A control method of a mobile charging robot based on a license plate of a vehicle to be charged

Abstract

The application relates to the field of computer motion control, in particular to a control method of a mobile charging robot based on a license plate of a vehicle to be charged, which comprises the following steps: S100, if s1<=D1, entering S200; S200, obtaining a position adjustment value and a posture adjustment value of the mobile charging robot when switching from a current pose to a target pose according to the license plate in the first image; S300, adjusting the current pose of the mobile charging robot, if the pose of the mobile charging robot after adjustment is the target pose, entering S400; S400, controlling the mobile charging robot to move towards the vehicle to be charged in a direction parallel to the normal vector, and obtaining s2; S500, if s2<=D2, entering S600; S600, continuing to move towards the vehicle to be charged by a third preset distance. The application solves the problem that the mobile charging robot cannot normally charge the vehicle to be charged.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of computer motion control, in particular to a control method of a mobile charging robot based on a license plate of a vehicle to be charged. BACKGROUND

[0002] The existing mobile charging robot needs to identify the vehicle to be charged in the garage, and then relies on the mechanical arm to carry the charging gun to aim at the charging port of the vehicle to be charged for automatic charging. One of the key links is that the chassis of the mobile charging robot needs to be accurately positioned according to the pose of the vehicle to be charged. The existing method is to preset a target position (for example, the midpoint of the end of the parking space close to the charging port) on the parking space where the vehicle to be charged is located, so that the mobile charging robot navigates to the target position, and then performs the subsequent related operations based on the target position. However, the above-mentioned preset target position is based on the premise that the vehicle to be charged is parked at the standard parking angle on the parking space. Since different users have different parking habits, the vehicle to be charged may not be parked at the standard parking angle on the parking space. If the mobile charging robot continues to navigate to the above-mentioned preset target position, the mechanical arm of the mobile charging robot may not be able to stretch normally, resulting in that the mobile charging robot cannot normally charge the vehicle to be charged. SUMMARY

[0003] The present application aims to provide a control method of a mobile charging robot based on a license plate of a vehicle to be charged, which can solve the problem that the mobile charging robot cannot normally charge the vehicle to be charged due to the vehicle to be charged not being parked at the standard parking angle on the parking space.

[0004] According to the present application, a control method of a mobile charging robot based on a license plate of a vehicle to be charged is provided, comprising the following steps:

[0005] S100, obtaining the distance s1 between the mobile charging robot and the preset target position during the movement of the mobile charging robot according to the preset trajectory, if s1≤D1, then entering S200; D1 is a first preset distance, the preset target position is the midpoint of the end of the parking space where the vehicle to be charged is located, the end is the end close to the charging port of the vehicle to be charged, and the end point of the preset trajectory is the preset target position.

[0006] S200, acquire a first image of the vehicle to be charged, and acquire a position adjustment value and a pose adjustment value of the mobile charging robot when the mobile charging robot switches from a current pose to a target pose based on a license plate in the first image; the target pose is a pose of the mobile charging robot that satisfies a first preset condition and a second preset condition, the first preset condition is that a vertical center axis of the mobile charging robot intersects with a normal vector of a license plate of the vehicle to be charged passing through a center of the license plate of the vehicle to be charged, and the second preset condition is that a shell close to a side of the vehicle to be charged of the mobile charging robot is parallel to the license plate of the vehicle to be charged.

[0007] S300, adjust the current pose of the mobile charging robot according to the position adjustment value and the pose adjustment value, and if the pose of the mobile charging robot after adjustment is the target pose, proceed to S400.

[0008] S400, control the mobile charging robot to move towards the vehicle to be charged in a direction parallel to the normal vector, and acquire a distance s2 between the mobile charging robot and the vehicle to be charged during movement.

[0009] S500, if s2≤D2, proceed to S600; D2 is a second preset distance.

[0010] S600, control the mobile charging robot to continue moving towards the vehicle to be charged in a direction parallel to the normal vector by a third preset distance.

[0011] Compared with the prior art, the method provided by the application has obvious beneficial effects, and at least has the following beneficial effects:

[0012] The application acquires an image of the vehicle to be charged at a preset target position of the mobile charging robot, acquires a corresponding position adjustment value and pose adjustment value of the mobile charging robot based on a license plate in the image, corrects the pose of the mobile charging robot, and controls the mobile charging robot to move to a position opposite to the center of the vehicle to be charged and a certain distance away from the vehicle to be charged based on the corrected pose, thereby ensuring the space for the mechanical arm of the mobile charging robot to perform subsequent charging operations, and solving the problem that the mobile charging robot cannot normally charge the vehicle to be charged due to the vehicle to be charged not being parked at a standard parking angle in the parking space. BRIEF DESCRIPTION OF DRAWINGS

[0013] In order to make the technical solutions in the embodiments of the present application clearer, the accompanying drawings needed in the embodiments will be briefly introduced below. Obviously, the accompanying drawings in the following description only represent some of the embodiments of the present application, and all other drawings obtained by those skilled in the art without creative effort based on these drawings also fall within the protection scope of the present application.

[0014] Figure 1 A flowchart of a control method of a mobile charging robot based on a license plate of a vehicle to be charged is provided in the embodiments of the present application. DETAILED DESCRIPTION

[0015] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments only represent some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative effort also fall within the protection scope of the present application.

[0016] According to the present application, as shown in Figure 1 The control method of the mobile charging robot based on the license plate of the vehicle to be charged includes:

[0017] S100, a distance s1 between the mobile charging robot and a preset target position in a movement process of the mobile charging robot according to a preset trajectory is acquired, if s1≤D1, S200 is entered; D1 is a first preset distance, the preset target position is a midpoint of an end of a parking space where the vehicle to be charged is located, the end is an end close to a charging port of the vehicle to be charged, and an end point of the preset trajectory is the preset target position.

[0018] According to the present application, the end of the parking space is a position for parking a head or a tail of the vehicle to be charged. It should be understood that the position of the mobile charging robot is taken as a starting point, the preset target position is taken as an end point, and the preset trajectory can be acquired by using a trajectory planning method; and those skilled in the art know that any trajectory planning method in the prior art falls within the protection scope of the present application.

[0019] Preferably, D1 is a position where a camera of the mobile charging robot can clearly capture the license plate of the vehicle to be charged and the license plate is a complete license plate in an acquired image. Thus, the mobile charging robot can acquire a clear and complete license plate image at a position with a distance of D1 from the preset target position.

[0020] According to the present application, if s1>D1, the distance between the mobile charging robot and the preset target position is continuously acquired.

[0021] S200, acquire a first image of the vehicle to be charged, and acquire the position adjustment value and attitude adjustment value of the mobile charging robot when switching from the current pose to the target pose based on the license plate in the first image; the target pose is the pose of the mobile charging robot that satisfies a first preset condition and a second preset condition, the first preset condition is that the vertical central axis of the mobile charging robot intersects with the normal vector of the license plate of the vehicle to be charged that passes through the center of the license plate of the vehicle to be charged, and the second preset condition is that the shell of the mobile charging robot on the side closer to the vehicle to be charged is parallel to the license plate of the vehicle to be charged.

[0022] According to the present invention, the size and shape of the license plate are relatively fixed and known information about the vehicle to be charged. According to the present invention, obtaining the position adjustment value and attitude adjustment value of the mobile charging robot when switching from the current pose to the target pose based on the license plate in the first image includes:

[0023] S210, using a trained neural network model, obtain the pixel points A = ((x1,y1),(x2,y2),…,(x...) of the license plate in the first image. N ,y N )), (x n ,y n Let x be the nth pixel of the license plate. n Let y be the x-coordinate of the nth pixel of the license plate. n Let be the ordinate of the nth pixel of the license plate, where n ranges from 1 to N, and N is the number of pixels in the license plate.

[0024] According to the present invention, the training process of the neural network model includes: acquiring a set of vehicle image samples with license plates; labeling the license plate regions in each sample; and training the neural network using the labeled samples. The process of training the neural network model is a supervised training process. Those skilled in the art will understand that any prior art method for supervised training of a neural network model falls within the protection scope of the present invention.

[0025] S220, Obtain the x-coordinate of the license plate center in the first image. and ordinate

[0026] S230, Obtain the position adjustment vector x0 is the x-coordinate of the center of the license plate in the image of the vehicle to be charged, obtained when the distance between the mobile charging robot and the preset target position is D1 and the mobile charging robot is in the target pose; y0 is the y-coordinate of the center of the license plate in the image of the vehicle to be charged, obtained when the distance between the mobile charging robot and the preset target position is D1 and the mobile charging robot is in the target pose.

[0027] S240, obtain the position adjustment value of the mobile charging robot when switching from the current pose to the target pose according to (Δx, Δy) and the first coordinate transformation relationship, where the first coordinate transformation relationship is the coordinate transformation relationship from the first image coordinate system to the mobile charging robot coordinate system.

[0028] According to the present invention, the viewing angle of the mobile charging robot's camera in acquiring images of the vehicle to be charged remains unchanged. The position adjustment value of the mobile charging robot when switching from its forward pose to the target pose can be obtained based on the difference between (x,y) and (x0,y0). The methods for constructing the image coordinate system and the mobile charging robot coordinate system, and the methods for obtaining the coordinate transformation relationship between the two coordinate systems, are prior art. Those skilled in the art will understand that any prior art method for constructing a coordinate system and obtaining the coordinate transformation relationship falls within the protection scope of this invention.

[0029] S250, obtain the edge pixels A' = ((x'1,y'1),(x'2,y'2),…,(x'...). M ,y' M )), (x' m ,y' m Let x' be the m-th edge pixel in A. m Let y' be the x-coordinate of the m-th edge pixel in A. m Let be the ordinate of the m-th edge pixel in A, where m ranges from 1 to M, and M is the number of pixels in the license plate.

[0030] According to the present invention, edge pixels in A can be obtained using an edge detection method. Those skilled in the art will understand that any edge detection method in the prior art falls within the protection scope of the present invention.

[0031] S260, using the Hough transform to detect straight lines in A', we obtain L = {l1, l2, ..., l Q}, l q Let q be the q-th line obtained, where q ranges from 1 to Q, and Q is the number of lines obtained.

[0032] According to the present invention, the license plate is a quadrilateral, and L includes at least two vertical lines and two horizontal lines.

[0033] S270, acquire features (b1, b2, b3, b4) of the license plate in the first image, b1 is an included angle of the first straight line and the second straight line, b2 is an included angle of the first straight line and the third straight line, b3 is an included angle of the second straight line and the fourth straight line, and b4 is an included angle of the third straight line and the fourth straight line, the first straight line is a straight line with the largest average longitudinal coordinate in L, the second straight line is a straight line with the smallest average transverse coordinate in L, the third straight line is a straight line with the largest average transverse coordinate in L, and the fourth straight line is a straight line with the smallest average longitudinal coordinate in L.

[0034] According to the application, the first straight line and the fourth straight line are respectively the upper and lower straight lines of the license plate, and the second straight line and the third straight line are respectively the left and right straight lines of the license plate; b1, b2, b3 and b4 are respectively four internal angles corresponding to the license plate.

[0035] S280, match (b1, b2, b3, b4) with each feature in the target feature library, if the similarity of the matched feature and (b1, b2, b3, b4) is greater than a preset similarity threshold, the pose adjustment value corresponding to the feature matched with (b1, b2, b3, b4) in the target feature library is used as the pose adjustment value of the mobile charging robot when switching from the current pose to the target pose; the target stores a plurality of features and the pose adjustment value corresponding to each feature.

[0036] Optionally, the above-mentioned preset similarity threshold is greater than 95%.

[0037] According to the application, the target feature storage library stores the features of the license plate corresponding to the first image acquired when the mobile charging robot is in different poses, and stores the pose adjustment value corresponding to each feature, the feature is the license plate feature corresponding to the same vehicle as the vehicle to be charged, and the pose adjustment value can realize the conversion of the current pose to the target pose corresponding to each feature.

[0038] According to the application, matching (b1, b2, b3, b4) with each feature in the target feature library includes:

[0039] S281, use cosine similarity to obtain the similarity of (b1, b2, b3, b4) and each feature in the target feature library.

[0040] S282, the feature with the largest similarity is used as the feature matched with (b1, b2, b3, b4).

[0041] S300, adjust the current pose of the mobile charging robot according to the position adjustment value and the pose adjustment value, if the pose of the mobile charging robot after adjustment is the target pose, enter S400.

[0042] According to the application, if the pose of the mobile charging robot after adjustment is not the target pose, steps S200 and S300 are repeated until the pose of the mobile charging robot after adjustment is the target pose.

[0043] S400, controlling the mobile charging robot to move towards the vehicle to be charged in a direction parallel to the normal vector, and acquiring a distance s2 between the mobile charging robot and the vehicle to be charged during the movement.

[0044] Optionally, the distance s2 between the mobile charging robot and the vehicle to be charged during the movement is acquired by using a radar sensor, and the distance s2 between the mobile charging robot and the vehicle to be charged during the movement is acquired by using the radar sensor, comprising:

[0045] S410, acquiring a distance S' = {s'1, s'2,..., s'P} between an outer contour of the vehicle to be charged and the vehicle to be charged. P},s' i s'i is a distance between an i-th point on the outer contour of the vehicle to be charged and the vehicle to be charged, i is in a range of 1 to P, and P is a number of points on the outer contour of the vehicle to be charged acquired by using the radar sensor.

[0046] S420, acquiring s2 = min(S'), min() is a minimum value.

[0047] S500, if s2≤D2, entering S600; D2 is a second preset distance.

[0048] Optionally, D2 is about 15 cm.

[0049] According to the application, if s2>D2, the distance between the mobile charging robot and the vehicle to be charged during the movement is continuously acquired by using the radar sensor.

[0050] S600, controlling the mobile charging robot to continue moving towards the vehicle to be charged in a direction parallel to the normal vector by a third preset distance.

[0051] Preferably, the encoder of the mobile charging robot is used to control the mobile charging robot to continue moving towards the vehicle to be charged in a direction parallel to the normal vector by a third preset distance. The encoder of the mobile charging robot can achieve simple and accurate control of the mobile charging robot.

[0052] Preferably, the third preset distance is in a range of 10-12 cm. Thus, the distance between the mobile charging robot and the protruding position of the outer contour of the vehicle to be charged is 3-5 cm, which is more suitable for the mechanical arm of the mobile charging robot to charge the vehicle to be charged.

[0053] The application guarantees the space for the mechanical arm of the mobile charging robot to perform the subsequent charging operation, and solves the problem that the mobile charging robot cannot normally charge the vehicle to be charged due to the vehicle to be charged not being parked at the standard parking angle in the parking space.

[0054] While some specific embodiments of the application have been described in detail by way of example with reference to the accompanying drawings, it should be understood that the examples are for illustrative purposes only and are not intended to limit the scope of the application. Those skilled in the art will understand that various modifications can be made to the embodiments without departing from the scope and spirit of the application. The scope of the application is defined by the appended claims.

Claims

1. A control method of a mobile charging robot based on a license plate of a vehicle to be charged, characterized by, The method comprises the following steps: S100, acquiring a distance s1 between the mobile charging robot and a preset target position during movement of the mobile charging robot according to a preset trajectory, and if s1≤D1, entering S200; D1 is a first preset distance, the preset target position is a midpoint of an end portion of a parking space where a vehicle to be charged is located, the end portion is an end portion close to a charging port of the vehicle to be charged, and a terminal point of the preset trajectory is the preset target position; S200, acquiring a first image of the vehicle to be charged, and acquiring a position adjustment value and a posture adjustment value of the mobile charging robot when the mobile charging robot switches from a current pose to a target pose according to a license plate in the first image; the target pose is a pose of the mobile charging robot that meets first preset conditions and second preset conditions, the first preset conditions are that a vertical central axis of the mobile charging robot intersects with a normal vector of a license plate of the vehicle to be charged passing through a center of the license plate of the vehicle to be charged, and the second preset conditions are that a shell on a side of the mobile charging robot close to the vehicle to be charged is parallel to the license plate of the vehicle to be charged; S300, adjusting the current pose of the mobile charging robot according to the position adjustment value and the posture adjustment value, and if the pose of the mobile charging robot after adjustment is the target pose, entering S400; S400, controlling the mobile charging robot to move towards the vehicle to be charged in a direction parallel to the normal vector, and acquiring a distance s2 between the mobile charging robot and the vehicle to be charged during movement; S500, if s2≤D2, entering S600; D2 is a second preset distance; S600, controlling the mobile charging robot to continue moving towards the vehicle to be charged in the direction parallel to the normal vector by a third preset distance.

2. The method of claim 1, wherein, The position adjustment value and the posture adjustment value of the mobile charging robot when the mobile charging robot switches from the current pose to the target pose according to the license plate in the first image comprise: S210, acquiring pixel points A = ((x1, y1), (x2, y2), …, (xN, yN)) of the license plate in the first image by using the trained neural network model, (x, y) being the nth pixel point of the license plate, x being the horizontal coordinate of the nth pixel point of the license plate, y being the vertical coordinate of the nth pixel point of the license plate, n being in a range of 1 to N, N being the number of pixel points of the license plate. N N n n n n ​​​​​​ S220, obtaining the horizontal coordinate of the center of the license plate in the first image and the vertical coordinate S230, acquire a position adjustment vector x0 is the horizontal coordinate of the center of the license plate in the image of the vehicle to be charged when the mobile charging robot is at the target pose and the distance between the mobile charging robot and the preset target position is D1, and y0 is the vertical coordinate of the center of the license plate in the image of the vehicle to be charged when the mobile charging robot is at the target pose and the distance between the mobile charging robot and the preset target position is D1. S240, acquiring the position adjustment value of the mobile charging robot when the mobile charging robot switches from the current pose to the target pose according to (Δx, Δy) and a first coordinate conversion relationship, and the first coordinate conversion relationship is a coordinate conversion relationship of conversion from a first image coordinate system to a mobile charging robot coordinate system.

3. The method of claim 2, wherein, The position adjustment value and the posture adjustment value of the mobile charging robot when the mobile charging robot switches from the current pose to the target pose according to the license plate in the first image comprise: S250, obtain the edge pixels A' = ((x'1,y'1),(x'2,y'2),…,(x'...). M ,y' M )), (x' m ,y' m Let x' be the m-th edge pixel in A. m Let y' be the x-coordinate of the m-th edge pixel in A. m Let m be the ordinate of the m-th edge pixel in A, where m ranges from 1 to M, and M is the number of pixels in the license plate. S260, using the Hough transform to detect the straight line in A', obtaining L={l1, l2, …, l Q}, l q For the qth straight line obtained, the value range of q is 1 to Q, and Q is the number of obtained straight lines; S270, acquiring features (b1, b2, b3, b4) of the license plate in the first image, b1 is an included angle between a first straight line and a second straight line, b2 is an included angle between the first straight line and a third straight line, b3 is an included angle between the second straight line and a fourth straight line, and b4 is an included angle between the third straight line and the fourth straight line, the first straight line is a straight line with the largest average longitudinal coordinate in L, the second straight line is a straight line with the smallest average transverse coordinate in L, the third straight line is a straight line with the largest average transverse coordinate in L, and the fourth straight line is a straight line with the smallest average longitudinal coordinate in L; S280, match (b1, b2, b3, b4) with each feature in the target feature library, if the similarity between the matched feature and (b1, b2, b3, b4) is greater than a preset similarity threshold, the pose adjustment value corresponding to the feature in the target feature library matched with (b1, b2, b3, b4) is used as the pose adjustment value of the mobile charging robot when switching from the current pose to the target pose.

4. The method of claim 3, wherein, In S280, (b1, b2, b3, b4) is matched with each feature in the target feature library, including: S281, use cosine similarity to obtain the similarity between (b1, b2, b3, b4) and each feature in the target feature library; S282, the feature with the maximum similarity is taken as the feature matched with (b1, b2, b3, b4).

5. The method of claim 1, wherein, In S400, the radar sensor is used to obtain the distance s2 of the mobile charging robot from the vehicle to be charged during the movement.

6. The method of claim 5, wherein, In S400, the distance s2 of the mobile charging robot from the vehicle to be charged during the movement is obtained, including: S410, obtaining a distance S' = {s'1, s'2,..., s'P} of the outer contour of the vehicle to be charged from the vehicle to be charged, wherein s'1, s'2,..., s'P are distances of the first P points on the outer contour of the vehicle to be charged obtained by the radar sensor. P},s' i s' is a distance of an i-th point on the outer contour of the vehicle to be charged obtained by the radar sensor, i is an integer in a range of 1 to P, and P is a number of points on the outer contour of the vehicle to be charged obtained by the radar sensor. S420, obtain s2 = min(S'), min() is the minimum value.

7. The method of claim 5, wherein, In S600, the encoder of the mobile charging robot is used to control the mobile charging robot to continue moving towards the vehicle to be charged in a direction parallel to the normal vector by a third preset distance D3.

Images