Robot path planning method based on machine vision

Abstract

The invention belongs to the technical field of robot control, and discloses a robot path planning method based on machine vision. The method comprises the following steps: step 1, acquiring image data of a planning area, and performing preprocessing operation on the image data; 2, analyzing the image data to generate a to-be-selected path; step 3, analyzing according to characteristic parameters contained in the generated to-be-selected path, and selecting an optimal path; 4, analyzing according to the associated data on the optimal path, generating a deviation degree, judging whether the driving path of the robot deviates or not based on the deviation degree, and generating an alarm signal when the path deviates; and step 5, making a corresponding response according to the alarm signal. According to the method, comprehensive analysis can be carried out in combination with multiple influence factors on the to-be-selected path, and each path is evaluated from multiple angles, so that the actual advantages and disadvantages of the paths are reflected more comprehensively, the optimal path is selected, and the driving efficiency of the robot is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of robot control, and more particularly, to a robot path planning method based on machine vision. BACKGROUND

[0002] A robot is an intelligent execution system integrating mechanical design, artificial intelligence, sensor fusion and automatic control technology. With the continuous development of science, the technology of robots is constantly matured, and the robot technology is widely used in medical, service, industrial and other fields. For example, in the industrial field, the robot technology can be used to realize the carrying of goods and improve the industrial efficiency.

[0003] In the process of carrying goods, the driving path of the robot is generally planned in advance. The existing path planning is mostly based on the starting point and the ending point distance of the robot, and then the nearest passable road is selected to determine. However, due to the high sensitivity of the robot, the robot will be disturbed by various influencing factors during driving, resulting in abnormal driving of the robot. The robot is generally slow when turning. If the road has too many corners, although the path is short, it will still affect the driving of the robot, thereby affecting the carrying efficiency of the robot. In addition, the existing robot generally alarms after the path deviation occurs when planning the path, and cannot extract the prediction, resulting in a certain hysteresis of the response.

[0004] In view of this, the present application provides a robot path planning method based on machine vision to solve the above problems. SUMMARY

[0005] In order to overcome the above-mentioned defects of the prior art and achieve the above-mentioned purposes, the present application provides the following technical scheme: A robot path planning method based on machine vision, characterized in that the method comprises: Step one, collecting image data of the planning area and performing preprocessing operation on the image data; Step two, analyzing the preprocessed image data, identifying the obstacle area and the non-obstacle area, and generating a selected path according to the starting point and the target point of the robot; Step three, analyzing the feature parameters contained in each selected path generated, and selecting the optimal path; Step four, collecting and analyzing the associated data on the optimal path in real time to generate a deviation degree, and judging whether the driving path of the robot deviates based on the deviation degree, and generating an alarm signal when there is path deviation; Step five, responding to the alarm signal.

[0006] Furthermore, the image data includes RGB image data acquired by an RGB camera and depth image data acquired by a depth camera.

[0007] Furthermore, the method for generating candidate paths in step two is as follows: Based on the image data of the planning area, the planning area is divided into multiple grids of equal size to obtain a grid map of the planning area; according to the identified obstacle area and non-obstacle area, the grids in the grid map are divided into obstacle grids and non-obstacle grids. Based on the robot's starting point and target point, a path composed of non-obstacle grids is searched on the grid map to generate multiple different candidate paths.

[0008] Furthermore, the method for selecting the optimal path in step three is as follows: Based on the characteristic parameters contained in each candidate path, a matching value is generated for each candidate path. The candidate paths are sorted from largest to smallest according to their corresponding matching values, and the candidate path with the highest ranking is selected as the optimal path.

[0009] Furthermore, the method for generating the matching value is as follows: The feature parameters include positive gain feature parameters and negative gain feature parameters. The negative gain feature parameters include path length, travel time, number of path turns, number of obstacles, and obstacle volume. The positive gain feature parameters include shortest safe distance, path smoothness, and path width. The feature parameters of each candidate path are normalized and weights are assigned to each feature parameter. The feature parameters in the negative gain feature parameters are weighted and summed to obtain the negative gain score. The feature parameters in the positive gain feature parameters are weighted and summed to obtain the positive gain score. The ratio of the positive gain score to the negative gain score is calculated to obtain the matching value of each candidate path.

[0010] Furthermore, the method for generating the degree of deviation in step four is as follows: The associated data includes image data of the robot traveling on the optimal path; based on the selected optimal path, a standard route is set on the optimal path, and multiple standard route center points are set on the center line of the standard route. n consecutive images of the robot traveling along the optimal path are collected. Based on the distance between the robot's position in each image and the center point of the corresponding standard route, the deviation value of each image is determined. Based on the deviation values ​​of the n images, a curve function showing the change of the deviation value with the images is fitted. The integral and the maximum first derivative of the curve function are calculated. The integral and the maximum first derivative are normalized and weighted. The integral and the maximum first derivative are then weighted and summed according to the weights to calculate the degree of deviation.

[0011] Furthermore, the method for generating the alarm signal in step four is as follows: Pre-set deviation thresholds A, B, and C. When the deviation exceeds the preset deviation threshold A, it is determined that the robot's path has deviated, and an alarm signal is generated.

[0012] Furthermore, the alarm signals include a first alarm signal, a second alarm signal, and a third alarm signal, and the severity of the third alarm signal is higher than that of the second alarm signal, and the severity of the second alarm signal is higher than that of the first alarm signal; When the third alarm signal is generated, the robot stops moving. When the first and second alarm signals are generated, the robot's position is corrected.

[0013] Furthermore, the method for determining the first alarm signal, the second alarm signal, and the third alarm signal is as follows: the alarm signal is determined based on the degree of deviation; When the degree of deviation is between the deviation threshold A and the deviation threshold B, a first alarm signal is generated; when the degree of deviation is between the deviation threshold B and the deviation threshold C, a second alarm signal is generated; and when the degree of deviation exceeds the deviation threshold C, a third alarm signal is generated.

[0014] Furthermore, the method for correcting the robot's position is as follows: The deviation adjustment coefficient is calculated based on the difference between the degree of deviation and the corresponding deviation threshold. This coefficient is then multiplied by a preset distance conversion coefficient to obtain the adjustment distance. Based on the adjustment distance, the robot is controlled to move towards the center point of the standard route corresponding to the current position. When the first alarm signal is generated, the deviation adjustment coefficient is the deviation threshold B minus the deviation degree; when the second alarm signal is generated, the deviation adjustment coefficient is the deviation threshold C minus the deviation degree.

[0015] The technical effects and advantages of the robot path planning method based on machine vision of the present invention are as follows: This invention can combine multiple influencing factors on the candidate path for comprehensive analysis, evaluate each path from multiple perspectives, and thus more comprehensively reflect the actual advantages and disadvantages of the path, thereby selecting the optimal path and improving the robot's driving efficiency.

[0016] This invention can comprehensively analyze the cumulative deviation of the robot's path and the speed of deviation during the robot's movement, thereby more accurately predicting the degree of deviation risk of the robot's movement and thus providing early warning and handling to ensure the robot's driving efficiency. Attached Figure Description

[0017] Figure 1This is a flowchart of a robot path planning method based on machine vision according to Embodiment 1 of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Please see Figure 1 As shown in this embodiment, a robot path planning method based on machine vision is described. The method includes: Step 1: Collect image data of the planned area and perform preprocessing operations on the image data; Step 2: Analyze the preprocessed image data to identify obstacle areas and non-obstacle areas, and generate candidate paths based on the robot's starting point and target point; Step 3: Analyze the characteristic parameters of each generated candidate path and select the optimal path; Step 4: Collect and analyze the associated data on the optimal path in real time to generate the degree of deviation. Based on the degree of deviation, determine whether the robot's driving path has deviated. When there is a path deviation, generate an alarm signal. Step 5: Respond accordingly to the alarm signal.

[0020] Through the above technical solution, this application uses a visual sensor to acquire image data of the planned area. This image data mainly consists of RGB image data obtained through an RGB camera and depth image data obtained through a depth camera. Preprocessing operations are performed on the acquired image data, including but not limited to time synchronization processing, noise reduction, and enhancement processing of the image data, and hole filling, smoothing, and outlier removal processing of the depth image data. These processing methods are all based on existing technologies and will not be described here. Based on the processed image data of the planned area, obstacle areas and non-obstacle areas are identified. The identification method can utilize existing technologies such as semantic segmentation. The process involves several steps: First, an obstacle and a non-obstacle labeled dataset is prepared to train the model, constructing a deep learning semantic segmentation model. Processed image data is input into the model for automatic segmentation, outputting obstacle and non-obstacle regions. Based on the robot's starting point and target point within the planned area, multiple candidate paths are automatically generated. The feature parameters of each candidate path are analyzed to select the optimal path for the robot. This comprehensive analysis of various influencing factors along the path more accurately reflects the suitability of each path for the robot's movement, effectively improving efficiency. While the robot is on the optimal path, key data is analyzed to generate the corresponding deviation level. Based on this deviation, the system determines if the robot has deviated from its path. If deviation is detected, an alarm signal is generated, allowing the terminal to respond with corresponding instructions, such as correcting the robot's course or stopping it, ensuring normal operation.

[0021] The method for generating candidate paths in step two is as follows: Based on the image data of the planning area, the planning area is divided into multiple grids of equal size to obtain a grid map of the planning area; according to the identified obstacle area and non-obstacle area, the grids in the grid map are divided into obstacle grids and non-obstacle grids. Based on the robot's starting point and target point, a path composed of non-obstacle grids is searched on the grid map to generate multiple different candidate paths.

[0022] The above scheme provides a specific method for generating candidate paths. First, based on the image data of the planning area, the planning area is divided into multiple equally sized grids, thus forming a grid map of the planning area. According to the identified obstacle and non-obstacle areas, the grids within the grid map are further divided into obstacle grids and non-obstacle grids. Based on the robot's starting point and target point, the RRT algorithm is used to search for paths composed of non-obstacle grids on the grid map, thereby generating multiple different candidate paths. This allows for the automatic generation of multiple driving path selections based on the non-obstacle areas between the robot's starting point and target point, improving the selection options.

[0023] The method for selecting the optimal path in step three is as follows: based on the characteristic parameters contained in each candidate path, analyze the matching value of each candidate path, sort each candidate path from largest to smallest according to the corresponding matching value, and select the candidate path with the highest ranking as the optimal path. The matching value is generated as follows: the feature parameters include positive gain feature parameters and negative gain feature parameters. The negative gain feature parameters include path length, travel time, number of path turns, number of obstacles, and obstacle volume. The positive gain feature parameters include shortest safe distance, path smoothness, and path width. The feature parameters of each candidate path are normalized and weights are assigned to each feature parameter. The feature parameters in the negative gain feature parameters are weighted and summed to obtain the negative gain score. The feature parameters in the positive gain feature parameters are weighted and summed to obtain the positive gain score. The ratio of the positive gain score to the negative gain score is calculated to obtain the matching value of each candidate path.

[0024] The above technical solution provides a specific method for selecting the optimal path. First, based on the generated candidate paths, the characteristic parameters contained in each candidate path are obtained. These characteristic parameters include positive gain characteristic parameters and negative gain characteristic parameters. Positive gain characteristic parameters indicate that the larger the value of the characteristic parameter, the more suitable it is for robot driving; negative gain characteristic parameters indicate that the smaller the value of the characteristic parameter, the more suitable it is for robot driving. For example, negative gain characteristic parameters include path length, driving time, number of path turns, and the number and volume of obstacles within a specific range of the path (e.g., within 1m of the path). The smaller these values, the less impact they have on the robot, and the more suitable it is for robot driving. Positive gain characteristic parameters include the shortest safe distance, path smoothness, and path width. The larger these values, the less impact they have on robot driving, and the more suitable it is for robot driving. Among these data, the path length can be obtained from the accumulated grid distance, the driving time can be determined according to the robot's set driving speed and path length, the number of path turns can be determined according to the number of turns in the generated path, and the number and volume of obstacles within a specific range of the path... Volume can be obtained through depth data. The shortest safe distance can be calculated by uniformly selecting sampling points on the candidate path, calculating the distance from each sampling point to the nearest obstacle, and taking the minimum value as the shortest safe distance. Path smoothness can be obtained by calculating the curvature change of the path; the smaller the curvature change, the smoother the path and the higher the smoothness. Path width can be determined based on the average passable width of the candidate paths. After obtaining these data, the feature parameters of each candidate path are normalized to ensure consistent dimensions. Then, weights are assigned to each feature parameter according to its influence on the robot. The negative gain feature parameters are weighted and summed to obtain the negative gain score, and the positive gain feature parameters are weighted and summed to obtain the positive gain score. Using the negative gain score as the denominator and the positive gain score as the numerator, the matching value of each candidate path is calculated based on the ratio. It can be seen that the larger the matching value, the more suitable the path is for the current robot to drive. Therefore, the candidate paths are sorted in descending order of matching value, and the candidate path with the highest ranking is selected as the optimal path. This method allows for a comprehensive analysis of various influencing factors on the candidate path, enabling the path to be evaluated from multiple perspectives. This provides a more complete picture of the path's actual advantages and disadvantages, thereby improving the robot's driving efficiency.

[0025] The method for generating the degree of deviation in step four is as follows: The associated data includes image data of the robot traveling on the optimal path; based on the selected optimal path, a standard route is set on the optimal path, and multiple standard route center points are set on the center line of the standard route. n consecutive images of the robot traveling along the optimal path are collected. Based on the distance between the robot's position in each image and the center point of the corresponding standard route, the deviation value of each image is determined. Based on the deviation values ​​of the n images, a curve function showing the change of the deviation value with the images is fitted. The integral and the maximum first derivative of the curve function are calculated. The integral and the maximum first derivative are normalized and weighted. The integral and the maximum first derivative are then weighted and summed according to the weights to calculate the degree of deviation.

[0026] The above technical solution provides a specific method for obtaining the degree of deviation. First, the associated data includes image data of the robot traveling on the optimal path. Based on the selected optimal path, a standard route is pre-defined on this optimal path, and multiple points are set on the center line of the standard route, denoted as the center point of the standard route. Then, n images of the robot traveling on the optimal path are continuously collected in chronological order. Based on the distance between the robot's position in each image and the corresponding center point of the standard route, the robot's position can be determined according to the coordinates of the corresponding grid in the image, thus determining the deviation value for each image. Obviously, the larger the deviation value, the greater the possibility that the robot has deviated from the route. To improve detection accuracy, based on the continuously collected n... For each image, a curve function is fitted to represent the deviation value as a function of the number of images. Based on this curve function, the integral and the maximum first derivative are calculated. The integral value represents the cumulative deviation of the robot; a larger value indicates a greater likelihood of deviation. The maximum first derivative represents the maximum rate of deviation; a larger value indicates a rapidly increasing deviation, further increasing the likelihood of deviation. To improve analysis, the integral and the maximum first derivative are normalized and weighted. The weights are manually assigned based on the proportion of influence of each factor when deviations occur in historical data. Finally, a weighted sum is calculated to determine the degree of deviation. A greater degree of deviation indicates a higher probability of path deviation. This method allows for a comprehensive analysis of the robot's cumulative path deviation and deviation rate during operation, enabling more accurate prediction of the risk of deviation and allowing for early warning and intervention to ensure the robot's operational efficiency.

[0027] The method for generating an alarm signal in step four is as follows: pre-set deviation thresholds A, B, and C. When the deviation exceeds the preset deviation threshold A, it is determined that the robot's path has deviated, and an alarm signal is generated. The alarm signals include a first alarm signal, a second alarm signal, and a third alarm signal, with the severity of the third alarm signal being higher than that of the second alarm signal, and the severity of the second alarm signal being higher than that of the first alarm signal. When the third alarm signal is generated, the robot stops moving; when the first alarm signal and the second alarm signal are generated, the robot's position is corrected. The method for determining the first alarm signal, the second alarm signal, and the third alarm signal is as follows: the alarm signal is determined based on the degree of deviation. When the degree of deviation is between the deviation threshold A and the deviation threshold B, a first alarm signal is generated; when the degree of deviation is between the deviation threshold B and the deviation threshold C, a second alarm signal is generated; when the degree of deviation exceeds the deviation threshold C, a third alarm signal is generated. The method for correcting the robot's position is as follows: calculate the deviation adjustment coefficient based on the difference between the degree of deviation and the corresponding deviation threshold, and multiply it by a preset distance conversion coefficient to obtain the adjustment distance; based on the adjustment distance, control the robot to move towards the center point of the standard route corresponding to the current position; When the first alarm signal is generated, the deviation adjustment coefficient is the deviation threshold B minus the deviation degree; when the second alarm signal is generated, the deviation adjustment coefficient is the deviation threshold C minus the deviation degree.

[0028] The above technical solution provides a specific method for generating alarm signals and a specific method for responding. First, three deviation thresholds are set based on experience: deviation threshold A, deviation threshold B, and deviation threshold C. Deviation threshold A... Deviation threshold B A deviation threshold C is set. When the detected deviation exceeds the deviation threshold A, the robot's path is determined to have deviated, and an alarm signal is generated. These alarm signals include a first alarm signal, a second alarm signal, and a third alarm signal, with the third alarm signal being more severe than the second alarm signal, and the second alarm signal being more severe than the first alarm signal. The generation of the first and second alarm signals indicates that although the robot is deviating, it can still function normally after correcting its position. The generation of the third alarm signal indicates abnormal robot behavior, so the robot is stopped and an investigation is initiated. Specifically, when the deviation is between the deviation thresholds A and B, it indicates a small deviation trend, and the first alarm signal is generated. When the deviation is between deviation thresholds B and C, it indicates a large deviation trend, and a second alarm signal is generated. When the deviation exceeds deviation threshold C, it indicates a serious deviation trend in the robot's movement, and a third alarm signal is generated. When a third alarm signal is generated, a shutdown check is performed. When the first and second alarm signals are generated, PID control commands are used for automatic early correction. The specific correction method is as follows: A deviation adjustment coefficient is calculated based on the difference between the current deviation magnitude and the current deviation threshold, and then multiplied by a preset distance conversion coefficient. The distance conversion coefficient can be obtained through comparative experiments using historical data. The adjustment distance is then calculated, and the robot is moved towards the center point of the standard route by the corresponding adjustment distance based on its current position. For example, if a second alarm signal is generated, the deviation is K, the deviation thresholds B and C are KB and KC respectively, and the preset distance conversion coefficient is... So, the actual adjustment distance Then, the robot will be moved a distance L towards the center point of the standard route. This method allows for the determination of different deviation severity trends based on the generated alarm signals, enabling corresponding countermeasures and providing a more accurate understanding of the robot's movement.

[0029] This invention can comprehensively analyze multiple influencing factors on candidate paths to evaluate each path from multiple perspectives, thereby more comprehensively reflecting the actual advantages and disadvantages of each path and selecting the optimal path to improve the robot's driving efficiency. Furthermore, this invention can comprehensively analyze the robot's cumulative path deviation and deviation speed during its journey to more accurately predict the degree of deviation risk, enabling early warning and intervention to ensure the robot's driving efficiency.

[0030] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0031] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0032] In the description of this invention, it should be understood that the terms "first," "second," etc., are used only for distinguishing descriptions and should not be construed as indicating or implying relative importance.

[0033] In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0034] In the description of this invention, "several" means one or more, and "a large number" means two or more.

[0035] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0036] All formulas in this manual are dimensionless and calculated numerically. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0037] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A method for robot path planning based on machine vision, characterized in that, The method comprises: Step one, collecting image data of the planning area and performing preprocessing operation on the image data; Step two, analyzing the preprocessed image data, identifying the obstacle area and the non-obstacle area, and generating candidate paths according to the starting point and the target point of the robot; Step three, analyzing the feature parameters contained in each candidate path to select the optimal path; Step four, collecting and analyzing the associated data on the optimal path in real time to generate the deviation degree, judging whether the travel path of the robot deviates based on the deviation degree, and generating an alarm signal when there is path deviation; Step five, responding to the alarm signal.

2. The robot path planning method based on machine vision according to claim 1, characterized in that, The image data includes RGB image data obtained by an RGB camera and depth image data obtained by a depth camera.

3. The robot path planning method based on machine vision according to claim 2, characterized in that, The method for generating candidate paths in step two is: Based on the image data of the planning area, the planning area is divided into a plurality of equal-sized grids to obtain a grid map of the planning area; the grids in the grid map are divided into obstacle grids and non-obstacle grids according to the identified obstacle area and non-obstacle area; According to the starting point and the target point of the robot, search for the path composed of non-obstacle grids on the grid map to generate a plurality of different candidate paths.

4. The robot path planning method based on machine vision according to claim 3, characterized in that, The method for selecting the optimal path in step three is: Based on the feature parameters contained in each candidate path, the matching value of each candidate path is generated, each candidate path is sorted according to the corresponding matching value from large to small, and the candidate path with the highest ranking is selected as the optimal path.

5. The robot path planning method based on machine vision according to claim 4, characterized in that, The method for generating the matching value is: The feature parameters include positive gain feature parameters and negative gain feature parameters, wherein the negative gain feature parameters include path length, travel time, number of path corners, number of obstacles, and obstacle volume, and the positive gain feature parameters include shortest safety distance, path smoothness, and path width; The feature parameters of each candidate path are normalized, and each feature parameter is assigned a weight; the feature parameters in the negative gain feature parameters are weighted and summed to obtain a negative gain score, the feature parameters in the positive gain feature parameters are weighted and summed to obtain a positive gain score, and the ratio of the positive gain score to the negative gain score is calculated to obtain the matching value of each candidate path.

6. The robot path planning method based on machine vision according to claim 5, wherein, The method for generating the deviation degree in step four is: The associated data includes picture data of the robot traveling on the optimal path; according to the selected optimal path, a standard route is set on the optimal path, and a plurality of standard route center points are set on the center line of the standard route; n pictures of the robot traveling on the optimal path are continuously collected, the deviation value of each picture is determined according to the distance between the position of the robot in each picture and the corresponding standard route center point, the deviation value is fitted with a curve function that changes with the picture based on the deviation values of the n pictures, the integral and the maximum first derivative of the curve function are calculated, and the integral and the maximum first derivative are normalized and weighted; the integral and the maximum first derivative are weighted and summed according to the assigned weights to calculate the deviation degree.

7. The robot path planning method based on machine vision according to claim 6, characterized in that, The method for generating the alarm signal in step four is: The deviation degree threshold A, the deviation degree threshold B and the deviation degree threshold C are preset, when the deviation degree exceeds the preset deviation degree threshold A, it is judged that the path of the robot deviates, and an alarm signal is generated.

8. The robot path planning method based on machine vision according to claim 7, characterized in that, The alarm signal includes a first alarm signal, a second alarm signal and a third alarm signal, and the third alarm signal has a higher severity than the second alarm signal, and the second alarm signal has a higher severity than the first alarm signal. When the third alarm signal is generated, the robot movement is stopped, and when the first alarm signal and the second alarm signal are generated, the robot driving position is corrected.

9. The robot path planning method based on machine vision according to claim 8, characterized in that, The first alarm signal, the second alarm signal and the third alarm signal determination method is to determine the alarm signal according to the deviation degree; When the deviation degree is between the deviation degree threshold A and the deviation degree threshold B, the first alarm signal is generated, when the deviation degree is between the deviation degree threshold B and the deviation degree threshold C, the second alarm signal is generated, and when the deviation degree exceeds the deviation degree threshold C, the third alarm signal is generated.

10. The robot path planning method based on machine vision according to claim 9, wherein, The method for correcting the robot driving position is: According to the difference between the deviation degree and the corresponding deviation degree threshold, a deviation adjustment coefficient is obtained, and multiplied by a preset distance conversion coefficient to obtain an adjustment distance; according to the adjustment distance, the robot is controlled to move to the standard route center point corresponding to the current position; When the first alarm signal is generated, the deviation adjustment coefficient is the deviation degree threshold B minus the deviation degree; when the second alarm signal is generated, the deviation adjustment coefficient is the deviation degree threshold C minus the deviation degree.

Images