Robot vision line patrol navigation method and vision line patrol robot

Abstract

The present invention discloses a robot visual line-patrol navigation method and a visual line-patrol robot. The method comprises: setting a navigation line on the ground; acquiring ground image information using a camera on the robot; analyzing the position of the navigation line in the image information, and determining the robot's direction of travel based on the position of the navigation line. The navigation line is a visually identifiable object, such as one directly sprayed on the ground, significantly reducing the installation and maintenance costs of the visual line patrol. The camera used to acquire image information on the robot can often also be used for other purposes, reducing equipment duplication and redundancy and lowering the overall cost of the robot.

Description

Technical Field

[0001] The invention relates to a robot visual line patrol navigation algorithm and a robot adopting the algorithm. Background Art

[0002] Line-following navigation is a commonly used robot navigation method. Its principle is to set a guideline on the ground, which the robot identifies and moves along. Traditional line-following navigation methods use magnetic guidance and electromagnetic induction. These methods all involve laying magnetic guide rails, which requires modification to the existing environment. This is costly and is susceptible to electromagnetic fields in the installation environment, making navigation failures a common problem. Summary of the Invention

[0003] In view of this, the present invention provides a robot visual line patrol navigation method and a visual line patrol robot, which uses visual line patrol to replace traditional electromagnetic line patrol navigation, which is more convenient and economical.

[0004] To solve the above technical problems, the present invention provides a robot visual line patrol navigation method, comprising: setting a navigation line on the ground; obtaining ground image information using a camera on the robot; analyzing the position of the navigation line on the image information, and determining the robot's travel direction based on the position of the navigation line.

[0005] As an improvement, the color of the navigation line contrasts with the color of the ground.

[0006] As a further improvement, the analysis of the position of the navigation line on the image information and determining the direction of movement of the robot based on the position of the navigation line includes: dividing the image information from left to right into a left-turn area, a straight-ahead area, and a right-turn area; judging the position of the starting point of the navigation line in the image information, if it is in the left-turn area, the robot turns left; if it is in the straight-ahead area, the robot goes straight; if it is in the right-turn area, the robot turns right.

[0007] As another further improvement, the left-turn area is divided into several left rotor areas from left to right, and the right-turn area is divided into several right rotor areas from right to left; the turning radius of the robot in each left rotor area increases from left to right, and the turning radius of the robot in each right rotor area increases from right to left.

[0008] As a preferred embodiment, the left-turn area is divided into a left small bend, a left middle bend area, and a left large bend area from left to right, and the right-turn area is divided into a right small bend area, a right middle bend area, and a right large bend area from right to left; if the navigation line is in the left large bend area, the robot turns a large left bend; if the navigation line is in the left middle bend area, the robot turns a medium left bend; if the navigation line is in the left small bend area, the robot turns a small left bend; if the navigation line is in the right small bend area, the robot turns a small right bend; if the navigation line is in the right middle bend area, the robot turns a medium right bend; if the navigation line is in the right large bend area, the robot turns a large right bend.

[0009] As a preference, the proportion of the left small curve area, the left middle curve area, the left large curve area, the straight area, the right large curve area, the right middle curve area, and the right small curve area is 110:20:20:20:20:20:110.

[0010] As a preferred embodiment, the speeds of the left and right wheels of the robot are equal when the robot is moving straight; the speeds of the left and right wheels are equal when the robot is moving straight; the ratio of the left and right wheel speeds when the robot turns left slightly is 1:3; the ratio of the left and right wheel speeds when the robot turns left in a medium turn is 3:5; the ratio of the left and right wheel speeds when the robot turns left in a large turn is 7:9; the ratio of the left and right wheel speeds when the robot turns right slightly is 3:1; the ratio of the left and right wheel speeds when the robot turns right in a medium turn is 5:3; and the ratio of the left and right wheel speeds when the robot turns right in a large turn is 9:7.

[0011] As an improvement, the robot turns by using the speed difference between the left and right wheels:

[0012] V l =V s +V t

[0013] V r =V s -V t

[0014] Among them, V l is the speed of the left wheel, V r is the speed of the right wheel, V s is the robot's straight-line speed, V t To adjust the speed;

[0015] V t =K p *E

[0016] E=X1-X m

[0017] Among them, K p is the proportional adjustment constant, E is the error amount, X1 is the horizontal coordinate of the starting point of the navigation line in the image information, X m is the horizontal coordinate of the midpoint of the image information.

[0018] As an improvement, the robot turns by using the speed difference between the left and right wheels:

[0019] V l =V s +V t

[0020] V r =V s -V t

[0021] Among them, Vl is the speed of the left wheel, V r is the speed of the right wheel, V s is the robot's straight-line speed, V t To adjust the speed;

[0022] V t =K p *E n +K d *(E n -E n-1 )

[0023] E n =X n -X m

[0024] E n-1 =X n-1 -X m

[0025] Among them, K p is the proportional adjustment constant, K d is the differential adjustment constant, E n is the error amount, E n-1 is the last error, X n The horizontal coordinate of the starting point of the navigation line in the image information of this frame, X n-1 The horizontal coordinate of the starting point of the navigation line in the previous frame image information, X m is the horizontal coordinate of the midpoint of the image information.

[0026] The present invention also provides a visual line patrol robot, comprising a visual line patrol navigation module and a motion module; the motion module moves according to the forward and turning instructions provided by the visual line patrol navigation module; the visual line patrol navigation module comprises an image information acquisition module for acquiring image information, and an analysis module for analyzing the position of a navigation line on the image information and determining the robot's travel direction based on the position of the navigation line.

[0027] The present invention is beneficial in that the guidance lines are visually identifiable objects, such as those sprayed directly on the ground, significantly reducing the installation and maintenance costs of visual inspection lines. The cameras used to acquire image information can often be used for other purposes on the robot, reducing equipment duplication and redundancy, and lowering the overall cost of the robot. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 Flowchart of the present invention.

[0029] Figure 2 Schematic diagram of image information in the present invention.

[0030] Figure 3It is a structural principle diagram of the present invention.

[0031] Mark in the figure: 1 DETAILED DESCRIPTION

[0032] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention is further described in detail below in conjunction with specific implementation methods.

[0033] like Figure 1 、 Figure 2 As shown, the present invention provides a robot visual line patrol navigation method, comprising:

[0034] S1 sets a guideline on the ground; it is expected that the guideline should be visually recognizable. It is best if the color of the guideline contrasts with the ground color to facilitate visual detection of the guideline.

[0035] S2 uses the camera on the robot to obtain ground image information;

[0036] S3 analyzes the position of the navigation line in the image information and determines the robot's direction of travel based on the position of the navigation line. Specifically, the image information is divided into a left-turn zone, a straight-ahead zone, and a right-turn zone from left to right. The starting point of the navigation line in the image information is determined (i.e., the bottom portion of the navigation line in the image information). If it is in the left-turn zone, the robot turns left; if it is in the straight-ahead zone, the robot goes straight; if it is in the right-turn zone, the robot turns right.

[0037] Example 1

[0038] The left-turn zone is divided into several left-turn zones from left to right, and the right-turn zone is divided into several right-turn zones from right to left. The robot's turning radius within each left-turn zone increases from left to right, and the robot's turning radius within each right-turn zone increases from right to left. In other words, the further the guideline deviates, the larger the turning radius required and the smaller the turning radius. Conversely, the smaller the guideline deviates, the smaller the turning radius required and the larger the turning radius. When the guideline is centered in the image, its turning radius approaches infinity, approaching a straight line.

[0039] In this embodiment, the left-turn area is divided into a left small bend, a left middle bend area, and a left large bend area from left to right, and the right-turn area is divided into a right small bend area, a right middle bend area, and a right large bend area from right to left; if the navigation line is in the left large bend area, the robot turns a large left bend; if the navigation line is in the left middle bend area, the robot turns a medium left bend; if the navigation line is in the left small bend area, the robot turns a small left bend; if the navigation line is in the right small bend area, the robot turns a small right bend; if the navigation line is in the right middle bend area, the robot turns a medium right bend; if the navigation line is in the right large bend area, the robot turns a large right bend.

[0040] Specifically, the ratio of the left small curve area, left medium curve area, left large curve area, straight area, right large curve area, right medium curve area, and right small curve area is 110:20:20:20:20:20:110. When the robot is traveling straight, the speeds of the left and right wheels are equal; when the robot makes a small left turn, the speed ratio of the left and right wheels is 1:3; when the robot makes a medium left turn, the speed ratio of the left and right wheels is 3:5; when the robot makes a large left turn, the speed ratio of the left and right wheels is 7:9; when the robot makes a small right turn, the speed ratio of the left and right wheels is 3:1; when the robot makes a medium right turn, the speed ratio of the left and right wheels is 5:3; and when the robot makes a large right turn, the speed ratio of the left and right wheels is 9:7.

[0041] Example 2

[0042] It can be concluded from Example 1 that the finer the turning partitions, the better the turning control, and the better the line patrol effect, which will prevent the robot from swinging left and right along the navigation line.

[0043] Therefore, the robot turns by the difference in left and right wheel speeds:

[0044] V l =V s +V t

[0045] V r =V s -V t

[0046] Among them, V l is the speed of the left wheel, V r is the speed of the right wheel, V s is the robot's straight-line speed, V t To adjust the speed;

[0047] V t =K p *E

[0048] E=X1-X m

[0049] Among them, K p is the proportional adjustment constant, E is the error, X1 is the horizontal coordinate of the starting point of the navigation line, X m is the horizontal coordinate of the midpoint of the image information.

[0050] Likewise Figure 2 As shown, assuming that in this frame, the horizontal coordinate of the starting point of the navigation line in the image information is X1=50, and the horizontal axis coordinate of the midpoint of the image information is X m =100, the error is -50. p The proportional adjustment constant is adjusted according to the actual situation. In this embodiment, it is 0.5, so the adjustment speed V t=-25. Assume the robot's linear speed is 200, then the left wheel speed V in this frame l =175, and the right wheel speed V r =225, thus turning left.

[0051] Example 3

[0052] Through the above adjustment method, the robot's left and right swing along the navigation line has been greatly improved, but there is still a little shaking. The reason is that the greater the speed, the greater the inertia, and the farther the starting point of the navigation line is from the center of the image, the more it adjusts the speed V. t The larger the speed, the greater the inertia. Therefore, it is necessary to adjust the speed V t Make restrictions.

[0053] Similarly, the robot turns by the difference in left and right wheel speed:

[0054] V l =V s +V t

[0055] V r =V s -V t

[0056] Among them, V l is the speed of the left wheel, V r is the speed of the right wheel, V s is the robot's straight-line speed, V t To adjust the speed;

[0057] V t =K p *E n +K d *(E n -E n-1 )

[0058] E n =X n -X m

[0059] E n-1 =X n-1 -X m

[0060] Among them, K p is the proportional adjustment constant, K d is the differential adjustment constant, E n is the error amount, E n-1 is the last error, X n The horizontal coordinate of the starting point of the navigation line in the image information of this frame, X n-1The horizontal coordinate of the starting point of the navigation line in the previous frame image information, X m is the horizontal coordinate of the midpoint of the image information.

[0061] Assume that in this frame, the horizontal coordinate of the starting point of the navigation line in the image information is X n =50, the horizontal axis coordinate X of the midpoint of the image information m =100, the error is -50. p The proportional adjustment constant is adjusted according to the actual situation. In this embodiment, it is 0.5. d The differential adjustment constant also needs to be set and adjusted according to the actual situation. In this embodiment, it is 3. n-1 =51 is the horizontal coordinate of the starting point of the navigation line in the previous frame of image information, then E n-1 =-49 is the last error. t =0.5*-50+3*(-50+49)=-25-3=-28, assuming the robot's linear speed is 200, then the left wheel speed V in this frame l =172, and the right wheel speed V r =228, thus turning left.

[0062] like Figure 3 As shown, the present invention also provides a visual line patrol robot, including a visual line patrol navigation module and a motion module; the motion module moves according to the forward and turning instructions provided by the visual line patrol navigation module; the visual line patrol navigation module includes an image information acquisition module for acquiring image information, and an analysis module for analyzing the position of the navigation line on the image information and determining the robot's travel direction according to the position of the navigation line.

[0063] The above are merely preferred embodiments of the present invention. It should be noted that the above preferred embodiments should not be construed as limiting the present invention, and the scope of protection of the present invention should be determined by the scope defined in the claims. Persons skilled in the art will appreciate that improvements and modifications may be made without departing from the spirit and scope of the present invention, and such improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A robot visual line patrol navigation method, characterized in that include: Setting navigation lines on the ground; Use the camera on the robot to obtain ground image information; Analyze the position of the navigation line on the image information and determine the robot's travel direction based on the position of the navigation line; The analyzing the position of the navigation line on the image information and determining the direction of travel of the robot according to the position of the navigation line includes: dividing the image information from left to right into a left-turning area, a straight-ahead area, and a right-turning area; Determine the starting point of the navigation line in the image information. If it is in the left-turn area, the robot turns left; if it is in the straight-ahead area, the robot goes straight; if it is in the right-turn area, the robot turns right; The robot turns by using the speed difference between the left and right wheels: V l =V s +V t V r =V s -V t Among them, V l is the speed of the left wheel, V r is the speed of the right wheel, V s is the robot's straight-line speed, V t To adjust the speed; V t =K p *E n +K d *(E n -E n-1 ) E n =X n -X m E n-1 =X n-1 -X m Among them, K p is the proportional adjustment constant, K d is the differential adjustment constant, E n is the error amount, E n-1 is the last error, X n The horizontal coordinate of the starting point of the navigation line in the image information of this frame, X n-1 The horizontal coordinate of the starting point of the navigation line in the previous frame image information, X m is the horizontal coordinate of the midpoint of the image information.

2. A robot visual line patrol navigation method according to claim 1, characterized in that: The color of the navigation line contrasts with the ground color.

3. The robot vision line patrol navigation method according to claim 1, characterized in that: The left-turn area is divided into several left rotor areas from left to right, and the right-turn area is divided into several right rotor areas from right to left; the turning radius of the robot in each left rotor area increases from left to right, and the turning radius of the robot in each right rotor area increases from right to left.

4. A robot visual line patrol navigation method according to claim 3, characterized in that: The left-turn area is divided into a left small bend area, a left middle bend area, and a left large bend area from left to right, and the right-turn area is divided into a right small bend area, a right middle bend area, and a right large bend area from right to left; if the navigation line is in the left large bend area, the robot turns a large left bend; if the navigation line is in the left middle bend area, the robot turns a medium left bend; if the navigation line is in the left small bend area, the robot turns a small left bend; if the navigation line is in the right small bend area, the robot turns a small right bend; if the navigation line is in the right middle bend area, the robot turns a medium right bend; if the navigation line is in the right large bend area, the robot turns a large right bend.

5. A robot visual line patrol navigation method according to claim 4, characterized in that: The proportions of the left small curve area, left middle curve area, left large curve area, straight area, right large curve area, right middle curve area, and right small curve area are 110:20:20:20:20:20:20:

110.

6. A robot visual line patrol navigation method according to claim 3, characterized in that: When the robot is moving straight, the speeds of the left and right wheels are equal; when the robot turns left slightly, the speed ratio of the left and right wheels is 1:3; when the robot turns left in a medium turn, the speed ratio of the left and right wheels is 3:5; when the robot turns left in a large turn, the speed ratio of the left and right wheels is 7:9; when the robot turns right slightly, the speed ratio of the left and right wheels is 3:1; when the robot turns right in a medium turn, the speed ratio of the left and right wheels is 5:3; when the robot turns right in a large turn, the speed ratio of the left and right wheels is 9:

7.

7. A visual line patrol robot, used to deploy the robot visual line patrol navigation method according to any one of claims 1 to 6, characterized in that It includes a visual line patrol navigation module and a motion module; the motion module moves according to the forward and turning instructions provided by the visual line patrol navigation module; the visual line patrol navigation module includes an image information acquisition module for acquiring image information, and an analysis module for analyzing the position of the navigation line on the image information and determining the robot's travel direction based on the position of the navigation line.

Images