Obstacle crossing control method and system of power transmission line maintenance robot

Abstract

The invention discloses an obstacle crossing control method for a power transmission line maintenance robot. The obstacle crossing control method comprises the following steps: acquiring front view image data of the power transmission line maintenance robot; marking the view image data to obtain an obstacle judgment training data set; setting a label for the view image data based on the distance between the obstacle and the robot to obtain an obstacle distance judgment training data set; constructing an initial obstacle judgment model and an initial obstacle distance judgment model; training the initial obstacle judgment model by using the obstacle judgment training data set to obtain an obstacle judgment model; training the initial obstacle distance judgment model by using the obstacle distance judgment training data set to obtain an obstacle distance judgment model; and on the basis of the obstacle judgment model and the obstacle distance judgment model, obstacle crossing control of the power transmission line maintenance robot is completed according to the robot obstacle crossing operation strategy. The invention also discloses a system for realizing the obstacle crossing control method of the power transmission line maintenance robot. According to the method, the maintenance robot can quickly and smoothly cross the obstacle to reach the target operation position, the automation and intelligence level of the robot is improved, the degree of dependence on people is reduced, and the operation efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the field of robot control, and in particular relates to an obstacle crossing control method and system for a power transmission line maintenance robot. Background Art

[0002] With the development of science and technology, the application of robots in power systems is becoming more and more common; power companies are using robots to perform inspection and maintenance tasks of high-voltage power lines, reducing the need for personnel to be exposed to high-risk environments. Although the application prospects of robots in power systems are broad, there are still problems such as complex technology, high cost, poor environmental adaptability, limited data processing capabilities, and low naturalness of human-computer interaction.

[0003] Regarding a modular live-working robot with a current publication number of CN115582846A, the robot system is currently used to perform maintenance operations on power transmission lines. During actual operations, the robot has a low level of intelligence and a high level of human participation. The robot is still very dependent on humans, and there is still much room for improvement in its efficiency when performing tasks. Summary of the invention

[0004] One of the purposes of the present invention is to provide an obstacle crossing control method for a power transmission line maintenance robot, which can effectively improve the working efficiency of the robot and further promote the intelligent application of power system robots.

[0005] A second object of the present invention is to provide a system for implementing the obstacle crossing control method of the power transmission line maintenance robot.

[0006] The present invention provides an obstacle crossing control method for a power transmission line maintenance robot, comprising the following steps:

[0007] S1. Obtaining image data of the front field of view of the power transmission line maintenance robot;

[0008] S2. Marking the visual field image data obtained in step S1 to obtain an obstacle judgment training data set;

[0009] S3. Label the field of view image data obtained in step S1 based on the distance between the obstacle and the robot to obtain an obstacle distance judgment training data set;

[0010] S4. Constructing an initial obstacle judgment model and an initial obstacle distance judgment model;

[0011] S5. Using the obstacle judgment training data set obtained in step S2, the initial obstacle judgment model obtained in step S4 is trained to obtain an obstacle judgment model; using the obstacle distance judgment training data set obtained in step S3, the initial obstacle distance judgment model obtained in step S4 is trained to obtain an obstacle distance judgment model;

[0012] S6. Based on the obstacle judgment model and the obstacle distance judgment model obtained in step S5, and according to the robot obstacle crossing operation strategy, complete the obstacle crossing control of the power transmission line maintenance robot.

[0013] In step S1, the image data of the front field of view of the power transmission line maintenance robot is obtained by the onboard camera of the power transmission line maintenance robot.

[0014] Step S2 specifically includes: marking the forward visual field image data obtained in step S1, if there is an obstacle in the image, marking it as "there is an obstacle", otherwise marking it as "there is no obstacle", and obtaining the marked forward visual field image data as an obstacle judgment training data set.

[0015] Step S3 specifically includes: setting a label for the front field image data obtained in step S1, wherein the label is the distance between the obstacle and the robot, and obtaining the front field image data with the label as an obstacle distance judgment training data set.

[0016] Step S4 specifically includes: constructing an initial obstacle judgment model and an initial obstacle distance judgment model based on the Yolov8 model.

[0017] Step S6 is specifically as follows: using the obstacle judgment model obtained in step S5 to make a real-time judgment on whether there is an obstacle in the robot's field of view; then when an obstacle appears, using the obstacle distance judgment model obtained in step S5 to make a distance judgment, and according to the robot's obstacle crossing operation strategy, completing the transmission line maintenance robot's obstacle crossing control.

[0018] Step S6 specifically includes the following steps:

[0019] When the power transmission line maintenance robot is actually in operation, the real-time front view image data is obtained through the onboard front camera, and the data is input into the obstacle judgment model obtained in step S5 to judge whether there is an obstacle in the front view. If there is no obstacle, the maintenance robot drives the walking wheels to move forward at the maximum speed until the result of the judgment is that there is an obstacle, and then enters the next step;

[0020] When the obstacle judgment model outputs that there is an obstacle, the real-time front field image data of the maintenance robot is input into the obstacle distance judgment model obtained in step S5 to obtain the current actual distance between the obstacle and the maintenance robot; according to the obtained current actual distance between the obstacle and the maintenance robot, the maintenance robot decelerates to a preset speed and then travels at a constant speed, which is expressed by the following formula:

[0021]

[0022] Where a is the acceleration of the maintenance robot deceleration; V maxis the maximum speed of the maintenance robot; V0 is the preset uniform speed and V0<V max ; X is the actual distance between the obstacle and the maintenance robot when the obstacle appears in the robot's front field of view; X0 is the preset buffer distance;

[0023] When the sensor on the front wheel of the maintenance robot detects that an obstacle has entered its trigger distance, the maintenance robot moves at a speed of V s The robot moves forward at a constant speed, and the wire clamping mechanism on the front wheel starts to drop to the bottom; when the sensor on the rear wheel detects that the obstacle enters the triggering distance of the rear wheel sensor, the wire clamping mechanism on the front wheel of the robot starts to rise and resume the wire clamping state, and the wire clamping mechanism on the rear wheel starts to drop to the bottom; after continuing to run for the preset time t0, the obstacle crossing control of the power transmission line maintenance robot is completed, and the parameters of the movement process of the maintenance robot through the obstacle meet the following conditions:

[0024]

[0025] Where s is the sensor trigger distance on the walking wheel of the maintenance robot; t s It is the time required for the wire clamping mechanism of the robot's walking wheel to drop to the bottom from the wire clamping state; L0 is the length of the obstacle.

[0026] The present invention also provides a system for implementing the obstacle crossing control method of the power transmission line maintenance robot, comprising a field of view image data acquisition module, an obstacle judgment marking module, an obstacle distance marking module, a model building module, a model training module and a robot obstacle crossing control module which are sequentially connected in series;

[0027] The visual field image data acquisition module acquires the visual field image data in front of the power transmission line maintenance robot and uploads the data to the obstacle judgment and marking module;

[0028] The obstacle judgment and marking module marks the field of view image data according to the received data, obtains the obstacle judgment training data set, and uploads the data to the obstacle distance marking module;

[0029] The obstacle distance labeling module sets labels for the field of view image data based on the distance between the obstacle and the robot according to the received data, obtains the obstacle distance judgment training data set, and uploads the data to the model building module;

[0030] The model building module builds an initial obstacle judgment model and an initial obstacle distance judgment model based on the received data, and uploads the data to the model training module;

[0031] The model training module uses the obstacle judgment training data set to train the initial obstacle judgment model based on the received data to obtain the obstacle judgment model; uses the obstacle distance judgment training data set to train the initial obstacle distance judgment model to obtain the obstacle distance judgment model, and uploads the data to the robot obstacle crossing control module;

[0032] The robot obstacle control module completes the obstacle control of the transmission line maintenance robot according to the received data, based on the obstacle judgment model and the obstacle distance judgment model and according to the robot obstacle crossing operation strategy.

[0033] The present invention discloses an obstacle crossing control method and system for a power transmission line maintenance robot, which can enable the maintenance robot to quickly and smoothly cross obstacles to reach a target operating position, thereby improving the automation and intelligence level of the robot, reducing its dependence on humans, and improving operating efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a schematic diagram of the process of the present invention;

[0035] Figure 2 It is a structural schematic diagram of the system of the present invention;

[0036] Figure 3 This is a rendering of the entire process of the robot overcoming obstacles when the method of the present invention is actually applied; Figure 3 a is the picture when the obstacle passes by the front wheel; Figure 3 b is the picture when the obstacle is located between the front and rear wheels; Figure 3 c is the picture of the wheel after the obstacle passes; Figure 3 d is a picture of the robot successfully overcoming obstacles. DETAILED DESCRIPTION

[0037] The present invention provides an obstacle crossing control method for a power transmission line maintenance robot, the flow chart of which is as follows: Figure 1 As shown, the following steps are included:

[0038] S1. Obtaining image data of the front field of view of the power transmission line maintenance robot;

[0039] In step S1, the image data of the front field of view of the power transmission line maintenance robot is obtained by the onboard camera of the power transmission line maintenance robot.

[0040] S2. Mark the visual field image data obtained in step S1 to obtain an obstacle judgment training data set, specifically:

[0041] The forward visual field image data obtained in step S1 is marked. If there is an obstacle in the image, it is marked as "there is an obstacle", otherwise it is marked as "there is no obstacle", and the marked forward visual field image data is obtained as an obstacle judgment training data set.

[0042] S3. Label the field of view image data obtained in step S1 based on the distance between the obstacle and the robot to obtain an obstacle distance judgment training data set, specifically:

[0043] A label is set for the front field image data obtained in step S1, and the label is the distance between the obstacle and the robot, so that the front field image data with the label is obtained as a training data set for obstacle distance judgment.

[0044] S4. Construct an initial obstacle judgment model and an initial obstacle distance judgment model, specifically:

[0045] Based on the Yolov8 model, an initial obstacle judgment model and an initial obstacle distance judgment model are constructed; Yolov8 is an existing model.

[0046] S5. Using the obstacle judgment training data set obtained in step S2, the initial obstacle judgment model obtained in step S4 is trained to obtain an obstacle judgment model; using the obstacle distance judgment training data set obtained in step S3, the initial obstacle distance judgment model obtained in step S4 is trained to obtain an obstacle distance judgment model;

[0047] S6. Based on the obstacle judgment model and the obstacle distance judgment model obtained in step S5, according to the robot's obstacle crossing operation strategy, the power transmission line maintenance robot obstacle crossing control is completed. Specifically: the obstacle judgment model obtained in step S5 is used to perform real-time judgment on whether there is an obstacle in the robot's field of view; then when an obstacle appears, the obstacle distance judgment model obtained in step S5 is used to perform distance judgment, and according to the robot's obstacle crossing operation strategy, the power transmission line maintenance robot obstacle crossing control is completed.

[0048] Step S6 specifically includes the following steps:

[0049] When the power transmission line maintenance robot is actually in operation, the real-time front view image data is obtained through the onboard front camera, and the data is input into the obstacle judgment model obtained in step S5 to judge whether there is an obstacle in the front view. If there is no obstacle, the maintenance robot drives the walking wheels to move forward at the maximum speed until the result of the judgment is that there is an obstacle, and then enters the next step;

[0050] When the obstacle judgment model outputs that there is an obstacle, the real-time front field image data of the maintenance robot is input into the obstacle distance judgment model obtained in step S5 to obtain the current actual distance between the obstacle and the maintenance robot; according to the obtained current actual distance between the obstacle and the maintenance robot, the maintenance robot decelerates to a preset speed and then travels at a constant speed, which is expressed by the following formula:

[0051]

[0052] Where a is the acceleration of the maintenance robot deceleration; V max is the maximum speed of the maintenance robot; V0 is the preset uniform speed and V0<V max ; X is the actual distance between the obstacle and the maintenance robot when the obstacle appears in the robot's front field of view; X0 is the preset buffer distance;

[0053] When the sensor on the front wheel of the maintenance robot detects that an obstacle has entered its trigger distance, the maintenance robot moves at a speed of V s The robot moves forward at a constant speed, and the wire clamping mechanism on the front wheel starts to drop to the bottom; when the sensor on the rear wheel detects that the obstacle enters the triggering distance of the rear wheel sensor, the wire clamping mechanism on the front wheel of the robot starts to rise and resume the wire clamping state, and the wire clamping mechanism on the rear wheel starts to drop to the bottom; after continuing to run for the preset time t0, the obstacle crossing control of the power transmission line maintenance robot is completed, and the parameters of the movement process of the maintenance robot through the obstacle meet the following conditions:

[0054]

[0055] Where s is the sensor trigger distance on the walking wheel of the maintenance robot; t s It is the time required for the wire clamping mechanism of the robot's walking wheel to drop to the bottom from the wire clamping state; L0 is the length of the obstacle.

[0056] Figure 3 This is a practical application effect diagram of the method of the present invention.

[0057] The present invention also provides a system for implementing the obstacle crossing control method of the power transmission line maintenance robot, and its structural schematic diagram is shown in FIG. Figure 2 As shown, it includes a field of view image data acquisition module, an obstacle judgment marking module, an obstacle distance marking module, a model building module, a model training module and a robot obstacle crossing control module which are connected in series in sequence;

[0058] The visual field image data acquisition module acquires the visual field image data in front of the power transmission line maintenance robot and uploads the data to the obstacle judgment and marking module;

[0059] The obstacle judgment and marking module marks the field of view image data according to the received data, obtains the obstacle judgment training data set, and uploads the data to the obstacle distance marking module;

[0060] The obstacle distance labeling module sets labels for the field of view image data based on the distance between the obstacle and the robot according to the received data, obtains the obstacle distance judgment training data set, and uploads the data to the model building module;

[0061] The model building module builds an initial obstacle judgment model and an initial obstacle distance judgment model based on the received data, and uploads the data to the model training module;

[0062] The model training module uses the obstacle judgment training data set to train the initial obstacle judgment model based on the received data to obtain the obstacle judgment model; uses the obstacle distance judgment training data set to train the initial obstacle distance judgment model to obtain the obstacle distance judgment model, and uploads the data to the robot obstacle crossing control module;

[0063] The robot obstacle control module completes the obstacle control of the transmission line maintenance robot according to the received data, based on the obstacle judgment model and the obstacle distance judgment model and according to the robot obstacle crossing operation strategy.

Claims

1. An obstacle crossing control method for a power transmission line maintenance robot, characterized in that: The following steps are involved: S1. Obtaining image data of the front field of view of the power transmission line maintenance robot; S2. Marking the visual field image data obtained in step S1 to obtain an obstacle judgment training data set; S3. Label the field of view image data obtained in step S1 based on the distance between the obstacle and the robot to obtain an obstacle distance judgment training data set; S4. Constructing an initial obstacle judgment model and an initial obstacle distance judgment model; S5. Using the obstacle judgment training data set obtained in step S2, the initial obstacle judgment model obtained in step S4 is trained to obtain an obstacle judgment model; Using the obstacle distance judgment training data set obtained in step S3, the initial obstacle distance judgment model obtained in step S4 is trained to obtain an obstacle distance judgment model; S6. Based on the obstacle judgment model and the obstacle distance judgment model obtained in step S5, and according to the robot obstacle crossing operation strategy, complete the obstacle crossing control of the power transmission line maintenance robot.

2. The obstacle crossing control method of the power transmission line maintenance robot according to claim 1, characterized in that: In step S1, the image data of the front field of view of the power transmission line maintenance robot is obtained by using the onboard camera of the power transmission line maintenance robot.

3. The obstacle crossing control method of the power transmission line maintenance robot according to claim 1, characterized in that: Step S2 is specifically: marking the forward visual field image data obtained in step S1, if there is an obstacle in the image, marking it as "obstacle", otherwise marking it as "no obstacle", and obtaining the marked forward visual field image data as an obstacle judgment training data set.

4. The obstacle crossing control method of the power transmission line maintenance robot according to claim 1, characterized in that: Step S3 specifically includes: setting a label for the front field image data obtained in step S1, wherein the label is the distance between the obstacle and the robot, and obtaining the front field image data with the label as an obstacle distance judgment training data set.

5. The obstacle crossing control method of the power transmission line maintenance robot according to claim 1, characterized in that: Step S4 specifically includes: constructing an initial obstacle judgment model and an initial obstacle distance judgment model based on the Yolov8 model.

6. The obstacle crossing control method of the power transmission line maintenance robot according to claim 5, characterized in that: Step S6 is specifically as follows: using the obstacle judgment model obtained in step S5 to make a real-time judgment on whether there is an obstacle in the robot's field of view; then when an obstacle appears, using the obstacle distance judgment model obtained in step S5 to make a distance judgment, and according to the robot's obstacle crossing operation strategy, completing the transmission line maintenance robot's obstacle crossing control.

7. The obstacle crossing control method of the power transmission line maintenance robot according to claim 6, characterized in that: Step S6 specifically includes the following steps: When the power transmission line maintenance robot is actually in operation, the real-time front view image data is obtained through the onboard front camera, and the data is input into the obstacle judgment model obtained in step S5 to judge whether there is an obstacle in the front view. If there is no obstacle, the maintenance robot drives the walking wheels to move forward at the maximum speed until the result of the judgment is that there is an obstacle, and then enters the next step; When the obstacle judgment model outputs that there is an obstacle, the real-time front field image data of the maintenance robot is input into the obstacle distance judgment model obtained in step S5 to obtain the current actual distance between the obstacle and the maintenance robot; according to the obtained current actual distance between the obstacle and the maintenance robot, the maintenance robot decelerates to a preset speed and then travels at a constant speed, which is expressed by the following formula: Where a is the acceleration of the maintenance robot deceleration; V max is the maximum speed of the maintenance robot; V0 is the preset uniform speed and V0<V max ; X is the actual distance between the obstacle and the maintenance robot when the obstacle appears in the robot's front field of view; X0 is the preset buffer distance; When the sensor on the front wheel of the maintenance robot detects that an obstacle has entered its trigger distance, the maintenance robot moves at a speed of V s The robot moves forward at a constant speed, and the wire clamping mechanism on the front wheel starts to drop to the bottom; when the sensor on the rear wheel detects that the obstacle enters the triggering distance of the rear wheel sensor, the wire clamping mechanism on the front wheel of the robot starts to rise and resume the wire clamping state, and the wire clamping mechanism on the rear wheel starts to drop to the bottom; after continuing to run for the preset time t0, the obstacle crossing control of the power transmission line maintenance robot is completed, and the parameters of the movement process of the maintenance robot through the obstacle meet the following conditions: Where s is the sensor trigger distance on the walking wheel of the maintenance robot; t s It is the time required for the wire clamping mechanism of the robot's walking wheel to drop to the bottom from the wire clamping state; L0 is the length of the obstacle.

8. A system for implementing the obstacle crossing control method of the power transmission line maintenance robot according to any one of claims 1 to 7, characterized in that: It includes a field of view image data acquisition module, an obstacle judgment marking module, an obstacle distance marking module, a model building module, a model training module and a robot obstacle crossing control module which are connected in series in sequence; The visual field image data acquisition module acquires the visual field image data in front of the power transmission line maintenance robot and uploads the data to the obstacle judgment and marking module; The obstacle judgment and marking module marks the field of view image data according to the received data, obtains the obstacle judgment training data set, and uploads the data to the obstacle distance marking module; The obstacle distance labeling module sets labels for the field of view image data based on the distance between the obstacle and the robot according to the received data, obtains the obstacle distance judgment training data set, and uploads the data to the model building module; The model building module builds an initial obstacle judgment model and an initial obstacle distance judgment model based on the received data, and uploads the data to the model training module; The model training module uses the obstacle judgment training data set to train the initial obstacle judgment model based on the received data to obtain the obstacle judgment model; uses the obstacle distance judgment training data set to train the initial obstacle distance judgment model to obtain the obstacle distance judgment model, and uploads the data to the robot obstacle crossing control module; The robot obstacle control module completes the obstacle control of the transmission line maintenance robot according to the received data, based on the obstacle judgment model and the obstacle distance judgment model and according to the robot obstacle crossing operation strategy.

Images