Automatic charging method and system and charging robot

Abstract

The invention provides an automatic charging method and system and a charging robot, and the method comprises the steps: controlling a first mechanical arm to grab a charging gun from a charging pile under the condition of receiving a charging request of a to-be-charged vehicle, and controlling the first mechanical arm to carry the charging gun to move towards a first preset pose; meanwhile, a second mechanical arm is controlled to carry the three-dimensional camera to move towards a second preset pose, and under the condition that the three-dimensional camera moves to the second preset pose, point cloud pictures of the charging gun and the charging base are obtained; determining an alignment state of the charging gun and the charging seat based on the point cloud atlas of the charging gun and the charging seat; under the condition that it is determined that the charging gun and the charging base are not aligned, the pose of the first mechanical arm is adjusted; and under the condition that the charging gun is aligned with the charging base, the first mechanical arm is controlled to insert the charging gun into the charging base so as to automatically charge the to-be-charged vehicle. According to the scheme, the efficiency of the charging robot can be remarkably improved, the requirement for the detection precision of the three-dimensional camera is lowered, and the environmental adaptability is enhanced.

Description

TECHNICAL FIELD

[0001] The embodiment of the present application relates to the technical field of automatic charging, in particular to an automatic charging method, system and charging robot. BACKGROUND

[0002] With the rapid development and large-scale application of new energy vehicles, the charging demand of electric vehicles is continuously rising. The traditional electric vehicle charging mode relies on manual operation, and the charging gun needs to be manually inserted into the charging seat of the vehicle to be charged. Not only is the operation process cumbersome, but also the charging seat and the charging gun are prone to be damaged, and there is a certain safety hazard.

[0003] To solve the above technical problems, an automatic device (for example, an automatic charging robot) can be used to complete the charging operation. Generally, the automatic device can adopt a combination scheme of a single mechanical arm + three-dimensional (3D) camera. In the case of ensuring that the mechanical arm is fixedly connected with the charging gun, before plugging and unplugging the charging gun, the charging seat of the vehicle is first detected by the high-precision 3D camera, and after calculating the deviation between the actual pose and the ideal pose of the charging seat, the pose of the robot arm is adjusted according to the deviation to realize the alignment of the charging gun and the charging seat, and the plugging and unplugging operation of the charging gun is completed at one time. However, the automatic charging mode of "one-time alignment and one-time plugging" has the following obvious deficiencies: first, the fixed connection between the arm and the charging gun is usually used to ensure that the pose of the charging gun and the camera is fixed, so that the detection data of the 3D camera can accurately guide the charging gun to complete the alignment and plugging. However, this fixed connection will cause the arm to be unable to separate from the charging gun during the entire charging process, and thus the charging service for other vehicles cannot be provided, greatly reducing the use efficiency of the equipment. Secondly, since the success rate of plugging and unplugging the charging gun is required to be very high, a high-precision 3D camera needs to be equipped to meet the accuracy requirement, which not only significantly increases the cost of the equipment, but also the high-precision 3D camera is usually large in size and heavy in weight, which brings inconvenience to the installation and use of the device. Moreover, the environmental adaptability is poor, and the dependence on the pose deviation is too strong. Once the detection of the pose deviation has an error, it may directly lead to the failure of plugging and unplugging, affecting the smooth progress of the charging process. SUMMARY

[0004] The embodiment of the present application provides an automatic charging method, system and charging robot, which can improve the use efficiency and reduce the cost in view of the problems of low use efficiency, high cost and poor environmental adaptability existing in the existing automatic charging mode of "one-time alignment and one-time plugging".

[0005] To achieve the above purpose, the embodiment of the present application adopts the following technical scheme: In a first aspect, the embodiments of the present application provide an automatic charging method applied to a charging robot, the charging robot comprising a mobile chassis, a first mechanical arm and a second mechanical arm, the second mechanical arm being provided with a three-dimensional camera, the method comprising: in response to receiving a charging request of a vehicle to be charged, controlling the mobile chassis to move the charging robot to a preset working area between the vehicle to be charged and a charging pile, wherein the vehicle to be charged is provided with a charging seat for coupling with a charging gun; in response to the charging robot being in the preset working area, controlling the first mechanical arm to pick up the charging gun from the charging pile and controlling the first mechanical arm to move the charging gun to a first preset pose; simultaneously, controlling the second mechanical arm to move the three-dimensional camera to a second preset pose, and in response to the three-dimensional camera moving to the second preset pose, controlling the three-dimensional camera to capture a point cloud map of the charging gun and the charging seat; determining an alignment state of the charging gun and the charging seat based on the point cloud map of the charging gun and the charging seat; in response to determining that the charging gun and the charging seat are misaligned, adjusting a pose of the first mechanical arm; in response to the charging gun and the charging seat being aligned, controlling the first mechanical arm to insert the charging gun into the charging seat to automatically charge the vehicle to be charged.

[0006] Based on the present scheme, the first mechanical arm picks up the charging gun from the charging pile, the three-dimensional camera provided on the second mechanical arm captures the point cloud map of the charging gun and the charging seat, and in response to the charging gun and the charging seat being aligned, the first mechanical arm is controlled to insert the charging gun into the charging seat. In this way, the operations between the first mechanical arm and the second mechanical arm do not interfere with each other, and after the first mechanical arm inserts the charging gun into the charging seat and returns to the charging pile, it can be released without being bound for a long time, which significantly improves the efficiency of the charging robot. Moreover, the three-dimensional camera simultaneously captures the point cloud map of the charging gun and the charging seat, which can more accurately determine the alignment state of the charging gun and the charging seat compared with the traditional method of detecting only the charging seat, reduces the requirement for the detection accuracy of the three-dimensional camera, and increases convenience while reducing costs. Furthermore, in response to determining that the charging gun and the charging seat are misaligned, the pose of the first mechanical arm is dynamically adjusted, which can gradually align the charging gun and the charging seat, and in response to the charging gun and the charging seat being aligned, the first mechanical arm is controlled to insert the charging gun into the charging seat. Compared with one-time alignment and one-time plugging, this method not only enhances environmental adaptability but also significantly improves the success rate of inserting the charging gun.

[0007] In some embodiments of the present application, determining the alignment state of the charging gun and the charging seat based on the point cloud map of the charging gun and the charging seat comprises: determining a point cloud map to be processed based on the point cloud map of the charging gun and the charging seat; processing the point cloud map to be processed to obtain a first pose deviation of the charging gun and a second pose deviation of the charging seat; in response to the first pose deviation being different from the second pose deviation, determining that the charging gun and the charging seat are misaligned; and in response to the first pose deviation being the same as the second pose deviation, determining that the charging gun and the charging seat are aligned.

[0008] Based on this solution, when the first position deviation of the charging gun can accurately reflect the difference between the actual and ideal positions of the charging gun, and the second position deviation of the charging base can accurately reflect the difference between the actual and ideal positions of the charging base, it can be accurately determined that the charging gun and the charging base are misaligned when the first and second position deviations are different, and it can be accurately determined that the charging gun and the charging base are aligned when the first and second position deviations are the same.

[0009] In some embodiments of this application, the point cloud image to be processed is processed to obtain the first pose deviation of the charging gun and the second pose deviation of the charging base, including: performing point cloud registration processing on the point cloud image to be processed to obtain a first transformation matrix corresponding to the charging gun and a second transformation matrix corresponding to the charging base; determining the first current pose of the charging gun in the robot coordinate system based on the first transformation matrix; determining the first pose deviation of the charging gun based on the first current pose and the first reference pose corresponding to the charging gun; determining the second current pose of the charging base in the robot coordinate system based on the second transformation matrix; and determining the second pose deviation of the charging base based on the second current pose and the second reference pose corresponding to the charging base.

[0010] Based on this scheme, when the first transformation matrix accurately reflects the deviation between the actual pose and the reference pose of the charging gun, the first current pose of the charging gun in the robot coordinate system can be accurately determined based on the first transformation matrix; and the first pose deviation of the charging gun can be accurately determined based on the first current pose and the corresponding first reference pose of the charging gun. When the second transformation matrix accurately reflects the deviation between the actual pose and the reference pose of the charging base, the second current pose of the charging base in the robot coordinate system can be accurately determined based on the second transformation matrix; and the second pose deviation of the charging base can be accurately determined based on the second current pose and the corresponding second reference pose of the charging base.

[0011] In some embodiments of this application, point cloud registration processing is performed on the point cloud image to be processed to obtain a first transformation matrix corresponding to the charging gun, including: preprocessing the point cloud image to be processed to obtain a first point cloud image corresponding to the charging gun; performing coarse matching on the first point cloud image and a first reference image corresponding to the charging gun to obtain a first initial transformation matrix; and performing fine matching on the first point cloud image and the first reference image based on the first initial transformation matrix to obtain a first transformation matrix.

[0012] Based on this scheme, by preprocessing the point cloud image to be processed, noise can be filtered out and the effective point cloud of the charging gun can be segmented, laying a high-quality data foundation for subsequent matching; coarse matching can quickly narrow the spatial search range and generate the first initial transformation matrix, avoiding fine matching from getting trapped in local optima and improving efficiency; fine matching is based on the first initial matrix iterative optimization, and through the elimination of fine difference pairs and the calculation of minimizing spatial distance, the registration accuracy is greatly improved. The first transformation matrix obtained at the end can accurately reflect the deviation between the actual pose of the charging gun and the reference pose.

[0013] In some embodiments of this application, point cloud registration processing is performed on the point cloud image to be processed to obtain a second transformation matrix corresponding to the charging dock. This includes: preprocessing the point cloud image to be processed to obtain a second point cloud image corresponding to the charging dock; performing coarse matching between the second point cloud image and a second reference image corresponding to the charging dock to obtain a second initial transformation matrix; and performing fine matching between the second point cloud image and the second reference image based on the second initial transformation matrix to obtain a second transformation matrix.

[0014] Based on this scheme, by preprocessing the point cloud image to be processed, noise can be filtered out and the effective point cloud of the charging seat can be segmented, laying a high-quality data foundation for subsequent matching; coarse matching can quickly narrow the spatial search range and generate a second initial transformation matrix, avoiding fine matching from getting stuck in local optima and improving efficiency; fine matching is based on the initial matrix iterative optimization, and through the elimination of fine difference pairs and the calculation of minimizing spatial distance, the registration accuracy is greatly improved. The final second transformation matrix can accurately reflect the deviation between the actual pose of the charging seat and the reference pose.

[0015] In some embodiments of this application, determining the first current pose of the charging gun in the robot coordinate system based on the first transformation matrix includes: obtaining the coordinates of the first center point cloud of the charging gun, the normal vector corresponding to the first center point cloud, and the coordinate transformation matrix between the camera coordinate system and the robot coordinate system; determining the pose of the charging gun in the camera coordinate system based on the coordinates of the first center point cloud of the charging gun, the normal vector corresponding to the first center point cloud, and the first transformation matrix; and determining the first current pose of the charging gun in the robot coordinate system based on the pose of the charging gun in the camera coordinate system and the coordinate transformation matrix.

[0016] Based on this scheme, the pose of the charging gun in the camera coordinate system is determined by the coordinates of the first center point cloud (the key part of the charging gun), the normal vector corresponding to the first center point cloud, and the first transformation matrix. Then, the first current pose of the charging gun in the robot coordinate system is determined, which can improve the robustness of the charging gun pose determination in complex scenarios.

[0017] In some embodiments of this application, determining the second current pose of the charging dock in the robot coordinate system based on the second transformation matrix includes: obtaining the coordinates of the second center point cloud of the charging dock, the normal vector corresponding to the second center point cloud, and the coordinate transformation matrix between the camera coordinate system and the robot coordinate system; determining the pose of the charging dock in the camera coordinate system based on the coordinates of the second center point cloud of the charging dock, the normal vector corresponding to the second center point cloud, and the second transformation matrix; and determining the second current pose of the charging dock in the robot coordinate system based on the pose of the charging dock in the camera coordinate system and the coordinate transformation matrix.

[0018] Based on this scheme, the pose of the charging seat in the camera coordinate system is determined by the coordinates of the second center point cloud (the key part of the charging seat), the normal vector corresponding to the second center point cloud, and the second transformation matrix. Then, the second current pose of the charging seat in the robot coordinate system is determined, which can improve the robustness of the charging seat pose determination in complex scenarios.

[0019] In some embodiments of this application, when it is determined that the charging gun and the charging base are not aligned, adjusting the pose of the first robotic arm includes: determining the offset of the first robotic arm based on the first pose deviation and the second pose deviation when it is determined that the charging gun and the charging base are not aligned; and adjusting the pose of the first robotic arm based on the offset of the first robotic arm.

[0020] Based on this solution, the offset of the robotic arm can be determined by the positional deviation of the charging gun and the charging base, and its position can be adjusted. This can accurately pinpoint the root cause of the deviation, realize the "deviation-adjustment" closed-loop control, and quickly reduce the positional deviation between the charging gun and the charging base.

[0021] In some embodiments of this application, after controlling the first robotic arm to insert the charging gun into the charging socket, the method further includes: controlling the first robotic arm to release the charging gun.

[0022] Based on this solution, after the first robotic arm inserts the charging gun into the charging socket, it releases the charging gun. This not only prevents the charging socket from being damaged due to continuous force, but also allows the first robotic arm to flexibly perform other tasks, so as to charge other vehicles.

[0023] In some embodiments of this application, the point cloud image of the charging gun and the charging base is a single point cloud image that includes both the charging gun and the charging base. Determining the point cloud image to be processed based on the point cloud image of the charging gun and the charging base includes: analyzing the integrity of the point cloud image of the charging gun and the charging base to obtain an integrity analysis result; if the charging gun or the charging base is incomplete in the point cloud image of the charging gun and the charging base, based on the integrity analysis result, controlling the second robotic arm to adjust its pose, and controlling the 3D camera to recapture the point cloud image of the charging gun and the charging base when the second robotic arm moves to the target shooting pose; if both the charging gun and the charging base are complete in the recaptured point cloud image of the charging gun and the charging base, the recaptured point cloud image of the charging gun and the charging base is used as the point cloud image to be processed.

[0024] Based on this solution, the integrity of the point cloud image is first analyzed, and then the robotic arm is controlled to adjust the shooting pose and recapture for incomplete cases until a complete point cloud image is obtained as the point cloud image to be processed. This process can actively identify and compensate for point cloud defects caused by occlusion, viewing angle deviation, etc., ensuring that the point cloud image to be processed contains complete three-dimensional features of the charging gun and charging base, providing a reliable data foundation for subsequent pose calculation, alignment judgment, and other stages. At the same time, this closed-loop mechanism of dynamically adjusting the shooting pose enhances the system's adaptability to complex environments, reduces subsequent processing errors caused by incomplete initial point clouds, and thus improves the stability and success rate of the entire charging alignment process.

[0025] Secondly, embodiments of this application provide a charging robot, which includes a mobile chassis, a first robotic arm, a second robotic arm, a three-dimensional camera and a controller mounted on the second robotic arm; A mobile chassis is used to move the charging robot to a preset working area between the vehicle to be charged and the charging pile when a charging request is received from a vehicle to be charged. The first robotic arm is used to grab the charging gun from the charging pile, carry the charging gun to the first preset position, and insert the charging gun into the charging socket after aligning the charging gun with the charging socket of the vehicle to be charged to achieve automatic charging. The second robotic arm is used to carry the 3D camera to a second preset pose. A 3D camera is used to capture point cloud images of the charging gun and charging base when the device is moved to a second preset pose. A controller is used to execute the automatic charging method provided in the first aspect embodiment described above.

[0026] Thirdly, embodiments of this application provide an automatic charging system, which includes a charging pile and a charging robot. The charging pile is used to charge the vehicle when the charging gun is inserted into the charging socket on the vehicle to be charged.

[0027] Fourthly, embodiments of this application provide an automatic charging device, which includes: The first control module is used to control the mobile chassis of the charging robot to move the charging robot to a preset working area between the vehicle to be charged and the charging pile when a charging request is received from the vehicle to be charged; wherein the vehicle to be charged is equipped with a charging socket for coupling with the charging gun. The second control module is used to control the first robotic arm of the charging robot to grab the charging gun from the charging pile when the charging robot is in the preset working area, and to control the first robotic arm of the charging robot to move with the charging gun to the first preset pose; at the same time, it controls the second robotic arm of the charging robot to move with the 3D camera to the second preset pose, and when the 3D camera moves to the second preset pose, it controls the 3D camera to capture the point cloud map of the charging gun and the charging base. The first determining module is used to determine the alignment state of the charging gun and the charging base based on the point cloud map of the charging gun and the charging base. The adjustment module is used to adjust the pose of the first robotic arm of the charging robot when it is determined that the charging gun and the charging base are not aligned. The third control module is used to control the first robotic arm of the charging robot to insert the charging gun into the charging socket when the charging gun is aligned with the charging socket, so as to automatically charge the vehicle to be charged.

[0028] This application provides a storage medium storing a computer program for executing the automatic charging method provided in the first aspect above.

[0029] Fifthly, embodiments of this application provide a computer program product that, when instructions in the computer program product are executed by a processor, performs the automatic charging method provided in the first aspect above. Attached Figure Description

[0030] Figure 1 This is a schematic diagram of a charging robot provided in an embodiment of this application.

[0031] Figure 2 This is a schematic flowchart of an automatic charging method provided in an embodiment of this application.

[0032] Figure 3 This is a schematic flowchart of another automatic charging method provided in an embodiment of this application.

[0033] Figure 4 This is a schematic flowchart of another automatic charging method provided in an embodiment of this application.

[0034] Figure 5 This is a schematic flowchart of another automatic charging method provided in an embodiment of this application.

[0035] Figure 6 This is a schematic flowchart of another automatic charging method provided in an embodiment of this application.

[0036] Figure 7 This is a schematic flowchart of another automatic charging method provided in an embodiment of this application.

[0037] Figure 8A This is a schematic flowchart of another automatic charging method provided in an embodiment of this application.

[0038] Figure 8B This is a schematic flowchart of another automatic charging method provided in an embodiment of this application.

[0039] Figure 9 This is a schematic diagram of an automatic charging device provided in an embodiment of this application.

[0040] Figure 10 This is a structural schematic diagram of a charging robot provided in an embodiment of this application.

[0041] Figure 11 This is a schematic diagram of an automatic charging system provided in an embodiment of this application. Detailed Implementation

[0042] The technical solutions of the embodiments of this application will now be described with reference to the accompanying drawings. To facilitate a clear description of the technical solutions of the embodiments of this application, the use of terms such as "first," "second," etc., in the embodiments of this application is for illustrative purposes and to distinguish the objects being described. There is no particular order between them, nor does it indicate a specific limitation on the number of devices in the embodiments of this application, and they do not constitute any limitation on the embodiments of this application.

[0043] The following describes the relevant technical terms used in the embodiments of this application: Charging stations, the core energy replenishment equipment for electric vehicles, function to safely and efficiently transmit power from the grid to the electric vehicle's battery. They can be classified according to current type as DC charging stations and AC charging stations, and according to installation method as wall-mounted, pole-mounted, and mobile charging stations.

[0044] A charging gun is used to connect a charging station to an electric vehicle, enabling the transfer of electricity from the power source to the vehicle's battery; it is often referred to as a charging plug or charging device, and is inserted into the vehicle's charging port to charge the power battery.

[0045] Automatic charging robots are used for the automatic charging of new energy vehicles. The core of the robot is to replace the human in automatically plugging and unplugging the charging gun. The robot needs to use visual detection to locate the deviation, and then insert and pull out the charging gun through a robotic arm.

[0046] A 3D camera is an imaging device that can capture three-dimensional spatial information of a scene. Through specific technologies (such as structured light, time-of-flight, and binocular vision), it can not only acquire two-dimensional planar images of objects, but also simultaneously calculate the depth information corresponding to each pixel in the image, thereby constructing three-dimensional point cloud data containing the object's length, width, height, and spatial position relationship, providing accurate spatial data support for applications such as 3D reconstruction, object recognition, and pose detection.

[0047] This application provides an automatic charging method. Upon receiving a charging request from a vehicle to be charged, a mobile chassis is controlled to move a charging robot to a preset working area between the vehicle and a charging pile. The vehicle to be charged has a charging socket for coupling with a charging gun. When the charging robot is in the preset working area, a first robotic arm is controlled to grab the charging gun from the charging pile and move it towards a first preset pose. Simultaneously, a second robotic arm is controlled to move it towards a second preset pose with a 3D camera. When the 3D camera reaches the second preset pose, it captures point cloud images of the charging gun and the charging socket. Based on the point cloud images, the alignment of the charging gun and the charging socket is determined. If the charging gun and the charging socket are not aligned, the pose of the first robotic arm is adjusted. When the charging gun and the charging socket are aligned, the first robotic arm inserts the charging gun into the charging socket to automatically charge the vehicle.

[0048] This automated charging method uses a first robotic arm to pick up the charging gun from the charging station, and a 3D camera mounted on a second robotic arm to capture point cloud images of the charging gun and charging base. Once the charging gun and charging base are aligned, the first robotic arm is controlled to insert the charging gun into the charging base. In this way, the operations of the first and second robotic arms do not interfere with each other, and the first robotic arm can be released after inserting the charging gun into the charging base and returning it to the charging station, eliminating the need for long-term binding and significantly improving the efficiency of the charging robot. Furthermore, the 3D camera simultaneously captures point cloud images of both the charging gun and charging base, which, compared to traditional methods that only detect the charging base, can more accurately determine the alignment status of the charging gun and charging base, reducing the requirements for the accuracy of the 3D camera detection and increasing convenience while reducing costs. Moreover, if the charging gun and charging base are determined to be misaligned, the posture of the first robotic arm can be dynamically adjusted to gradually align the charging gun and charging base. The first robotic arm is only controlled to insert the charging gun into the charging base once alignment is achieved. Compared to a one-time alignment and insertion / removal, this not only enhances environmental adaptability but also significantly improves the success rate of charging gun insertion.

[0049] Figure 1 This is a schematic diagram of a charging scenario for a charging robot provided in an embodiment of this application. Figure 1As shown, the charging robot 10 includes a movable base 101, a left arm 102, a right arm 103, and a 3D camera 104 mounted on the right arm 103. The movable base 101 can move the entire charging robot 10. The left arm 102 and the right arm 103 can move independently. The left arm 102 grasps the charging gun 11 through an end effector and adjusts the pose of the left arm 102 grasping the charging gun 11 using the point cloud map captured by the 3D camera 104 on the right arm 103, so that the charging gun 11 is aligned with the vehicle's charging dock. When the charging gun 11 is aligned with the vehicle's charging dock 12, the charging gun 11 is inserted into the vehicle's charging dock.

[0050] Figure 2 This is a schematic flowchart illustrating an automatic charging method provided in an embodiment of this application. This automatic charging method can be applied to, for example... Figure 1 The controller in the charging robot 10 shown. The first robotic arm corresponds to... Figure 1 The left arm 102 in the middle, the second robotic arm corresponds to Figure 1 The right arm is 103. (As shown in the image) Figure 2 As shown, the automatic charging method may include the following steps 201 to 205.

[0051] Step 201: Upon receiving a charging request from a vehicle to be charged, control the mobile chassis to move the charging robot to the preset working area between the vehicle to be charged and the charging pile.

[0052] The vehicle to be charged is equipped with a charging socket for coupling with the charging gun.

[0053] A vehicle awaiting charging can refer to a vehicle that has entered a dedicated charging parking space near a charging station and requires recharging. In some examples, the vehicle awaiting charging may include electric vehicles, electric trucks, electric buses, and other electrified vehicles. This application does not limit the specific type of vehicle awaiting charging; however, this application uses an electric vehicle as an example for illustrative purposes.

[0054] The preset work area refers to a specific space area that is pre-planned and set in the working scenario of the charging robot, located between the vehicle to be charged and the charging pile. It is the "dedicated workstation" for the charging robot to carry out charging operations.

[0055] In some examples, the charging request for a vehicle to be charged can be a charging request that the user manually enters through a mobile application or in-vehicle terminal, including information such as the vehicle model. This charging request is then forwarded to the corresponding charging robot via a cloud platform or local server.

[0056] In other examples, the charging robot can monitor the vehicle entry status of a dedicated charging parking space in real time through sensors deployed in the space or by linking with a parking management system. Once it detects that a vehicle has parked, it sends a charging request confirmation to the user and confirms receipt of the charging request in response to the user's confirmation.

[0057] In other examples, the vehicle to be charged can establish a wireless communication connection between its onboard system and the control system of the charging robot, and directly send a charging request to the charging robot.

[0058] This application does not limit the method of transmitting the charging request of the vehicle to be charged in this embodiment. This application uses the method of transmitting the charging request through the user terminal command as an example for illustrative explanation.

[0059] Taking the method of transmitting charging requests via user-end commands as an example, the charging robot can generate movement commands for its mobile chassis upon receiving a charging request forwarded via a cloud platform or local server. The charging robot will then move to a preset work area along a pre-defined path.

[0060] Step 202: When the charging robot is in the preset working area, control the first robotic arm to grab the charging gun from the charging pile and control the first robotic arm to move towards the first preset pose with the charging gun; at the same time, control the second robotic arm to move towards the second preset pose with the 3D camera, and when the 3D camera moves to the second preset pose, control the 3D camera to capture the point cloud map of the charging gun and the charging base.

[0061] The first preset pose can be the initial position of the first robotic arm, which is convenient for aligning with the charging dock of the vehicle to be charged.

[0062] The second preset pose can be a preset angle that allows the camera to clearly capture the entire view of the charging gun and the charging dock of the vehicle to be charged.

[0063] The point cloud map to be processed can be a clear 3D point cloud map including the complete charging gun and the vehicle being charged.

[0064] The charging robot can stop moving once it reaches a preset working area and generate a first motion command for the first robotic arm and a second motion command for the second robotic arm. This drives the joint movements of the first robotic arm (such as translation and rotation) to grasp the charging gun from the charging pile and move it towards a first preset pose. Simultaneously, it drives the joint movements of the second robotic arm (such as translation and rotation) to move it towards a second preset pose, carrying the 3D camera. Upon detecting that the 3D camera has moved to the second preset pose, the charging robot can send a start-shooting command to the 3D camera, allowing the 3D camera to respond and capture a 3D point cloud image of the charging gun and charging base.

[0065] Step 203: Based on the point cloud map of the charging gun and the charging base, determine the alignment status of the charging gun and the charging base.

[0066] The alignment status of the charging gun and charging base includes both misalignment and alignment. Misalignment refers to a situation where there is a positional deviation between the charging gun and the charging base, preventing a smooth and accurate physical connection. Alignment refers to a situation where the charging gun and charging base meet spatial matching requirements, enabling a safe and smooth physical connection. Meeting spatial matching requirements can mean that the positions of the charging gun and the charging base are completely aligned or the deviation is within acceptable limits.

[0067] The charging robot can directly process the point cloud images of the charging gun and charging base, and determine the alignment state of the charging gun and charging base based on the processing result. Alternatively, it can first filter out the point cloud images to be processed from the point cloud images of the charging gun and charging base, then perform targeted processing on the point cloud images to be processed, and finally determine the alignment state of the charging gun and charging base based on the processing result. This application embodiment does not limit the specific implementation method of determining the alignment state of the charging gun and charging base based on the point cloud images of the charging gun and charging base. This application embodiment uses the example of first filtering out the point cloud images to be processed from the point cloud images of the charging gun and charging base, then performing targeted processing on the point cloud images to be processed, and finally determining the alignment state of the charging gun and charging base based on the processing result to illustrate the process.

[0068] Step 204: If it is determined that the charging gun and the charging base are not aligned, adjust the position of the first robotic arm.

[0069] When the charging robot determines that the charging gun and the charging base are not aligned, it can generate a pose adjustment command for the first robotic arm and change the spatial pose (including position and orientation) of the charging gun by driving the joint movement of the first robotic arm.

[0070] Step 205: With the charging gun aligned with the charging base, control the first robotic arm to insert the charging gun into the charging base for automatic charging of the vehicle to be charged.

[0071] Once the charging gun and the charging base are aligned, the charging robot generates a charging gun insertion command for the first robotic arm. By driving the joint movement of the first robotic arm, the charging gun is inserted into the charging base to charge the vehicle to be charged.

[0072] The automatic charging method provided in this application involves a first robotic arm grasping a charging gun from a charging station, and a 3D camera mounted on a second robotic arm capturing point cloud images of the charging gun and charging base. Once the charging gun and charging base are aligned, the first robotic arm inserts the charging gun into the charging base. Thus, the operations of the first and second robotic arms do not interfere with each other, and the first robotic arm can be released after inserting the charging gun into the charging base and returning it to its position on the charging station, eliminating the need for long-term binding and significantly improving the efficiency of the charging robot. Furthermore, the 3D camera simultaneously captures point cloud images of both the charging gun and charging base, which, compared to traditional methods that only detect the charging base, allows for more accurate determination of the alignment status between the charging gun and charging base, reducing the requirements for the accuracy of the 3D camera detection and increasing convenience while lowering costs. Furthermore, when it is determined that the charging gun and the charging base are not aligned, the position of the first robotic arm can be dynamically adjusted to gradually align the charging gun and the charging base. Only when the charging gun and the charging base are aligned can the first robotic arm be controlled to insert the charging gun into the charging base. Compared with one-time alignment and one-time insertion and removal, this not only enhances environmental adaptability but also significantly improves the success rate of charging gun insertion.

[0073] like Figure 3 As shown above, in the above Figure 2 Based on the embodiment shown, step 203 determines the alignment state of the charging gun and the charging base based on the point cloud map of the charging gun and the charging base, which may include the following steps 2031 to 2034.

[0074] Step 2031: Based on the point cloud maps of the charging gun and charging dock, determine the point cloud map to be processed.

[0075] For example, the charging robot can filter the original point cloud images of the charging gun and the charging base according to the point cloud integrity requirements. If the original point cloud image is determined not to meet the integrity requirements after filtering, the re-acquisition process is triggered to continuously collect the point cloud images of the two and repeat the filtering until a point cloud image that meets the point cloud integrity requirements is obtained, and it is used as the point cloud image to be processed.

[0076] Step 2032: Process the point cloud image to be processed to obtain the first pose deviation of the charging gun and the second pose deviation of the charging base.

[0077] The first pose deviation can be the deviation between the current pose of the charging gun and its reference pose. The second pose deviation can be the deviation between the current pose of the charging dock and its reference pose. The reference pose of the charging dock refers to its pose under ideal conditions (when the vehicle is parked in a designated charging parking space with a preset pose). The reference pose of the charging gun refers to its pose under ideal conditions (when the charging dock is in its reference pose), ensuring alignment with the charging dock.

[0078] In some examples, the charging robot can perform point cloud registration, matrix transformation, and other processing on the point cloud map to be processed, so as to obtain the first pose deviation of the charging gun and the second pose deviation of the charging base.

[0079] Step 2033: If the first position deviation and the second position deviation are different, determine that the charging gun and the charging base are not aligned.

[0080] The charging robot can determine that the charging gun and charging base are misaligned if the difference between the first and second posture deviations is outside a preset deviation range. The preset deviation range can be determined comprehensively based on the hardware characteristics of the charging system, the required docking accuracy, and the actual application scenario. This application does not limit the size of the preset deviation range; however, this embodiment uses a preset deviation range of ±1mm to ±5mm for positional deviation and ±0.5° to ±2° for angular deviation as an example for illustrative purposes.

[0081] Step 2034: If the first and second positional deviations are the same, determine that the charging gun and the charging base are aligned.

[0082] The charging robot can determine that the charging gun and the charging base are aligned when the difference between the first and second posture deviations is within a preset deviation range.

[0083] The automatic charging method provided in this application embodiment can accurately determine that the charging gun and the charging base are misaligned when the first posture deviation of the charging gun can accurately reflect the difference between the actual posture and the ideal posture of the charging gun, and the second posture deviation of the charging base can accurately reflect the difference between the actual posture and the ideal posture of the charging base. When the first posture deviation and the second posture deviation are different, it can accurately determine that the charging gun and the charging base are aligned.

[0084] like Figure 4 As shown above, in the above Figure 3 Based on the embodiment shown, step 2032 processes the point cloud image to be processed to obtain the first pose deviation of the charging gun and the second pose deviation of the charging base, which may include the following steps 401 to 405.

[0085] Step 401: Perform point cloud registration processing on the point cloud map to be processed to obtain the first transformation matrix corresponding to the charging gun and the second transformation matrix corresponding to the charging base.

[0086] Point cloud registration is the process of precisely aligning two sets of point cloud data from different viewpoints in the same coordinate system by calculating spatial transformations (such as translation, rotation, and scaling). In some examples, point cloud registration may include preprocessing, coarse matching, and fine matching.

[0087] The first transformation matrix can be a matrix describing the spatial transformation relationship between the charging gun point cloud map and the charging gun reference map in the point cloud map to be processed. The charging gun point cloud map refers to the point cloud map obtained by extracting charging gun features from the point cloud map to be processed and performing data simplification and noise reduction. The charging gun reference map refers to the point cloud map of the charging gun under ideal conditions.

[0088] The second transformation matrix can be a matrix describing the spatial transformation relationship between the charging dock point cloud map and the charging dock reference map in the point cloud map to be processed. The charging dock point cloud map refers to the point cloud map obtained by extracting the charging dock features from the point cloud map to be processed and performing data simplification and noise reduction. The charging dock reference map refers to the point cloud map of the charging dock under ideal conditions.

[0089] In some examples, both the first transformation matrix and the second transformation matrix can be 4. A matrix of size 4.

[0090] Step 402: Based on the first transformation matrix, determine the first current pose of the charging gun in the robot coordinate system.

[0091] The robot coordinate system is a reference coordinate system that describes the spatial position of the robot itself and its surrounding environment. The zero point of the robot coordinate system is the fixed position of the robot arm's base. The X-axis points "forward" towards the robot, which is the main working direction when the robot arm is extended, such as the direction facing the work area; the Y-axis is perpendicular to the X-axis and points "to the side" of the robot, such as the right side; the Z-axis is perpendicular to both the X and Y axes, points "up," and is perpendicular to the ground or mounting surface, representing the robot's height.

[0092] In some examples, the zero point of the robot coordinate system is the geometric center of the robot arm base and the mounting surface (such as the ground or workbench). In other examples, the zero point of the robot coordinate system is a distinctive mechanical interface of the base (such as the center of the base flange). This application does not limit the specific location of the zero point of the robot coordinate system; this application uses the example of the robot arm base and the mounting surface as the zero point of the robot coordinate system for illustrative purposes.

[0093] The charging robot can process the point cloud representing the position and orientation of the charging gun based on the first transformation matrix to obtain the first current pose of the charging gun in the robot coordinate system.

[0094] Step 403: Determine the first pose deviation of the charging gun based on the first current pose and the first reference pose corresponding to the charging gun.

[0095] The first reference pose can correspond to the reference pose of the charging gun. The charging robot can use the difference between the first current pose and the first reference pose corresponding to the charging gun as the first pose deviation of the charging gun.

[0096] Step 404: Based on the second transformation matrix, determine the second current pose of the charging base in the robot coordinate system.

[0097] The charging robot can process the point cloud representing the position and orientation of the charging seat based on the second transformation matrix to obtain the second current pose of the charging seat in the robot coordinate system.

[0098] Step 405: Determine the second pose deviation of the charging stand based on the second current pose and the second reference pose corresponding to the charging stand.

[0099] The second reference pose can correspond to the reference pose of the charging dock. The charging robot can use the difference between the second current pose and the second reference pose corresponding to the charging dock as the second pose deviation of the charging dock.

[0100] The automatic charging method provided in this application, when the first transformation matrix accurately reflects the deviation between the actual pose and the reference pose of the charging gun, can accurately determine the first current pose of the charging gun in the robot coordinate system based on the first transformation matrix; and can accurately determine the first pose deviation of the charging gun based on the first current pose and the first reference pose corresponding to the charging gun. When the second transformation matrix accurately reflects the deviation between the actual pose and the reference pose of the charging base, can accurately determine the second current pose of the charging base in the robot coordinate system based on the second transformation matrix; and can accurately determine the second pose deviation of the charging base based on the second current pose and the second reference pose corresponding to the charging base.

[0101] like Figure 5 As shown above, in the above Figure 4 Based on the embodiment shown, step 401 performs point cloud registration processing on the point cloud image to be processed to obtain a first transformation matrix corresponding to the charging gun and a second transformation matrix corresponding to the charging base, which may include the following steps 4011 to 4016.

[0102] Step 4011: Preprocess the point cloud image to be processed to obtain the first point cloud image corresponding to the charging gun.

[0103] The charging robot first performs a pass-through filter on the point cloud image to be processed, removing point cloud data other than that of the charging gun. Next, it performs voxel grid downsampling on the obtained initial charging gun point cloud image, simplifying the data volume while retaining important features. Finally, it removes outlier point clouds from the simplified initial charging gun point cloud image to obtain the first point cloud image. This reduces interference for subsequent matching steps, improving matching efficiency and accuracy to a certain extent.

[0104] Step 4012: Perform coarse matching between the first point cloud map and the first reference map corresponding to the charging gun to obtain the first initial transformation matrix.

[0105] The first reference map corresponds to the aforementioned charging gun reference map. Coarse matching refers to the process of determining the approximate spatial transformation relationship between the first point cloud map and the first reference map. It is used to narrow down the global search range, eliminate obvious misalignments, provide effective initial values ​​for fine matching, and avoid fine matching from getting trapped in local optima.

[0106] The charging robot first extracts intrinsic shape signatures (ISS) key points from the first point cloud map, including key points such as contours, cubes, faces, points on faces, intersections, and turning points, to obtain a key point cloud map. Next, it calculates the fast point feature histogram (FPFH) of each key point cloud in the key point cloud map. Then, it performs sample consensus initial alignment (SAC-IA) between the FPFH features of each key point cloud and the corresponding FPFH features of each reference point cloud in the first reference map, obtaining the first initial transformation matrix.

[0107] Step 4013: Perform fine matching on the first point cloud map and the first reference map based on the first initial transformation matrix to obtain the first transformation matrix.

[0108] Fine matching refers to the process of iteratively adjusting transformation parameters using a high-precision optimization algorithm based on the approximate spatial transformation relationship obtained from coarse matching. This minimizes the geometric error between the point cloud of the charging gun or / and charging base and the point cloud of the standard template, ultimately obtaining a precise spatial pose relationship that meets the charging alignment accuracy requirements. It also eliminates residual errors from coarse matching, ensuring that the alignment deviation between the charging gun and the charging base is within the pluggable threshold. In some examples, fine matching requires higher accuracy than coarse matching, but its matching speed is lower.

[0109] The charging robot can start with the coarse matching result and then iteratively optimize it using the iterative closest point (ICP) algorithm to accurately align the point cloud to be registered with the reference point cloud, ultimately obtaining the first transformation matrix (the precise spatial transformation relationship between the two), so that the overlapping areas of the two sets of point clouds achieve high-precision alignment.

[0110] Step 4014: Preprocess the point cloud image to be processed to obtain a second point cloud image corresponding to the charging dock.

[0111] The implementation of step 4014 is similar to that of step 4011, except that the object retained by filtering in step 4011 is the charging gun, while the object retained by filtering in step 4014 is the charging dock. The specific implementation details of this application will not be repeated here; please refer to the description of step 4011 for further information.

[0112] Step 4015: Perform coarse matching between the second point cloud map and the second reference map corresponding to the charging dock to obtain the second initial transformation matrix.

[0113] The second reference diagram can correspond to the aforementioned charging dock reference diagram. The implementation method of step 4015 is similar to that of step 4012, and will not be described again here in this embodiment.

[0114] Step 4016: Perform fine matching on the second point cloud map and the second reference map based on the second initial transformation matrix to obtain the second transformation matrix.

[0115] The implementation of step 4016 is similar to that of step 4013, and will not be described again in this embodiment.

[0116] The automatic charging method provided in this application preprocesses the point cloud image to be processed, which can filter out noise and segment the effective point cloud of the charging gun, laying a high-quality data foundation for subsequent matching. Coarse matching can quickly narrow the spatial search range and generate a first initial transformation matrix, avoiding fine matching from getting trapped in local optima and improving efficiency. Fine matching is based on the iterative optimization of the first initial matrix, and significantly improves the registration accuracy through the elimination of fine difference pairs and the minimization of spatial distance. The final first transformation matrix can accurately reflect the deviation between the actual pose of the charging gun and the reference pose. Preprocessing the point cloud image to be processed can filter out noise and segment the effective point cloud of the charging base, laying a high-quality data foundation for subsequent matching. Coarse matching can quickly narrow the spatial search range and generate a second initial transformation matrix, avoiding fine matching from getting trapped in local optima and improving efficiency. Fine matching is based on the iterative optimization of the initial matrix, and significantly improves the registration accuracy through the elimination of fine difference pairs and the minimization of spatial distance. The final second transformation matrix can accurately reflect the deviation between the actual pose of the charging base and the reference pose.

[0117] like Figure 6 As shown above, in the above Figure 4 Based on the embodiment shown, step 402 determines the first current pose of the charging gun in the robot coordinate system based on the first transformation matrix, which may include the following steps 4021 to 4023.

[0118] Step 4021: Obtain the coordinates of the first center point cloud of the charging gun, the normal vector corresponding to the first center point cloud, and the coordinate transformation matrix between the camera coordinate system and the robot coordinate system.

[0119] The first center point cloud of the charging gun corresponds to the center point of the charging gun. In some examples, the coordinates of the first center point cloud can be (x1, y1, z1), and the corresponding normal vector can be (ΔRx1, ΔRy1, ΔRz1). Here, ΔRx1, ΔRy1, and ΔRz1 represent the roll angle, pitch angle, and yaw angle, respectively.

[0120] The camera coordinate system is a reference system that describes the position of an object in three-dimensional space relative to the camera. The origin of the camera coordinate system is the optical center of the camera, that is, the image-side principal point of the lens. The X-axis is horizontal to the right, parallel to the horizontal direction of the image plane, pointing to the right side of the camera; the Y-axis is vertically upward, parallel to the vertical direction of the image plane, pointing upward to the camera; the Z-axis is along the direction of the camera's optical axis, pointing in front of the camera, that is, the direction of the field of view during shooting, also known as the "depth direction".

[0121] The coordinate transformation matrix between the camera coordinate system and the robot coordinate system is a key parameter describing the spatial position and orientation relationship between the two 3D coordinate systems. In some examples, the coordinate transformation matrix between the camera coordinate system and the robot coordinate system can be 4. The homogeneous transformation matrix of 4.

[0122] Step 4022: Determine the pose of the charging gun in the camera coordinate system based on the coordinates of the first center point cloud of the charging gun, the normal vector corresponding to the first center point cloud, and the first transformation matrix.

[0123] The pose of the charging gun in the camera coordinate system can include six dimensions. In some examples, the pose of the charging gun in the camera coordinate system can be (x, y, z, ΔRx, ΔRy, ΔRz). Since the direction in which the charging gun is inserted is the X direction in the robot coordinate system, the five dimensions (y, z, ΔRx, ΔRy, ΔRz) are mainly considered when calculating the pose deviation.

[0124] In some examples, the charging robot can use the product of the coordinates of the first center point cloud of the charging gun, the normal vector corresponding to the first center point cloud, and the first transformation matrix as the pose of the charging gun in the camera coordinate system.

[0125] Step 4023: Based on the pose and coordinate transformation matrix of the charging gun in the camera coordinate system, determine the first current pose of the charging gun in the robot coordinate system.

[0126] For example, the product of the charging gun's pose in the camera coordinate system and the coordinate transformation matrix can be used as the charging gun's first current pose in the robot coordinate system.

[0127] The automatic charging method provided in this application determines the pose of the charging gun in the camera coordinate system by using the coordinates of the first center point cloud (a key part of the charging gun), the normal vector corresponding to the first center point cloud, and the first transformation matrix. This determines the first current pose of the charging gun in the robot coordinate system, which can improve the robustness of the charging gun pose determination in complex scenarios.

[0128] like Figure 7 As shown above, in the above Figure 4 Based on the embodiment shown, step 404 determines the second current pose of the charging base in the robot coordinate system based on the second transformation matrix, which may include the following steps 4041 to 4043.

[0129] Step 4041: Obtain the coordinates of the second center point cloud of the charging dock, the normal vector corresponding to the second center point cloud, and the coordinate transformation matrix between the camera coordinate system and the robot coordinate system.

[0130] The second center point cloud of the charging dock corresponds to the center point of the charging dock. In some examples, the coordinates of the second center point cloud can be (x2, y2, z2), and the normal vector corresponding to the first center point cloud can be (ΔRx2, ΔRy2, ΔRz2).

[0131] Step 4042: Determine the pose of the charging stand in the camera coordinate system based on the coordinates of the second center point cloud of the charging stand, the normal vector corresponding to the second center point cloud, and the second transformation matrix.

[0132] The implementation of step 4042 is similar to that of step 4022, and will not be described again in this embodiment.

[0133] Step 4043: Based on the pose and coordinate transformation matrix of the charging base in the camera coordinate system, determine the second current pose of the charging base in the robot coordinate system.

[0134] The implementation of step 4043 is similar to that of step 4023, and will not be described again in this embodiment.

[0135] The automatic charging method provided in this application determines the pose of the charging seat in the camera coordinate system by using the coordinates of the second center point cloud (a key part of the charging seat), the normal vector corresponding to the second center point cloud, and the second transformation matrix. This determines the second current pose of the charging seat in the robot coordinate system, which can improve the robustness of the charging seat pose determination in complex scenarios.

[0136] like Figure 8A As shown above, in the above Figure 2 Based on the embodiment shown, step 204, in the case that the charging gun and the charging base are not aligned, adjusts the pose of the first robotic arm, which may include the following steps 2041 and 2042.

[0137] Step 2041: If it is determined that the charging gun and the charging base are not aligned, determine the offset of the first robotic arm based on the first posture deviation and the second posture deviation.

[0138] The offset of the first robotic arm refers to the specific parameters that drive the movement of the end effector of the first robotic arm (where the charging gun is held). In some examples, the offset of the first robotic arm may include position offset and attitude offset.

[0139] The charging robot can first calculate the total deviation between the charging gun and the charging base based on the first and second pose deviations, and then decompose the total deviation into executable motion commands. In some examples, the difference between the second and first pose deviations can be used as the total deviation between the charging gun and the charging base.

[0140] Step 2042: Adjust the pose of the first robotic arm based on the offset of the first robotic arm.

[0141] The charging robot can convert the offset of the first robotic arm into the end-effector target pose, and calculate the joint angle increment through inverse kinematics to control the motor to perform angle adjustment.

[0142] The automatic charging method provided in this application determines the offset of the robotic arm by measuring the positional deviation between the charging gun and the charging base and then adjusts its position. This method can accurately pinpoint the root cause of the deviation, achieve "deviation-adjustment" closed-loop control, and quickly reduce the positional deviation between the charging gun and the charging base.

[0143] In some embodiments of this application, after step 205 controls the first robotic arm to insert the charging gun into the charging socket, the automatic charging method further includes: controlling the first robotic arm to release the charging gun.

[0144] The automatic charging method provided in this application embodiment releases the charging gun after the first robotic arm inserts it into the charging socket. This not only prevents the charging socket from being damaged by continuous force, but also allows the first robotic arm to flexibly perform other tasks, so as to charge other vehicles.

[0145] In some embodiments of this application, the point cloud image of the charging gun and the charging dock is a single point cloud image that includes both the charging gun and the charging dock. For example... Figure 8B As shown above, in the above Figure 3 Based on the embodiment shown, step 2031 determines the point cloud map to be processed based on the point cloud map of the charging gun and the charging base, which may include the following steps 801 to 803.

[0146] Step 801: Analyze the integrity of the point cloud map of the charging gun and charging base to obtain the integrity analysis results.

[0147] The integrity analysis results include incomplete charging gun and / or incomplete charging base in the point cloud maps of the charging gun and charging base.

[0148] The charging robot can assess the coverage of key structures by the point clouds of the charging gun and charging base. If the coverage is greater than or equal to a preset value, the charging gun or charging base is considered complete in the point cloud image. If the coverage is less than the preset value, the charging gun or charging base is considered incomplete. The preset value can be greater than or equal to 90%. This embodiment does not limit the magnitude of the preset value; however, it uses a preset value of 98% as an example for illustrative purposes.

[0149] Step 802: If the charging gun or charging base is incomplete in the point cloud map of the charging gun and charging base, based on the integrity analysis results, control the second robotic arm to adjust its pose, and when the second robotic arm moves to the target shooting pose, control the 3D camera to recapture the point cloud map of the charging gun and charging base.

[0150] The target shooting pose can be the pose of the second robotic arm corresponding to the integrity analysis result, and can be changed according to the integrity analysis result. If the integrity analysis result indicates that the charging gun is incomplete, the target shooting pose can be the pose that can capture the entire charging gun; if the integrity analysis result indicates that the charging base is incomplete, the target shooting pose can be the pose that can capture the entire charging base.

[0151] Step 803: If the recaptured point cloud images of the charging gun and charging dock are complete, use the recaptured point cloud images of the charging gun and charging dock as the point cloud images to be processed.

[0152] The automatic charging method provided in this application first analyzes the integrity of the point cloud image, then controls the robotic arm to adjust the shooting pose and recapture for incomplete cases, until a complete point cloud image is obtained as the point cloud image to be processed. This process can actively identify and compensate for point cloud defects caused by occlusion, viewing angle deviation, etc., ensuring that the point cloud image to be processed contains complete three-dimensional features of the charging gun and charging base, providing a reliable data foundation for subsequent pose calculation, alignment judgment, and other steps. At the same time, this closed-loop mechanism of dynamically adjusting the shooting pose enhances the system's adaptability to complex environments, reduces subsequent processing errors caused by incomplete initial point clouds, and thus improves the stability and success rate of the entire charging alignment process.

[0153] Corresponding to the embodiments of the aforementioned automatic charging method, this application also provides an automatic charging device, such as... Figure 9As shown, the automatic charging device 90 may include a first control module 901, a second control module 902, a first determination module 903, an adjustment module 904, and a third control module 905; The first control module 901 is used to control the mobile chassis of the charging robot to move the charging robot to a preset working area between the vehicle to be charged and the charging pile when a charging request is received from the vehicle to be charged; wherein the vehicle to be charged is provided with a charging socket for coupling with the charging gun. The second control module 902 is used to control the first robotic arm of the charging robot to grab the charging gun from the charging pile when the charging robot is in the preset working area, and to control the first robotic arm of the charging robot to move towards the first preset pose with the charging gun; at the same time, it controls the second robotic arm of the charging robot to move towards the second preset pose with the 3D camera, and controls the 3D camera to capture the point cloud map of the charging gun and the charging base when the 3D camera moves to the second preset pose. The first determining module 903 is used to determine the alignment state of the charging gun and the charging base based on the point cloud map of the charging gun and the charging base. The adjustment module 904 is used to adjust the pose of the first robotic arm of the charging robot when it is determined that the charging gun and the charging base are not aligned. The third control module 905 is used to control the first robotic arm of the charging robot to insert the charging gun into the charging socket when the charging gun is aligned with the charging socket, so as to automatically charge the vehicle to be charged.

[0154] In some embodiments, the first determining module 903 is specifically used to determine a point cloud map to be processed based on the point cloud map of the charging gun and the charging base, process the point cloud map to be processed to obtain a first pose deviation of the charging gun and a second pose deviation of the charging base; if the first pose deviation and the second pose deviation are different, it is determined that the charging gun and the charging base are not aligned; if the first pose deviation and the second pose deviation are the same, it is determined that the charging gun and the charging base are aligned.

[0155] In some embodiments, the first determining module 903 is specifically used to perform point cloud registration processing on the point cloud image to be processed, to obtain a first transformation matrix corresponding to the charging gun and a second transformation matrix corresponding to the charging base; based on the first transformation matrix, to determine the first current pose of the charging gun in the robot coordinate system; based on the first current pose and the first reference pose corresponding to the charging gun, to determine the first pose deviation of the charging gun; based on the second transformation matrix, to determine the second current pose of the charging base in the robot coordinate system; and based on the second current pose and the second reference pose corresponding to the charging base, to determine the second pose deviation of the charging base.

[0156] In some embodiments, the first determining module 903 is specifically used to preprocess the point cloud image to be processed to obtain a first point cloud image corresponding to the charging gun; perform coarse matching between the first point cloud image and the first reference image corresponding to the charging gun to obtain a first initial transformation matrix; and perform fine matching between the first point cloud image and the first reference image based on the first initial transformation matrix to obtain a first transformation matrix.

[0157] In some embodiments, the first determining module 903 is specifically used to obtain the coordinates of the first center point cloud of the charging gun, the normal vector corresponding to the first center point cloud, and the coordinate transformation matrix between the camera coordinate system and the robot coordinate system; determine the pose of the charging gun in the camera coordinate system based on the coordinates of the first center point cloud of the charging gun, the normal vector corresponding to the first center point cloud, and the first transformation matrix; and determine the first current pose of the charging gun in the robot coordinate system based on the pose of the charging gun in the camera coordinate system and the coordinate transformation matrix.

[0158] In some embodiments, the adjustment module 904 is specifically used to determine the offset of the first robotic arm based on the first pose deviation and the second pose deviation when it is determined that the charging gun and the charging base are not aligned; and to adjust the pose of the first robotic arm based on the offset of the first robotic arm.

[0159] In some embodiments, the third control module 905 is also used to control the first robotic arm to release the charging gun.

[0160] In some embodiments, the point cloud image of the charging gun and the charging base is a single point cloud image that includes both the charging gun and the charging base. The first determining module 903 is specifically used to analyze the integrity of the point cloud image of the charging gun and the charging base to obtain an integrity analysis result. If the charging gun or the charging base is incomplete in the point cloud image of the charging gun and the charging base, based on the integrity analysis result, the second robotic arm is controlled to adjust its pose, and when the second robotic arm moves to the target shooting pose, the 3D camera is controlled to recapture the point cloud image of the charging gun and the charging base. If both the charging gun and the charging base are complete in the recaptured point cloud image of the charging gun and the charging base, the recaptured point cloud image of the charging gun and the charging base is used as the point cloud image to be processed.

[0161] The beneficial technical effects corresponding to the exemplary embodiments of the above-described automatic charging device can be found in the corresponding beneficial technical effects in the above-described method embodiments section, and will not be repeated here.

[0162] Based on the aforementioned embodiments of the automatic charging method, this application also provides an embodiment of a charging robot. For example... Figure 10 As shown, the charging robot 100 may include a mobile chassis 1001, a first robotic arm 1002, a second robotic arm 1003, a controller 1004, and a three-dimensional camera 1005 mounted on the second robotic arm 1003. The mobile chassis 1001 is used to move the charging robot 100 to a preset working area between the vehicle to be charged and the charging pile when a charging request is received from the vehicle to be charged. The first robotic arm 1002 is used to grab the charging gun from the charging pile, carry the charging gun to the first preset position, and insert the charging gun into the charging socket after aligning the charging gun with the charging socket of the vehicle to be charged to achieve automatic charging. The second robotic arm 1003 is used to carry the 3D camera 1005 to a second preset pose. The 3D camera 1005 is used to acquire point cloud images of the charging gun and charging base when the device is moved to a second preset pose. The controller 1004 is used to execute the automatic charging method provided in the above embodiments.

[0163] The beneficial technical effects corresponding to the above exemplary embodiments of the automated robot can be found in the corresponding beneficial technical effects in the above method embodiment section, and will not be repeated here.

[0164] Based on the above embodiments of the charging robot, this application also provides an embodiment of an automatic charging system. For example... Figure 11 As shown, the automatic charging system 110 includes a charging pile 1101 and a charging robot 1102. The charging pile 1101 is used to charge the vehicle to be charged when the charging gun is inserted into the charging socket of the vehicle to be charged.

[0165] Furthermore, embodiments of this application may also be computer-readable storage media storing computer program instructions thereon, which, when executed by a processor, cause the processor to perform the steps of the automatic charging methods of various embodiments of this application described in the above-described method embodiment section.

[0166] Computer-readable storage media may take the form of any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may include, but is not limited to, systems, apparatuses, or devices that are electrical, magnetic, optical, electromagnetic, infrared, or semiconductor, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0167] The basic principles of this application have been described above with reference to specific embodiments. However, the advantages, benefits, and effects mentioned in this application are merely examples and not limitations, and should not be considered as essential features of each embodiment of this application. Furthermore, the specific details of the above embodiments are for illustrative and facilitative purposes only, and are not limitations. These details do not restrict this application from being implemented using the aforementioned specific details.

[0168] Those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

[0169] Furthermore, the embodiments described above are merely specific embodiments of this application and are not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made based on the technical solution of this application should be included within the scope of protection of this application.

Claims

1. An automatic charging method applied to a charging robot, the charging robot comprising a mobile chassis, a first robotic arm, and a second robotic arm; the second robotic arm is equipped with a three-dimensional camera, characterized in that, The method includes: Upon receiving a charging request from a vehicle to be charged, the mobile chassis is controlled to move the charging robot to a preset working area between the vehicle to be charged and the charging pile; wherein, the vehicle to be charged is equipped with a charging socket for coupling with the charging gun. When the charging robot is in the preset working area, the first robotic arm is controlled to grab the charging gun from the charging pile and move with the charging gun to the first preset pose; at the same time, the second robotic arm is controlled to move with the 3D camera to the second preset pose, and when the 3D camera moves to the second preset pose, the 3D camera is controlled to capture the point cloud map of the charging gun and the charging base. Based on the point cloud map of the charging gun and the charging base, the alignment state of the charging gun and the charging base is determined. If it is determined that the charging gun and the charging base are not aligned, the position of the first robotic arm is adjusted. When the charging gun is aligned with the charging socket, the first robotic arm is controlled to insert the charging gun into the charging socket to automatically charge the vehicle to be charged.

2. The method according to claim 1, characterized in that, Determining the alignment state of the charging gun and the charging base based on the point cloud map of the charging gun and the charging base includes: Based on the point cloud maps of the charging gun and the charging dock, the point cloud map to be processed is determined. The point cloud image to be processed is processed to obtain the first pose deviation of the charging gun and the second pose deviation of the charging base; If the first posture deviation is different from the second posture deviation, it is determined that the charging gun and the charging base are not aligned. If the first posture deviation and the second posture deviation are the same, it is determined that the charging gun is aligned with the charging base.

3. The method according to claim 2, characterized in that, The process of processing the point cloud image to obtain the first pose deviation of the charging gun and the second pose deviation of the charging base includes: The point cloud image to be processed is subjected to point cloud registration processing to obtain a first transformation matrix corresponding to the charging gun and a second transformation matrix corresponding to the charging base. Based on the first transformation matrix, the first current pose of the charging gun in the robot coordinate system is determined; Based on the first current pose and the first reference pose corresponding to the charging gun, the first pose deviation of the charging gun is determined. Based on the second transformation matrix, the second current pose of the charging dock in the robot coordinate system is determined; Based on the second current pose and the second reference pose corresponding to the charging dock, the second pose deviation of the charging dock is determined.

4. The method according to claim 3, characterized in that, The point cloud image to be processed is subjected to point cloud registration processing to obtain a first transformation matrix corresponding to the charging gun, including: The point cloud image to be processed is preprocessed to obtain a first point cloud image corresponding to the charging gun. A coarse matching is performed between the first point cloud map and the first reference map corresponding to the charging gun to obtain the first initial transformation matrix; The first point cloud map and the first reference map are precisely matched based on the first initial transformation matrix to obtain the first transformation matrix.

5. The method according to claim 3, characterized in that, Determining the first current pose of the charging gun in the robot coordinate system based on the first transformation matrix includes: Obtain the coordinates of the first center point cloud of the charging gun, the normal vector corresponding to the first center point cloud, and the coordinate transformation matrix between the camera coordinate system and the robot coordinate system; Based on the coordinates of the first center point cloud of the charging gun, the normal vector corresponding to the first center point cloud, and the first transformation matrix, the pose of the charging gun in the camera coordinate system is determined. Based on the pose of the charging gun in the camera coordinate system and the coordinate transformation matrix, the first current pose of the charging gun in the robot coordinate system is determined.

6. The method according to claim 2, characterized in that, The step of adjusting the pose of the first robotic arm when it is determined that the charging gun and the charging base are not aligned includes: If it is determined that the charging gun and the charging base are not aligned, the offset of the first robotic arm is determined based on the first pose deviation and the second pose deviation. The pose of the first robotic arm is adjusted based on the offset of the first robotic arm.

7. The method according to any one of claims 1-6, characterized in that, After controlling the first robotic arm to insert the charging gun into the charging socket, the method further includes: Control the first robotic arm to release the charging gun.

8. The method according to any one of claims 1-6, characterized in that, The point cloud image of the charging gun and the charging dock is a single point cloud image that includes both the charging gun and the charging dock; the step of determining the point cloud image to be processed based on the point cloud image of the charging gun and the charging dock includes: The integrity of the point cloud maps of the charging gun and the charging dock is analyzed to obtain the integrity analysis results; If the charging gun or the charging base is incomplete in the point cloud map of the charging gun and the charging base, based on the integrity analysis result, the second robotic arm is controlled to adjust its pose, and when the second robotic arm moves to the target shooting pose, the three-dimensional camera is controlled to recapture the point cloud map of the charging gun and the charging base. If both the charging gun and the charging dock are intact in the recaptured point cloud image, the recaptured point cloud image of the charging gun and the charging dock is used as the point cloud image to be processed.

9. A charging robot, characterized in that, The charging robot includes a mobile chassis, a first robotic arm, a second robotic arm, a controller, and a three-dimensional camera mounted on the second robotic arm. The mobile chassis is used to move the charging robot to a preset working area between the vehicle to be charged and the charging pile when a charging request is received from the vehicle to be charged. The first robotic arm is used to grab the charging gun from the charging pile, carry the charging gun to a first preset position, and insert the charging gun into the charging socket after the charging gun is aligned with the charging socket of the vehicle to be charged to achieve automatic charging. The second robotic arm is used to carry the 3D camera to a second preset pose; The three-dimensional camera is used to capture point cloud images of the charging gun and the charging dock when the camera moves to the second preset pose. The controller is used to execute the automatic charging method according to any one of claims 1-8.

10. An automatic charging system, characterized in that, It includes a charging station and a charging robot, wherein the charging station is used to charge the vehicle when the charging gun is inserted into the charging socket on the vehicle to be charged.

Images