includes: By formula Calculate the risk factor of the current route at the time of the i-th detection ; Among them, i is any risk factor detection, is the distance between the robot and the obstacle during the i-th detection, is the moving speed of the robot during the i-th detection, is the robot’s data processing time, is the warning distance between the robot and the obstacle, is the number of obstacles during the i-th detection, s is any obstacle, is the size of the sth obstacle during the i-th detection, for The standard value of is the position influence coefficient of the sth obstacle during the i-th detection, which is set based on empirical fitting.

5. The industrial robot obstacle avoidance control method based on sensor data analysis according to claim 4, characterized in that: The analysis process in S4 includes: By calculating the risk factor of the i-th time point on the current route The preset risk factor threshold range Make a comparison; like , it is determined that the risk factor of the current route at this time point is low, and it is possible to continue moving on the current route; like , judging that the risk factor of the current route at this point in time is high, the intelligent analysis module instructs the robot to stop moving and continuously monitors the risk factor of the current route.

6. The industrial robot obstacle avoidance control method based on sensor data analysis according to claim 5, characterized in that: The process of analyzing the change in the risk coefficient of the current route in S5 includes: The robot continuously monitors the risk factor of the current route and establishes a risk factor change curve for the current route. ; And through the formula Calculate the change in risk factor of the current route over a period of time ; Where n is the total number of risk factor detections during the continuous monitoring process after the robot stops moving. is the starting time point of the continuous monitoring process after the robot stops moving. For all The average value of It is the end time of the continuous monitoring process after the robot stops moving. is the proportionality coefficient, which is set based on empirical fitting.

7. The industrial robot obstacle avoidance control method based on sensor data analysis according to claim 6, characterized in that: The process of analyzing the change in the risk coefficient of the current route in S5 also includes: By calculating the risk factor change of the current route over a period of time The preset change threshold Make a comparison; like , judging that the obstacle risk factor in the current route is gradually decreasing; like , to determine whether the obstacle risk factor in the current route is gradually increasing or has not changed.

8. The industrial robot obstacle avoidance control method based on sensor data analysis according to claim 7, characterized in that: The decision-making process in S6 includes: When it is determined that the obstacle risk factor in the current route is gradually decreasing, it means that the obstacle is moving away from the robot, and the robot can be controlled to continue moving through the intelligent analysis module; When it is determined that the obstacle risk factor in the current route is gradually increasing or has not changed, it means that the obstacle has not moved or is moving towards the direction close to the robot, and the route is adjusted through the intelligent analysis module.

Images