Charging system and information processing device

Abstract

A charging system and an information processing apparatus are provided, which reduce the situation that a connector for charging is not properly inserted into a power supply port of an electric vehicle. An information processing apparatus (20) determines the position of a power supply port (201) provided to an electric vehicle (2), and controls an arm portion (18) based on the determination result to move a connector (17) to a position facing the power supply port (201). In addition, the information processing apparatus (20) controls the arm portion (18) to insert the connector (17) into the power supply port (201). At this time, the information processing apparatus (20) controls the arm portion (18) based on the detection value of a force sensor (30) so that one or both of the force and the torque in a direction other than the insertion direction of the connector (17) are below a threshold value.

Description

Technical Field

[0001] This invention relates to a charging system and an information processing device. Background Technology

[0002] Technology for charging electric vehicles has been proposed. Patent Document 1 describes an automated electric vehicle workstation. In this workstation, an RFID system reads the charging conditions of a vehicle waiting to be charged, and these conditions are transmitted to a control board. The control board then controls an automatic charging mechanism based on these read charging conditions to charge the vehicle.

[0003] Existing technical documents

[0004] Patent documents

[0005] Patent Document 1: Japanese Utility Model Registration No. 3169439 Summary of the Invention

[0006] The problem the invention aims to solve

[0007] However, when automating the charging of electric vehicles, there are situations where the connector cannot be properly inserted into the power supply port. This is because the position and insertion direction of the power supply port of the electric vehicle are sometimes different each time due to factors such as the parking position and height of the vehicle. In the technology described in Patent Document 1, there are instances where the connector cannot be properly inserted into the power supply port.

[0008] One objective of this invention is to reduce situations where the charging connector cannot be properly inserted into the power supply port of an electric vehicle.

[0009] Solution for solving the problem

[0010] To address the aforementioned problems, one aspect of the present invention relates to a charging system comprising an arm, a force sensor, a charging connector, and one or more processors. The charging connector is fixed to the arm via the force sensor and can be inserted into the power supply port of an electric vehicle. The processor executes a position determination step, a movement control step, and an insertion control step. In the position determination step, the position of the connector located at the power supply port of the electric vehicle is determined. In the movement control step, the arm is controlled based on the determination result in the position determination step to move the connector to a position facing the power supply port. In the insertion control step, the arm is controlled to insert the connector into the power supply port. Furthermore, in the insertion control step, the processor controls the arm based on the detection value of the force sensor, such that the magnitude of one or both of the force and torque in directions other than the insertion direction of the connector is below a threshold value.

[0011] The effects of the invention

[0012] According to one aspect of the present invention, it is possible to reduce the number of situations where the connector cannot be properly inserted into the power supply port of the electric vehicle. Attached Figure Description

[0013] Figure 1 This is a schematic diagram showing the structure of the charging system according to Embodiment 1 of the present invention.

[0014] Figure 2 This is a block diagram that schematically illustrates the structure of the charging system according to Embodiment 1 of the present invention.

[0015] Figure 3 This is a diagram illustrating the appearance of the power supply port.

[0016] Figure 4 This is a diagram illustrating the appearance of the connector.

[0017] Figure 5 This is a block diagram showing the structure of an information processing device.

[0018] Figure 6 This is a flowchart illustrating an example of a charging operation performed by a charging system.

[0019] Figure 7 This is a schematic diagram illustrating the insertion action of inserting a connector into a power supply port.

[0020] Figure 8 This is a schematic diagram illustrating the insertion action of inserting a connector into a power supply port.

[0021] Figure 9 This is a flowchart illustrating an example of the actions involved in building the learning model and generating teacher data.

[0022] Explanation of reference numerals in the attached figures

[0023] 1: Charging system; 2: Electric vehicle; 10: Collaborative robot; 17: Connector; 18: Arm; 20: Information processing device; 21: Processor; 30: Force sensor; 40: First vision sensor; 50: Second vision sensor; 201: Power supply port. Detailed Implementation

[0024] [Implementation Method 1]

[0025] [System Overview]

[0026] The following describes one embodiment of the present invention. Figure 1This is a schematic diagram showing the structure of a charging system 1 according to one embodiment of the present invention. The charging system 1 is a system for charging an electric vehicle 2. The electric vehicle 2 is a car that uses electricity as its energy source and an electric motor as its power source for propulsion. The electric vehicle 2 is, for example, a battery-powered vehicle that charges a secondary battery (rechargeable battery) from an external power supply and supplies power from the secondary battery to the electric motor. Alternatively, the electric vehicle 2 may be a so-called hybrid vehicle that uses both an internal combustion engine and a secondary battery. The electric vehicle 2 has a power supply port 201 into which a charging connector 17 can be inserted. The location of the power supply port 201 varies depending on the vehicle model and type (hereinafter referred to as "model") of the electric vehicle 2.

[0027] The charging system 1 includes a collaborative robot 10, an information processing device 20, and a force sensor 30. The collaborative robot 10 is used to charge an electric vehicle 2 parked at the charging station S. The collaborative robot 10 has an arm 18, and performs the following actions: inserting a charging connector 17 into the power supply port 201 of the electric vehicle 2 via the arm 18, or removing the connector 17 from the power supply port 201. The information processing device 20 is a device that performs various calculations to control the collaborative robot 10, such as a personal computer.

[0028] The force sensor 30 is a sensor used to detect the direction and magnitude of force and torque, such as a 6-axis force sensor. Furthermore, the force sensor 30 is not limited to a 6-axis force sensor; it can also be a 4-axis force sensor or other force sensors.

[0029] exist Figure 1 In this embodiment, the charging system 1 includes one collaborative robot 10 and one information processing device 20, but this is not a limitation of this embodiment. The charging system 1 may also have a structure with multiple collaborative robots 10 and multiple information processing devices 20.

[0030] Figure 2 This is a block diagram schematically illustrating the structure of a charging system 1 according to one embodiment of the present invention. The collaborative robot 10 includes a connector 17 and an arm 18. The connector 17 is a charging connector that can be inserted into the power supply port 201 of an electric vehicle. The arm 18 is used to move the connector 17, for example, it is the robotic arm of the collaborative robot 10. The arm 18 includes one or more joints, which are driven to perform actions. The connector 17 is fixed to the arm 18 via a force sensor 30.

[0031] The information processing device 20 includes a processor 21. The processor 21 executes a charging control method M1, which includes a position determination step M11, a movement control step M12, and an insertion control step M13. The position determination step M11 is the step of determining the position of the power supply port 201 provided on the electric vehicle 2. For example, the processor 21 determines the model of the electric vehicle 2 based on an image captured by a camera device used to photograph the electric vehicle 2, and determines the position of the power supply port 201 based on the determined model.

[0032] The movement control step M12 is a step of controlling the arm 18 to move the connector 17 to a position facing the power supply port 201 based on the determination result in the position determination step M11. The position facing the power supply port 201 is the position where the power supply port 201 and the connector 17 are opposite each other at the beginning of the insertion action of the connector 17.

[0033] Insertion control step M13 is the step of controlling arm 18 to insert connector 17 into power supply port 201. In insertion control step M13, processor 21 controls arm 18 based on the detection value of force sensor 30, such that the magnitude of force and / or torque in directions other than the insertion direction of connector 17 is below a threshold.

[0034] The arm 18 drives the joint according to the control information supplied by the processor 21, thereby changing the posture and position of the connector 17. The arm 18, according to the control information supplied by the processor 21, performs the action of inserting the connector 17 into the power supply port 201 while changing the posture and position of the connector 17. During the insertion action, the arm 18 changes the posture and position of the connector 17, therefore the insertion direction of the connector 17 into the power supply port 201 changes during the insertion action.

[0035] According to the above structure, the charging system 1 controls the arm 18 to move the connector 17 to the power supply port 201, and inserts the connector 17 into the power supply port 201 while controlling the force and / or torque in directions other than the insertion direction of the connector 17 to be below a threshold. During the insertion operation, the processor 21 controls the force and / or torque in directions other than the insertion direction of the connector 17 to be below a threshold, thus changing the insertion direction of the connector 17. In this way, the position and insertion direction of the power supply port 201 are corrected during the insertion operation, so that proper insertion can be performed.

[0036] [System Structure]

[0037] Next, refer to Figure 1 To illustrate the structure of charging system 1. For example... Figure 1As shown, in addition to the collaborative robot 10, the information processing device 20 and the force sensor 30, the charging system 1 also has a first vision sensor 40, a second vision sensor 50 and an unmanned transport vehicle 60.

[0038] The first vision sensor 40 is used to photograph the electric vehicle 2 parked at the charging station S. The first vision sensor 40 is positioned and positioned at the charging station S to photograph the electric vehicle 2 parked at the charging station S.

[0039] The first vision sensor 40 outputs data representing the captured image to the information processing device 20. The first vision sensor 40 can also perform the determination using other types of sensors, such as photoelectric sensors.

[0040] The second vision sensor 50 is used to capture images of the power supply port 201 of the electric vehicle 2. The second vision sensor 50 is mounted on the arm 18 of the collaborative robot 10. Furthermore, the second vision sensor 50 is positioned so that its imaging direction aligns with the direction in which the power supply port 201, into which the connector 17 can be inserted, is located. The second vision sensor 50 outputs data representing the captured image to the information processing device 20. The second vision sensor 50 can also be used to make judgments using other types of sensors, such as photoelectric sensors.

[0041] The unmanned transport vehicle 60 is an example of a transport mechanism for transporting the collaborative robot 10, including its arm 18. The unmanned transport vehicle 60 has a movement mechanism such as wheels or tracks. The unmanned transport vehicle 60 is parked at a predetermined position (hereinafter referred to as the "initial position") at the charging station S. When it is necessary to charge the electric vehicle 2, the unmanned transport vehicle 60 moves the collaborative robot 10 from the initial position to the vicinity of the power supply port 201 based on control information supplied from the information processing device 20. Furthermore, the unmanned transport vehicle 60 can be independent of the collaborative robot 10, or it can be integrated with the collaborative robot 10.

[0042] [Structure of Collaborative Robots]

[0043] As described above, the collaborative robot 10 includes a connector 17 and an arm 18. The connector 17 is, for example, located at the end of the arm 18. A charging cable is connected to the connector 17, and by inserting the connector 17 into the power supply port 201, the electric vehicle 2 is charged using power from the charging cable. Additionally, the collaborative robot 10 may also include a notification unit (not shown) that notifies the driver of the electric vehicle 2. This notification unit may be, for example, a speaker that outputs warning sounds or audio messages, or a display that shows images such as messages.

[0044] Figure 3This is a diagram illustrating the appearance of the power supply port 201 into which connector 17 is inserted. Figure 4 This is a diagram illustrating the appearance of connector 17. Figure 3 In the power supply port 201, there are a cover 201a and a recess 201b. The cover 201a is provided in a closable manner, and in the closed state, the cover 201a is fixed to cover the recess 201b. When the cover 201a is opened, the recess 201b is exposed to the outside. The driver of the electric vehicle 2 opens and closes the cover 201a by operating a designated control device (operation button, etc.) of the electric vehicle 2.

[0045] Connector 17 has a convex shape that is embedded in recess 201b. Connector 17 is inserted into power port 201 to charge electric vehicle 2. Figure 4 In this example, connector 17 is fixed to the end of the arm 18 of the collaborative robot 10 via force sensor 30. The joints of the arm 18 are driven, thereby changing the position and orientation of connector 17 with the movement of the arm 18. Furthermore, the shapes of the power supply port 201 and connector 17 are not limited to... Figure 3 and Figure 4 The illustrated shape allows for the application of various shapes of power supply ports 201 and connectors 17.

[0046] In this embodiment, the force sensor 30 is a 6-axis force sensor. Figure 4 In the example, force sensor 30 detects force components Fx, Fy, Fz in the x-axis, y-axis, and z-axis directions, and torque components Mx, My, Mz in the x-axis, y-axis, and z-axis directions, respectively, in a 3D space defined by the x-axis, y-axis, and z-axis. Figure 4 In this example, the force sensor 30 is fixed in an orientation such that the positive direction of the z-axis is the insertion direction of the connector 17. That is, when the orientation or position of the connector 17 changes due to the movement of the arm 18, the direction of each component detected by the force sensor 30 also changes with the orientation or position of the connector 17. The information processing device 20 acquires each component detected by the force sensor 30.

[0047] Furthermore, in this embodiment, the structure in which the force sensor 30 is independent of the collaborative robot 10 has been described, but it is also possible for the force sensor 30 to be built into the collaborative robot 10. In this case, the collaborative robot 10 outputs the components detected by the force sensor 30 to the information processing device 20.

[0048] [Structure of Information Processing Device 20]

[0049] Figure 5This is a block diagram showing the structure of the information processing device 20. The information processing device 20 is implemented using a general-purpose computer and includes a processor 21, a primary memory 22, a secondary memory 23, an input / output IF 24, a communication IF 25, and a bus 26. The processor 21, primary memory 22, secondary memory 23, input / output IF 24, and communication IF 25 are interconnected via the bus 26.

[0050] The secondary memory 23 stores the charging control program P1 and the learning completion model LM1. The processor 21 expands the charging control program P1 and the learning completion model LM1 stored in the secondary memory 23 onto the primary memory 22, and executes the steps of the charging control method M1 according to the commands contained in the charging control program P1 expanded on the primary memory 22. When the processor 21 executes the location determination step M11 of the charging control method M1, it utilizes the learning completion model LM1 expanded on the primary memory 22. Furthermore, storing the charging control program P1 in the secondary memory 23 means that the secondary memory 23 stores source code or a runtime file obtained by compiling source code. Additionally, storing the learning completion model LM1 in the secondary memory 23 means that the secondary memory 23 stores parameters used to define the learning completion model LM1.

[0051] As a device that can be used as processor 21, a CPU (Central Processing Unit) can be listed as an example. Additionally, as a device that can be used as primary memory 22, a semiconductor RAM (Random Access Memory) can be listed as an example. Furthermore, as a device that can be used as secondary memory 23, flash memory can be listed as an example.

[0052] Input and / or output devices are connected to input / output IF 24. Examples of input / output IF 24 include USB (Universal Serial Bus). In charging control method M1, data acquired from the collaborative robot 10, force sensor 30, first vision sensor 40, and second vision sensor 50 are input to information processing device 20 via input / output IF 24. Additionally, information provided to the user in charging control method M1 is output from information processing device 20 via input / output IF 24.

[0053] Communication IF 25 is an interface for communicating with other computers. Communication IF 25 can include an interface for communicating with other computers without a network or other networks, such as a Bluetooth (trademarked) interface. Alternatively, Communication IF 25 can include an interface for communicating with other computers via a LAN (Local Area Network), such as a Wi-Fi (trademarked) interface. The collaborative robot 10 and the information processing device 20 can be connected either via Input / Output IF 24 or via Communication IF 25.

[0054] Furthermore, in this embodiment, a structure is adopted in which a single processor (processor 21) is used to execute the charging control method M1, but the present invention is not limited to this. That is, a structure in which multiple processors are used to execute the charging control method M1 may also be adopted. In this case, the multiple processors that cooperate in executing the charging control method M1 may be configured to be located in a single computer and able to communicate with each other via a bus, or they may be configured to be distributed among multiple computers and able to communicate with each other via a network. As an example, consider the following: a processor built into a computer constituting a cloud server and a processor built into a computer owned by a user of the cloud server cooperate in executing the charging control method M1.

[0055] Furthermore, in this embodiment, a structure is adopted in which the learned completion model LM1 is stored in the memory (secondary memory 23) of the same computer as the processor (processor 21) that executes the charging control method M1, but the present invention is not limited to this. That is, a structure can also be adopted in which the learned completion model LM1 is stored in the memory of a different computer than the processor that executes the charging control method M1. In this case, the computer with the memory storing the learned completion model LM1 is configured to communicate with the computer with the processor that executes the charging control method M1 via a network. As an example, consider the following: the learned completion model LM1 is stored in the memory of the computer that constitutes the cloud server, and the charging control method M1 is executed by the processor of the computer owned by the user of the cloud server.

[0056] Furthermore, in this embodiment, a structure is adopted in which the learned model LM1 is stored in a single memory (secondary memory 23), but the present invention is not limited to this. That is, a structure in which the learned model LM1 is stored in multiple memories can also be adopted. In this case, the multiple memories storing the learned model LM1 can be located in a single computer (which may or may not be the computer with a built-in processor that executes the charging control method M1), or they can be distributed among multiple computers (which may or may not include the computer with a built-in processor that executes the charging control method M1). As an example, the following structure is considered: the learned model LM1 is stored in multiple memories in the memories of each of the multiple computers constituting the cloud server.

[0057] The learned model LM1 is a learned model used to determine the type of electric vehicle 2. The learned model LM1 is a model obtained by learning from images captured by the first vision sensor 40 as input and outputting the type of electric vehicle 2. The learned model LM1 can be derived using algorithms such as neural network models (e.g., convolutional neural networks, recurrent neural networks), regression models (e.g., linear regression), or tree models (e.g., regression trees).

[0058] [Charging action for electric vehicle 2]

[0059] Figure 6 This is a flowchart illustrating the charging operation performed by the charging system 1. When the electric vehicle 2 stops at the charging station S, in step S11, the first vision sensor 40 takes a picture of the electric vehicle 2 stopped at the charging station S and outputs data representing the captured image to the information processing device 20.

[0060] In step S12, the processor 21 determines the vehicle type of the electric vehicle 2 based on the image captured by the first vision sensor 40. The processor 21 may use a learned model LM1, for example, to determine the vehicle type.

[0061] The processor 21 can use either the image captured by the first vision sensor 40 or a portion of the image captured by the first vision sensor 40 as input. The processor 21 determines the model of the electric vehicle 2 based on the output value obtained by inputting the input image into the learning completion model LM1 constructed through machine learning.

[0062] Furthermore, the processor 21 may also determine the model of the electric vehicle 2 by using, for example, a rule base, instead of the learned model LM1. In this case, for example, image data representing images of the electric vehicle 2 captured by the processor 20 are pre-stored in the secondary memory 23 of the information processing device 20 for each model. The processor 21 analyzes the images captured by the first vision sensor 40 and uses the pre-stored image data to determine the model of the electric vehicle 2 by methods such as pattern matching.

[0063] In step S13, the processor 21 determines the stopping position of the electric vehicle 2 and the position of the power supply port 201 based on the vehicle model identified in step S12. In this embodiment, the processor 21 analyzes the image captured by the first vision sensor 40 to determine the approximate stopping position of the electric vehicle 2 and the position of the power supply port 201. For example, for each vehicle model, image data representing images captured of the electric vehicle 2 are pre-stored in the secondary memory 23 of the information processing device 20, and the processor 21 uses this image data for determination processing. In this case, for example, the processor 21 may also use the image captured by the first vision sensor 40 and the pre-stored image data to determine the parking position of the electric vehicle 2 and the orientation of the vehicle body using methods such as pattern matching. Alternatively, for example, the processor 21 refers to a table that stores the vehicle model of the electric vehicle 2 in correspondence with the position of the power supply port 201 to determine the position of the power supply port 201 corresponding to the vehicle model identified in step S12.

[0064] When the processor 21 determines the location of the power supply port 201, it outputs control information to the unmanned transport vehicle 60 to move the unmanned transport vehicle 60 to the vicinity of the power supply port 201. That is, the processor 21 uses the unmanned transport vehicle 60 to move the arm 18 to the vicinity of the power supply port 201, and uses the arm 18 to move the connector 17 to a position facing the power supply port 201. When the unmanned transport vehicle 60 obtains the control information from the information processing device 20, in step S14, it moves the collaborative robot 10 to the position indicated by the information processing device 20 based on the obtained control information.

[0065] Furthermore, when the unmanned transport vehicle 60 completes the transport, the collaborative robot 10 moves the connector 17 to a position facing the power supply port 201 and changes the posture of the connector 17 to the posture for initiating the insertion action. For example, the posture of the connector 17 when the arm 18 begins the insertion action is preset for each model of the electric vehicle 2. In this case, for example, the model of the electric vehicle 2 and the posture of the connector 17 are pre-stored in the secondary memory 23 in correspondence, and the processor 21 instructs the arm 18 to the posture corresponding to the model of the electric vehicle 2 to be charged. Through this movement, the positional relationship between the second vision sensor 50 and the power supply port 201 is changed to the positional relationship where the power supply port 201 is located in the camera direction of the second vision sensor 50.

[0066] At any time after parking the electric vehicle 2 at the charging station S, the driver of the electric vehicle 2 uses a control device (operation button, etc.) provided on the electric vehicle 2 to operate the cover 201a of the power supply port 201. The electric vehicle 2 opens the cover 201a based on the operation information output from the control device. The cover 201a is opened, thereby exposing the recess 201b to the outside.

[0067] In step S15, the second vision sensor 50 captures an image of the power supply port 201 and outputs image data representing the captured image. In step S16, the processor 21 determines the position of the power supply port 201 based on the image captured by the second vision sensor 50. Although the position of the power supply port 201 was determined in step S13, errors may occur due to factors such as the air pressure of the electric vehicle 2's tires and the tilt of the ground. Therefore, in step S16, the processor 21 analyzes the image captured by the second vision sensor 50 to determine the position of the power supply port 201 with higher accuracy.

[0068] Specifically, for example, image data representing images captured of the power supply port 201 of the electric vehicle 2 is pre-stored in the secondary memory 23 of the information processing device 20 for each vehicle model. The processor 21 uses the captured image from the second vision sensor 50 and the pre-stored image data to determine the position of the power supply port 201 on the electric vehicle 2 using methods such as pattern matching. In addition, the processor 21 stores information indicating the start position of the insertion action of the connector 17 (hereinafter referred to as "start position information") in the secondary memory 23. This information is referenced when the connector 17 is removed.

[0069] In step S16, when the cover 201a of the power supply port 201 is closed, the processor 21 may sometimes be unable to properly determine the position of the power supply port 201. If the position cannot be determined, the processor 21 may also notify the driver by controlling the collaborative robot 10 to output a warning sound or an audio message such as "Please open the cover of the power supply port".

[0070] In step S17, processor 21 controls arm 18 to insert connector 17 into power supply port 201. Arm 18 performs the action of inserting connector 17 into power supply port 201, which was positioned in step S16, based on control information supplied by processor 21. At this time, processor 21 controls arm 18 based on the detection value of force sensor 30, such that the magnitude of force and / or torque in directions other than the insertion direction of connector 17 is below a threshold.

[0071] Specifically, in Figure 4 In this example, processor 21 controls arm 18 to move connector 17 in the insertion direction (i.e., the positive direction of the z-axis of force sensor 30), thereby performing an insertion action. Furthermore, during this insertion action, processor 21 controls arm 18 such that the absolute values ​​of the force sensor 30's detected values, excluding the force component Fz (i.e., force components Fx, Fy and torque components Mx, My, Mz), are below a predetermined threshold. This predetermined threshold can be, for example, zero or a value greater than zero. As the threshold, different values ​​can be set for each component, or a threshold common to multiple components can be used.

[0072] Figure 7 and Figure 8 This is a schematic diagram illustrating an example of the insertion operation of connector 17 into power supply port 201. Figure 7 In the process, the arm 18, based on control information supplied by the processor 21, inserts the connector 17 into the positive direction of the z-axis of the force sensor 30, i.e., the insertion direction z1. During the insertion operation, if a deviation occurs between the desired insertion direction a1 and the insertion direction z1 of the connector 17, such as... Figure 8 As shown, a portion of connector 17 will collide with the inner wall of the recess 201b of the power supply port 201, thereby pressing connector 17 against the inner wall of the recess 201b. As a result, an external force Fa will be applied from the recess 201b to connector 17 and force sensor 30.

[0073] Force sensor 30 detects the force and torque components of the external force Fa. Processor 21 supplies control information to arm 18 for correcting the position and orientation of connector 17, ensuring that the detected values ​​of components other than the force component Fz from force sensor 30 are below a threshold. Figure 8In the example, the processor 21 provides control information to the arm 18 for correcting the position and orientation of the connector 17 based on the detection value of the force sensor 30, such that the difference between the positive direction of the z-axis of the force sensor 30, i.e. the insertion direction z1, and the desired insertion direction a1 becomes smaller.

[0074] Thus, based on the control information supplied by the processor 21, the arm 18 performs the action of inserting the connector 17 into the power supply port 201 while correcting the position and orientation of the connector 17. The processor 21 repeats the above-described correction process at predetermined unit time intervals. By repeatedly correcting the position and orientation of the connector 17 while performing the insertion action, even if there is a deviation between the insertion direction of the connector 17 (the z-direction of the force sensor 30) and the desired insertion direction a1, the insertion direction can be corrected while performing the insertion. This reduces the likelihood of the connector 17 not being properly inserted into the power supply port 201 of the electric vehicle 2.

[0075] exist Figure 4 In the example described above, the processor 21 controls the arm 18 such that the detected values ​​of the force sensor 30, excluding the force component Fz (i.e., force components Fx, Fy and torque components Mx, My, Mz), are below a predetermined threshold. The control method of the processor 21 is not limited to the method described above; the processor 21 can also control the arm 18 such that the detected values ​​of the force components (Fx, Fy) or torque components (Mx, My, Mz) are below the threshold.

[0076] return Figure 6 The following is an explanation. In step S18, the processor 21 determines whether the insertion of the connector 17 into the power supply port 201 has been completed. When the insertion of the connector 17 into the power supply port 201 is completed, an external force in the z-axis direction is applied from the recess 201b to the connector 17 and the force sensor 30. The processor 21 determines whether the insertion of the connector 17 has been completed based on whether the force sensor 30 detects an external force above a threshold in the insertion direction (z-axis direction). If the magnitude of the force in the z-axis direction detected by the force sensor 30 is above a predetermined threshold, the processor 21 determines that the insertion has been completed. On the other hand, if the magnitude of the force in the z-axis direction detected by the force sensor 30 is less than the threshold, the processor 21 determines that the insertion has not been completed. If the insertion has been completed ("Yes" in step S18), the processor 21 proceeds to the processing in step S19. On the other hand, if the insertion has not been completed ("No" in step S18), the processor 21 returns to the processing in step S17 and continues the insertion operation.

[0077] Connector 17 is inserted into power supply port 201 to supply power to electric vehicle 2. In step S19, processor 21 determines whether charging of electric vehicle 2 has been completed. For example, if the current supplied to electric vehicle 2 through connector 17 decreases to a predetermined value, processor 21 determines that charging has been completed.

[0078] If charging is complete (Yes in step S19), processor 21 proceeds to step S20. On the other hand, if charging is incomplete (No in step S19), processor 21 returns to step S19 and enters standby mode until charging is complete.

[0079] In step S20, the arm 18 performs the action of pulling the connector 17 out of the power supply port 201 based on the control information supplied by the processor 21. This pulling action is the opposite of the insertion action in step S17. That is, the processor 21 supplies the arm 18 with control information to move the connector 17 in the direction opposite to the insertion direction (i.e., the negative direction of the z-axis of the force sensor 30).

[0080] During the pull-out action, the processor 21 also performs the following actions: while controlling the arm 18 based on the detection value of the force sensor 30, so that the magnitude of the force and torque in directions other than the insertion direction of the connector 17 is below a threshold, the connector 17 is pulled out from the power supply port 201.

[0081] Specifically, the processor 21 controls the arm 18, for example, to move the connector 17 in the pull-out direction (i.e., the negative direction of the z-axis of the force sensor 30), thereby performing a pull-out action. Furthermore, during this pull-out action, the processor 21 controls the arm 18 such that the detected values ​​of the force sensor 30, excluding the force component Fz (i.e., force components Fx, Fy and torque components Mx, My, Mz), are below a predetermined threshold.

[0082] In step S21, the processor 21 determines whether the removal of the connector 17 has been completed. The processor 21, for example, refers to the start position information stored in the secondary memory 23 and determines whether the removal has been completed based on whether the connector 17 has moved to the start position of the insertion operation through the removal action. In this case, the processor 21 determines that the removal is complete if the connector 17 has moved to the start position, and determines that the removal is incomplete if the connector 17 has not moved to the start position.

[0083] If the unplugging action has been completed ("Yes" in step S21), the processor 21 proceeds to step S22. On the other hand, if the unplugging action has not been completed ("No" in step S21), the processor 21 returns to step S20 and continues the unplugging action.

[0084] The processor 21 outputs control information to the unmanned transport vehicle 60 to move the unmanned transport vehicle 60 to its initial position. When the unmanned transport vehicle 60 obtains the control information from the information processing device 20, in step S22, it moves the collaborative robot 10 to its initial position based on the obtained control information.

[0085] When the charging action performed by the collaborative robot 10 is completed, the driver of the electric vehicle 2 closes the cover 201a of the power supply port 201. The driver can close the cover 201a manually or by using a designated actuator to close the cover 201a.

[0086] [Generation of teacher data and construction of learning completion models]

[0087] Next, please refer to the attached diagram for explanation. Figure 6 The steps S12 include the construction of the learning completion model LM1 and the generation of teacher data used in the construction process. In this embodiment, the information processing device 20 performs the construction process of the learning completion model LM1 and the generation process of teacher data. However, the construction process of the learning completion model LM1 and the generation process of teacher data can also be performed by other devices besides the information processing device 20.

[0088] The teacher data used in the construction of the learning-complete model LM1 includes video images of electric vehicle 2 and label data representing the model of electric vehicle 2.

[0089] Figure 9 This is a flowchart illustrating an example of the construction actions of learning and completing the model LM1 and the generation actions of teacher data. In S31, the first visual sensor 40 captures an image of the electric vehicle 2 to generate a video image. The information processing device 20 can use the video image generated by the first visual sensor 40 directly as a learning image, or it can use an image obtained by extracting a portion of the video image as a learning image.

[0090] In S32, the information processing device 20 generates teacher data by associating image-based label data with learning data. The label data represents vehicle type data. The label data is input to the information processing device 20, for example, via input / output IF 24.

[0091] In S33, the information processing device 20 constructs a learning completion model LM1 using teacher-learned data. The learning completion model LM1 can be implemented using algorithms such as neural network models (e.g., convolutional neural networks, recurrent neural networks), regression models (e.g., linear regression), or tree models (e.g., regression trees).

[0092] [Effects of this implementation method]

[0093] Furthermore, when charging the electric vehicle 2, there are instances where the power supply port 201 cannot be properly aligned with the connector 17, or the connector 17 cannot be properly inserted into the power supply port 201. This is because the position of the power supply port 201 varies depending on the vehicle model, or sometimes the insertion direction may deviate slightly due to changes in the air pressure of the electric vehicle 2's tires.

[0094] In contrast, according to this embodiment, the processor 21 controls the insertion of the connector 17 during the insertion operation, ensuring that the magnitude applied in directions other than the insertion direction of the connector 17 is below a threshold, thereby changing the insertion direction of the connector 17. Therefore, even if there is a slight deviation between the position of the power supply port 201 and the insertion direction, the insertion direction can be corrected to ensure proper insertion. Thus, situations where the connector 17 cannot be properly inserted into the power supply port 201 of the electric vehicle 2 can be reduced.

[0095] Furthermore, according to this embodiment, the charging system 1 controls the insertion direction of the connector 17 based on one or both of the force and torque detected by the force sensor 30, which is a 6-axis force sensor. This reduces the likelihood of the connector 17 being improperly inserted into the power supply port 201 of the electric vehicle 2.

[0096] Furthermore, according to this embodiment, the processor 21 determines the model of the electric vehicle 2 based on images captured by a sensor (such as a first vision sensor) used to photograph the electric vehicle 2, and determines the position of the power supply port 201 based on the determined model. Therefore, it is possible to insert the connector 17 into the power supply port 201 for electric vehicles 2 of multiple models.

[0097] Furthermore, according to this embodiment, the processor 21 determines the position of the power supply port 201 based on the image captured by the second vision sensor 50. Therefore, even if the position of the power supply port 201 deviates due to factors such as the inclination of the ground where the electric vehicle 2 is parked or the height of the vehicle, the accuracy of determining the position of the power supply port 201 of the electric vehicle 2 can be improved.

[0098] Furthermore, according to this embodiment, the processor 21 uses the unmanned transport vehicle 60, which serves as a transport mechanism, to move the arm 18 to the vicinity of the power supply port 201, and uses the arm 18 to move the connector 17 to a position facing the power supply port 201. Therefore, even if the parking position of the electric vehicle 2 is different each time, the connector 17 can be inserted into the power supply port 201 of the electric vehicle 2 by moving the connector 17 to the vicinity of the power supply port 201.

[0099] Furthermore, by changing the insertion direction of the connector 17 based on the detection results of the force sensor 30 while performing the insertion action, it is possible to prevent the connector 17 from being inserted into the power supply port 201 in an incorrect insertion direction. The connector 17 will not be forced into the power supply port 201, thus preventing damage to the connector 17 and the power supply port 201.

[0100] Furthermore, in this embodiment, errors in the positional relationship between the connector 17 and the power supply port 201, as well as some degree of error in the insertion direction of the connector 17, are corrected during the insertion of the connector 17. Therefore, it is not necessary to use a high-precision sensor as the first visual sensor 40 for capturing images of the power supply port 201, nor is it necessary to use a high-precision collaborative robot 10.

[0101] Furthermore, in this embodiment, if the force sensor 30 detects an external force exceeding a threshold in the z-axis direction (insertion direction), the processor 21 determines that the insertion action has been completed. Thus, the processor 21 can detect whether the pushing action of the connector 17 into the power supply port 201 has been completed.

[0102] Furthermore, in this embodiment, the information processing device 20 determines the location of the power supply port 201, and the collaborative robot 10 moves to the determined location. Therefore, the collaborative robot 10 can charge the electric vehicle 2 parked at any location in the charging station S. That is, it is not necessary to pre-determine the parking location of the electric vehicle 2, and power can be supplied regardless of the parking location of the electric vehicle 2, thereby reducing the burden on the driver. In addition, the driver of the electric vehicle 2 does not need to perform power supply operations, so the driver will not accidentally damage the connector 17, and the driver does not need to perform complex operations such as handling the charging cable connected to the connector 17.

[0103] Furthermore, in this embodiment, the unmanned transport vehicle 60 moves to the power supply port 201 for charging, so the driver of, for example, the electric vehicle 2, does not need to get out of the vehicle to obtain power. Additionally, since no power supply operation is required, there is no risk of accidental electric shock to the driver.

[0104] Furthermore, in this embodiment, the processor 21 controls the force components (Fx, Fy) to be below a threshold value, thereby preventing excessive normal force from acting on both sides when the connector 17 contacts the sidewall of the power supply port 201. Additionally, the processor 21 controls the torque components (Mx, My, Mz) to be below a threshold value, thereby preventing excessive frictional force from acting on both sides when the connector 17 contacts the sidewall of the power supply port 201.

[0105] In addition, the processor 21 controls the absolute values ​​of the detected components of force (Fx, Fy) and torque (Mx, My, Mz) to be below the threshold, thereby reducing the possibility of damage to the connector or power supply port due to normal force or friction.

[0106] [Variation Example]

[0107] In the above embodiment, the collaborative robot 10 is shown to have a structure including the connector 17, but the connector 17 may also be constructed independently of the collaborative robot 10. In this case, the connector 17 may be held, for example, by a holding part that serves as the end effector of the arm 18 of the collaborative robot 10, for insertion into the power supply port 201.

[0108] Additionally, in the above-described embodiment, the processor 21 can also detect malfunctions or foreign object intrusion in the connector 17 based on the detection values ​​of the force sensor 30. In this case, for example, information representing the movement trajectory of the connector 17 when it is properly inserted into the power supply port 201 (hereinafter referred to as "trajectory information") is pre-stored in the secondary memory 23 for each vehicle model. The processor 21 can also compare the movement trajectory represented by the trajectory information stored in the secondary memory 23 with the actual movement trajectory of the connector 17, and determine that an abnormality has occurred if the difference between the two does not meet a predetermined condition. The predetermined condition is, for example, the following: the force sensor 30 detects a force of more than a threshold in the z-axis direction, and the difference between the movement distance of the connector 17 and the movement distance corresponding to the trajectory information stored in the secondary memory 23 is more than a predetermined threshold.

[0109] In the above embodiments, the collaborative robot 10 and the information processing device 20 are described as being configured as independent devices. The structure of the charging system 1 is not limited to the structure shown in the above embodiments. For example, the collaborative robot 10 and the information processing device 20 may also be configured as an integrated device.

[0110] Furthermore, in the above-described embodiment, the processor 21 executes the position determination step M11, the movement control step M12, and the insertion control step M13. However, these steps can also be performed by the information processing device 20 and one or more other devices. For example, the position determination step M11 can be executed by the processor 21, while the movement control step M12 and the insertion control step M13 can be executed by the processor disposed on the collaborative robot 10.

[0111] In the above embodiment, the driver operation of the electric vehicle 2 is provided on the operator's switch of the electric vehicle 2, thereby opening and closing the cover 201a. The triggering of opening and closing the cover 201a is not limited to the driver's prescribed operation, but can also be other. For example, the electric vehicle 2 may open the cover 201a when the connector 17 approaches the power supply port 201 and the distance between the connector 17 and the power supply port 201 becomes below a predetermined threshold. Alternatively, the electric vehicle may close the cover 201a when the distance between the connector 17 and the power supply port 201 becomes above the threshold. In this case, for example, the power supply port 201 may be equipped with a distance sensor structure, which measures the distance between the power supply port 201 and the connector 17 that has approached the power supply port 201. As the distance sensor, known distance sensors such as ToF (Time of Flight) sensors, Doppler sensors, and cameras can be used.

[0112] 〔Summarize〕

[0113] The charging system of Method 1 includes: an arm; a force sensor; a charging connector fixed to the arm via the force sensor and capable of being inserted into a power supply port of an electric vehicle; and one or more processors, wherein the processors perform the following steps: a position determination step, in which a position is determined at the power supply port of the electric vehicle; a movement control step, in which the arm is controlled based on the determination result in the position determination step to move the connector to a position facing the power supply port; and an insertion control step, in which the arm is controlled to insert the connector into the power supply port, wherein, in the insertion control step, the processor controls the arm based on the detection value of the force sensor such that the magnitude of one or both of the force and torque in directions other than the insertion direction of the connector is below a threshold value.

[0114] According to the above structure, during the connector insertion operation, the processor controls the process to ensure that the magnitude of one or both of the force and torque in directions other than the connector insertion direction is below a threshold. By controlling the force in directions other than the insertion direction to be below the threshold, excessive normal force can be prevented from acting on either side when the connector contacts the sidewall of the power supply port. Furthermore, by controlling the torque to be below the threshold, excessive frictional force can be prevented from acting on either side when the connector contacts the sidewall of the power supply port. By controlling the magnitude of both the force and torque in directions other than the insertion direction to be below the threshold, the possibility of damage to the connector or power supply port due to normal force or friction can be reduced.

[0115] The information processing device involved in Method 2 includes one or more processors, which perform the following steps: a position determination step, in which the position determination step is determined to be located at a power supply port of an electric vehicle; a movement control step, in which the arm of a charging connector that can be inserted into the power supply port is fixed by a force sensor based on the determination result in the position determination step, so that the connector moves to a position facing the power supply port; and an insertion control step, in which the arm is controlled to insert the connector into the power supply port, wherein, in the insertion control step, the processor controls the arm based on the detection value of the force sensor, such that the magnitude of one or both of the force and torque in a direction other than the insertion direction of the connector is below a threshold.

[0116] According to the above structure, the processor controls the process during the connector insertion operation to ensure that the magnitude of one or both of the force and torque in directions other than the connector insertion direction is below a threshold. Therefore, the insertion operation is performed while changing the connector insertion direction. This reduces the likelihood of the connector not being properly inserted into the electric vehicle's power supply port.

[0117] [Additional Notes]

[0118] This invention is not limited to the embodiments described above, and various modifications can be made within the scope of the claims. Other embodiments obtained by appropriately combining the above embodiments with the disclosed technical solutions are also included within the technical scope of this invention.

Claims

1. A charging system, characterized in that, have: arm; Force sensor; A charging connector is fixed to the arm via the force sensor and can be inserted into the power supply port of the electric vehicle. as well as One or more processors, The processor performs the following steps: The location determination step involves determining the location of the power supply port located on the electric vehicle. A movement control step, wherein the arm is controlled based on the determination result in the position determination step to move the connector to a position facing the power supply port; and An insertion control step, in which the arm is controlled to insert the connector into the power supply port. In the insertion control step, the processor controls the arm based on the detection value of the force sensor, such that the magnitude of one or both of the force and torque in directions other than the insertion direction of the connector is below a threshold. In the insertion control step, the processor detects a fault in the connector or the insertion of a foreign object based on the detection value of the force sensor.

2. The charging system according to claim 1, characterized in that, The force sensor is a 6-axis force sensor.

3. The charging system according to claim 1 or 2, characterized in that, In the location determination step, the processor determines the model of the electric vehicle based on images captured by a sensor used to photograph the electric vehicle, and determines the location of the power supply port based on the determined model.

4. The charging system according to any one of claims 1 to 3, characterized in that, In the location determination step, the processor determines the location of the power supply port based on an image captured by a sensor used to photograph the power supply port.

5. The charging system according to any one of claims 1 to 4, characterized in that, It also includes a conveying mechanism for transporting the arm. In the movement control step, the processor moves the arm to the vicinity of the power supply port via the conveying mechanism, and moves the connector to a position facing the power supply port via the arm.

6. The charging system according to any one of claims 1 to 4, characterized in that, In the insertion control step, if the force sensor detects an external force above a threshold in the insertion direction, the processor determines that the insertion of the connector has been completed.

7. An information processing device, characterized in that, Equipped with one or more processors The one or more processors perform the following steps: The location determination step involves determining the location of the power supply port located on the electric vehicle. The movement control step involves controlling the arm of a charging connector that can be inserted into the power supply port, which is fixed by a force sensor, based on the determination result in the position determination step, so that the connector moves to a position facing the power supply port. as well as An insertion control step, in which the arm is controlled to insert the connector into the power supply port. In the insertion control step, the processor controls the arm based on the detection value of the force sensor, such that the magnitude of one or both of the force and torque in directions other than the insertion direction of the connector is below a threshold. In the insertion control step, the processor detects a fault in the connector or the insertion of a foreign object based on the detection value of the force sensor.

Images