Electric vehicle charging system having reinforcement learning-based autonomous mobile charger, and operation method thereof

Abstract

The present invention relates to an electric vehicle charging system having a reinforcement learning-based autonomous mobile charger, and an operation method thereof, the system being capable of: conveniently and safely charging a plurality of vehicles by driving a mobile charger through autonomous driving using artificial intelligence; minimizing the occupation of parking space in a parking area by, unlike a fixed charging stand, not restricting a portion of parking space as exclusive to electric vehicles; and flexibly adding merely the mobile chargers according to increased demand for electric vehicle charging such that increased demand for charging infrastructure can be addressed at costs lower than that of a conventional charging stand. The electric vehicle charging system having a reinforcement learning-based autonomous mobile charger, of the present invention, comprises: a charging tower for charging a battery; and the mobile charger which uses the battery charged in the charging tower so as to move to a charging service area where an electric vehicle is parked and charge the electric vehicle.

Description

Electric vehicle charging system equipped with a reinforcement learning-based autonomous mobile charger and its operating method

[0001] The present invention relates to an electric vehicle charging system equipped with a mobile charger and an operating method thereof, and more specifically, to a system in which a mobile charger moves within a parking area to perform charging, without limiting a portion of a parking space to be exclusively used for electric vehicles, as is the case with fixed charging stands.

[0002] In addition, the present invention relates to an electric vehicle charging system equipped with an autonomous mobile charger that can conveniently charge by driving the mobile charger autonomously, and an operating method thereof.

[0003] Recently, the automobile industry has been showing increasing interest in electric vehicles (EVs) that use electricity instead of fossil fuels such as gasoline and diesel, and consumer sales are also increasing.

[0004] Electric vehicles are powered by electric energy rather than fossil fuels such as gasoline or diesel, so they are more environmentally friendly than conventional gasoline or diesel vehicles because they produce no exhaust fumes, and they also produce less noise.

[0005] To charge these electric vehicles, charging stands at charging stations, such as gas pumps at gas stations, are generally used. Most electric vehicle charging facilities currently being built are designed to install fixed charging stands in a certain area of ​​a parking space and designate the parking area as an electric vehicle zone.

[0006] However, fixed charging stands use a certain portion of the parking space exclusively for electric vehicles, which is not a major problem in the early stages of deployment, but as the number of vehicles deployed and the occupancy rate gradually increase, a significant number of parking spaces must always be kept empty for electric vehicles, resulting in inefficient use of parking spaces.

[0007] In particular, in a society like ours with a small land area and dense urban centers, if the distribution and market share of electric vehicles exceed a certain level, there is a high possibility that the electric vehicle charging infrastructure of this type will reach a saturation point where it will no longer be able to build a sufficient electric vehicle charging environment.

[0008] Therefore, it is a well-known fact that the main reason why the general public is reluctant to choose electric vehicles is the lack of charging infrastructure and the resulting inconvenience.

[0009] The perfect scenario for charging electric vehicles would be a slow overnight charge in a garage after work, but this is out of reach for most people living in densely populated urban areas.

[0010] Moreover, even in suburban areas with private garages, charging at garages is not easy due to the progressive electricity rate system, so efforts are being made to expand sufficient public charging infrastructure both domestically and internationally.

[0011] Currently, most charging infrastructure consists of rapid or slow charging stations. While these charging stations are not problematic in the current introductory phase when the number of electric vehicles is small, they are expected to present numerous problems in the maturity phase when the electric vehicle market share approaches half.

[0012] If the number of charging stands in public parking lots continues to increase in line with the increasing share of electric vehicles, parking spaces dedicated to electric vehicles with charging stands will seriously infringe on the parking rights of regular vehicles, and charging stands that have received a huge investment will be occupied by vehicles that are already fully charged and parked, or will have low utilization rates that make them useless during empty times.

[0013] In fact, fines are being imposed for illegal parking and obstruction of charging within electric vehicle charging zones, and for new and existing apartments, the installation of electric vehicle chargers is set at 5% and 2%, respectively.

[0014] Moreover, the problem isn't simply the space taken up by electric vehicle charging stations. The increasing number of high-power rapid charging stations is creating serious power supply problems, as the power grid struggles to cope with the increasing demand for charging power.

[0015] Just a few years ago, automotive battery capacities were only around 16 or 20 kWh, but today, they are increasing to over 80 kWh. These large-capacity batteries require a correspondingly high-power charging infrastructure, placing a significant burden on the power supply system.

[0016] Therefore, there is a need for an effective solution to the growing shortage of electric vehicle charging infrastructure.

[0017] As an example, Republic of Korea Patent Publication No. 10-2021-0059094 discloses a mobile charging device for electric vehicles that can charge small electric trucks at night by preventing collisions between charging devices, facilitating maintenance in the event of a charging device failure, and minimizing the risk of safety accidents by ensuring stable transport of the charging device.

[0018] However, even in this case, there is a problem in that the charging device is structured to move through rails on top of the vehicle, and thus does not properly resolve the problems of the current charging infrastructure as described above.

[0019] The purpose of the present invention is to provide an electric vehicle charging system equipped with a reinforcement learning-based autonomous mobile charger capable of conveniently and safely charging multiple vehicles by driving the mobile charger with autonomous driving using artificial intelligence, and an operating method thereof.

[0020] Another purpose of the present invention is to provide an electric vehicle charging system and an operating method thereof, which includes a reinforcement learning-based autonomous driving mobile charger that can minimize the occupancy of parking spaces in a parking area by not limiting a portion of a parking space to be exclusively used for electric vehicles, as in the case of fixed charging stands.

[0021] Another purpose of the present invention is to provide an electric vehicle charging system and an operating method thereof, which includes a reinforcement learning-based autonomous driving mobile charger that can flexibly add only mobile chargers according to the increasing demand for electric vehicle charging, thereby solving the increasing demand for charging infrastructure at a lower cost than existing charging stands.

[0022] An electric vehicle charging system equipped with a reinforcement learning-based autonomous driving mobile charger according to the present invention may include a charging tower that performs battery charging and a mobile charger that uses the battery charged in the charging tower to move to a charging service area where an electric vehicle is parked and performs electric vehicle charging.

[0023] Here, the mobile charger may include a charger body that enables movement, a charging plug that is plugged into an electric vehicle by a user to charge the electric vehicle, and a charging plug retrieval unit that retrieves the charging plug when charging of the electric vehicle is complete.

[0024] Additionally, the mobile charger can autonomously move to a charging service area where a user-designated electric vehicle is parked upon user request.

[0025] Here, the mobile charger can autonomously return to the charging tower when the electric vehicle charging is complete.

[0026] Additionally, when the mobile charger starts moving, a safety warning light can be activated at the start and end points of the mobile charger's moving path to alert the surroundings.

[0027] Here, the driver of an arriving vehicle can reserve charging using a user app (APP) or an HMI (Human Machine Interface) at the charging station while the mobile charger is charging the preceding vehicle.

[0028] Additionally, when the charging of a vehicle being charged is completed, a notification is sequentially sent to the driver who made the charging reservation, and if the driver cannot continue charging in his / her turn, the reservation can be canceled, or the charging turn can be automatically assigned to the next vehicle after a certain period of waiting.

[0029] Here, the driver can be provided with the location of electric vehicle charging stations near the area the driver has searched for, as well as whether the electric vehicle charger is currently in use and, if so, the estimated time until charging is complete, as well as the status of vehicles reserved for charging and the estimated travel time from the driver's current location to each charging station.

[0030] Additionally, the charging plug recovery unit can unplug the charging plug plugged into the charging port of the electric vehicle by the user when charging of the electric vehicle is completed.

[0031] Here, the charging service area can be designated in advance as a certain area of ​​an existing parking lot and used for a limited period.

[0032] Additionally, autonomous driving can be performed by a learning module that is installed in a mobile charger and performs reinforcement learning.

[0033] Here, reinforcement learning can be performed by training a DDQN (Dual Deep Q-Network) based on images recognized by a camera installed in the charger body, surrounding information recognized by a proximity sensor installed in the charger body, and driving information (output) of an actuator that moves the charger body.

[0034] Additionally, the learning module may include a Q-network that performs learning within an episode, and a target Q-network that has the same structure as the Q-network but does not perform learning within an episode but copies and uses the parameters of the Q-network after the end of one episode.

[0035] Here, the Q-network can update the parameters of the Q-network so that the difference between the results of the Q-network and the results of the target Q-network during the episode is minimized.

[0036] Additionally, the episode may include at least one of driving, obstacle avoidance, and stopping of the mobile charger.

[0037] According to another embodiment of the present invention, an operating method of an electric vehicle charging system having a reinforcement learning-based autonomous mobile charger may include: a charger moving step in which a mobile charger moves from a charging tower to an electric vehicle through reinforcement learning-based autonomous driving in response to a user's charging request; an electric vehicle charging step in which an electric vehicle is charged through a charging plug of the mobile charger plugged into a charging port of the electric vehicle by the user; an unplugging step in which the charging plug of the mobile charger plugged into the charging port of the electric vehicle is unplugged by a charging plug retrieval unit of the mobile charger; and a charger returning step in which, after the unplugging step is completed, the mobile charger returns to the charging tower through reinforcement learning-based autonomous driving.

[0038] An electric vehicle charging system and its operating method equipped with a reinforcement learning-based autonomous driving mobile charger according to the present invention can conveniently and safely charge multiple vehicles by driving the mobile charger with autonomous driving using artificial intelligence.

[0039] In addition, the electric vehicle charging system and its operating method equipped with a reinforcement learning-based autonomous driving mobile charger according to the present invention can minimize the occupation of parking spaces in a parking area by not limiting a portion of a parking space to electric vehicles only, as is the case with fixed charging stands.

[0040] In addition, the electric vehicle charging system and its operating method equipped with a reinforcement learning-based autonomous driving mobile charger according to the present invention can flexibly add only the mobile charger according to the increasing demand for electric vehicle charging, thereby solving the increasing demand for charging infrastructure at a lower cost than existing charging stands.

[0041] FIG. 1 is a drawing showing an electric vehicle charging system equipped with a reinforcement learning-based autonomous driving mobile charger according to one embodiment of the present invention.

[0042] FIG. 2 is a drawing showing the mobile charger of FIG. 1 in detail. FIG. 2(a) shows a charging plug being plugged into a charging port of an electric vehicle by a user to perform charging, and FIG. 2(b) shows a charging plug being plugged into a charging port of an electric vehicle being unplugged by a charging plug recovery unit.

[0043] Figure 3 is a drawing showing the location of charging towers installed around the charging service area of ​​Figure 1 and the movement path of a mobile charger.

[0044] FIG. 4 is a drawing for explaining a learning module within a mobile charger for autonomous driving based on reinforcement learning of the mobile charger of FIG. 1.

[0045] FIG. 5 is a flowchart illustrating an operation method of an electric vehicle charging system equipped with a reinforcement learning-based autonomous driving mobile charger according to one embodiment of the present invention.

[0046] Hereinafter, specific embodiments for carrying out the present invention will be described with reference to the attached drawings.

[0047] When describing the present invention, terms such as "first" and "second" may be used to describe various components. However, the components may not be limited by these terms. The terms are used solely to distinguish one component from another. For example, without departing from the scope of the present invention, the first component could be referred to as the "second component," and similarly, the second component could also be referred to as the "first component."

[0048] When it is said that a component is connected or connected to another component, it can be understood that it may be directly connected or connected to that other component, but there may also be other components in between.

[0049] The terminology used in this specification is for the purpose of describing specific embodiments only and is not intended to limit the present invention. Singular expressions may include plural expressions unless the context clearly dictates otherwise.

[0050] In this specification, terms such as “include” or “have” are intended to indicate the presence of a feature, number, step, operation, component, part, or combination thereof described in the specification, and can be understood as not excluding in advance the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0051] Additionally, the shape and size of elements in the drawing may be exaggerated for clearer explanation.

[0052] Hereinafter, with reference to the attached drawings, an electric vehicle charging system equipped with a reinforcement learning-based autonomous driving mobile charger according to the present invention and its operating method will be described in detail.

[0053]

[0054] FIG. 1 is a drawing showing an electric vehicle charging system equipped with a reinforcement learning-based autonomous driving mobile charger according to one embodiment of the present invention, and FIGS. 2 to 4 are detailed drawings for explaining FIG. 1 in detail.

[0055] Hereinafter, an electric vehicle charging system equipped with a reinforcement learning-based autonomous driving mobile charger according to the present invention will be described with reference to FIGS. 1 to 4.

[0056] First, referring to FIG. 1, the electric vehicle charging system equipped with a reinforcement learning-based autonomous driving mobile charger of the present invention is composed of a charging tower (100) that performs charging of a battery (240) and a mobile charger (200) that uses the battery (240) charged in the charging tower (100) to move to a charging service area (300) where an electric vehicle (500) is parked and performs electric vehicle charging.

[0057] Here, the mobile charger (200) is equipped with a charger body (210) that enables movement, a charging plug (230) that is plugged into an electric vehicle (500) by a user to charge the electric vehicle, and a charging plug retrieval unit (220) that retrieves the charging plug (230) when the electric vehicle is fully charged. When the user requests, the mobile charger (200) can autonomously move to a charging service area (300) where the electric vehicle (500) is parked and, when the electric vehicle charging is fully completed, autonomously return to the charging tower (100).

[0058] According to an implementation example, safety warning lights are installed at the starting and ending points of the moving section of the mobile charger to alert the surroundings, and when the mobile charger (200) starts moving, the safety warning lights are operated in the same way as the entry and exit of vehicles at the entrance and exit of an underground parking lot, thereby further ensuring the safety of surrounding vehicles and drivers.

[0059] FIG. 2 is a drawing showing in detail the mobile charger (200) of FIG. 1. FIG. 2(a) shows a charging plug (230) being plugged into a charging port (510) of an electric vehicle (500) by a user to perform charging, and FIG. 2(b) shows a charging plug (230) plugged into a charging port (510) of an electric vehicle (500) being unplugged by a charging plug recovery unit (220).

[0060] As can be seen in Fig. 2, when charging of an electric vehicle is completed, the mobile charger (200) of the present invention automatically unplugs and retrieves the charging plug (230) plugged into the charging port (510) of the electric vehicle (500) by the user using the charging plug recovery unit (220).

[0061] The present invention enables the plug-in operation of the charging plug (230) to be performed by the user, thereby enabling the implementation of a portable charger (200) with a simpler structure and more economical cost compared to the existing one.

[0062] In addition, after the electric vehicle charging is completed, the charging plug retrieval unit (220) of the mobile charger (200) automatically separates the charging plug (230), so there is an advantage that the user does not have to wait until charging is completed.

[0063]

[0064] FIG. 3 is a drawing showing the location of a charging tower (100) installed around a charging service area (300) of FIG. 1 and the movement path (400) along which a mobile charger (200) travels.

[0065] As can be seen in Fig. 3, the present invention pre-designates a certain area of ​​an existing parking lot as a charging service available area (300) and uses it for limited purposes.

[0066] That is, the electric vehicle charging system of the present invention can be used by designating in advance a certain area of ​​an existing parking lot, for example, a predetermined area consisting of approximately 10 to 15 parking spaces, as a charging service available area (300).

[0067] The electric vehicle charging system of the present invention requires manual plug-in by the user, and therefore, if the entire parking lot is set as a charging service area, a situation may arise where the user has to wait for a considerably long time.

[0068] In addition, if the entire parking lot is designated as a charging service area, the mobile charger will require a high level of autonomous driving performance, such as recognizing all surrounding objects on an irregular driving path and predicting and avoiding their movement in advance, which will require significant technical barriers and costs for system implementation and construction.

[0069] Therefore, in the present invention, a certain area of ​​an existing parking lot, that is, an area consisting of approximately 10 to 15 parking spaces as described above, is designated in advance as a charging service available area (300) and used.

[0070] Through this, the present invention not only drastically reduces the time a user spends waiting for a mobile charger to arrive after parking a car, but also simplifies the route and driving mechanism of the mobile charger, thereby drastically reducing the costs of system design, implementation, and construction.

[0071] Meanwhile, a magnetic tape or other marker may be installed on the moving path (400) so that the mobile charger (200) is guided therethrough and driven along the set path.

[0072] In this way, the present invention simplifies the movement path and mechanism of the mobile charger (200) so that the mobile charger (200) can be implemented in a manner in which it stops and waits only when it detects an obstacle or dangerous situation in a given driving direction and then drives again, making it very easy to ensure the driving safety of the mobile charger (200).

[0073] Meanwhile, in the present invention, it is preferable to designate a certain area of ​​an existing parking lot designated as a charging service area (300) from the outermost area of ​​the parking lot's business building, that is, the area least preferred by drivers parking their vehicles in the parking lot.

[0074] Through this, the present invention allows for the smooth charging of electric vehicles by minimizing the parking of general vehicles while allowing the parking of general vehicles in the charging service area without designating the charging service area as an electric vehicle-only parking area.

[0075] According to an embodiment, the electric vehicle charging system equipped with a reinforcement learning-based autonomous driving mobile charger of the present invention can further improve the efficiency of use of the system by providing a method in which a driver can reserve charging for a vehicle arriving while the charging system is charging a preceding vehicle using a user app (APP) or an on-site human machine interface (HMI).

[0076] Through this, the electric vehicle charging system of the present invention can significantly alleviate the increase in charging demand power, while further improving the efficiency of facility use through a charging reservation function for vehicles entering during charging, and can obtain an even greater alleviation effect of the increase in demand power.

[0077] In addition, the electric vehicle charging system equipped with the autonomous driving mobile charger based on reinforcement learning of the present invention can sequentially send notifications to drivers who have reserved charging when the charging of a vehicle in the middle of a charging service is completed, so that the charging service can continue smoothly. In addition, if a driver encounters a situation where he or she cannot continue charging in his or her turn, he or she can cancel the reservation to yield the charging turn to the next vehicle, or the charging turn can be automatically assigned to the next vehicle after a certain period of time systematically, so as to prevent long-term congestion in the charging service.

[0078] Meanwhile, depending on the implementation example, the electric vehicle charging system of the present invention may adopt a pre-charging reservation method to improve user convenience and facility utilization rate.

[0079] That is, when an electric vehicle driver anticipates the need to charge the vehicle while moving, it is common to look for the location of an electric vehicle charging station where charging is possible in advance, and therefore, the electric vehicle charging system of the present invention can provide the driver with not only the location of an electric vehicle charging station near the area inquired, but also whether the electric vehicle charger is currently in use and, if so, the estimated time until charging is completed, and can also provide the status of vehicles reserved for charging and the estimated travel time from the driver's current location to each charging station.

[0080] Through this, electric vehicle drivers can synthesize all of this information from the electric vehicle charging system to predict their arrival time and reserve charging services in advance. They can also cancel or change the reservation time at any time if their schedule changes during the trip.

[0081] The electric vehicle charging system equipped with the autonomous mobile charger based on reinforcement learning according to the present invention as described above can contribute to the early diffusion and stabilization of the electric vehicle charging system by simplifying the movement line of the mobile charger and providing a method for ensuring the safety of the electric vehicle charging system, thereby lowering the entry barriers such as various regulations and certifications, and can also secure customers using the electric vehicle charging system and increase the facility utilization rate by improving user convenience such as charging reservation and advance reservation of the electric vehicle charging system, while promoting market expansion.

[0082]

[0083] Next, Fig. 4 is a drawing for explaining a learning module within a mobile charger (200) for autonomous driving based on reinforcement learning of the mobile charger (200) of Fig. 1.

[0084] As can be seen in Fig. 4, autonomous driving in the mobile charger (200) of the present invention is performed by a learning module (600) that performs reinforcement learning within the mobile charger (200).

[0085] Here, reinforcement learning can be performed by training a DDQN (Double Deep Q-Network) based on images recognized by a camera provided within the charger body (210), surrounding information recognized by a proximity sensor provided within the charger body (210), and driving information (output) of an actuator that moves the charger body (210).

[0086] At this time, an ultrasonic sensor can be used as a representative proximity sensor.

[0087] In addition, DDQN is a complement to DQN (Deep Q-Network), which has only one learning model, and has the advantage of being able to learn stably.

[0088] That is, the learning module (600) of the present invention is equipped with two learning models, namely, a Q-network (610) that mainly performs learning within an episode, and a target Q-network (620) that has the same structure as the Q-network (610) but does not perform learning within an episode and copies and uses the parameters of the Q-network (610) after the end of one episode.

[0089] Here, the target Q-network (620) can be prevented from learning in an indefinite state by maintaining fixed parameters without performing learning during the episode, and for example, parameter updates of the Q-network (610) can be performed so that the difference between the results of the Q-network (610) and the results of the target Q-network (620) during the episode is minimized.

[0090] Meanwhile, the input of the learning module (600) of the present invention is the camera's image and the surrounding information of the proximity sensor, and the action can be distinguished by the actuator's operation signal. In addition, an episode may include at least one of driving, obstacle avoidance, and stopping of the mobile charger (200).

[0091] The electric vehicle charging system equipped with the autonomous mobile charger based on reinforcement learning of the present invention as described above enables autonomous driving for the mobile charger (200) and learns about it using DDQN (Double Deep Q-Network), thereby stably updating the parameters of the learning module (600).

[0092]

[0093] FIG. 5 is a flowchart illustrating an operation method of an electric vehicle charging system equipped with a reinforcement learning-based autonomous driving mobile charger according to one embodiment of the present invention.

[0094] As can be seen in Fig. 5, the operation method of the electric vehicle charging system having a reinforcement learning-based autonomous mobile charger includes a charger movement step (S100) in which a mobile charger (200) moves from a charging tower (100) to an electric vehicle (500) by reinforcement learning-based autonomous driving upon a user's charging request, an electric vehicle charging step (S200) in which electric vehicle charging is performed through a charging plug (230) of a mobile charger (200) plugged into a charging port (510) of an electric vehicle (500) by a user, an unplugging step (S300) in which the charging plug (230) of the mobile charger (200) plugged into the charging port (510) of the electric vehicle (500) by a charging plug retrieval unit (220) of the mobile charger (200) is unplugged, and after the unplugging step (S300) is completed, the mobile charger (200) moves to the charging tower (100) by reinforcement learning-based autonomous driving. It consists of a charging station return stage (S400) where the vehicle returns to autonomous driving.

[0095] Here, reinforcement learning-based autonomous driving in the charger call stage (S100) and the charger return stage (S400) is performed by a learning module (600) that performs reinforcement learning within the mobile charger (200) as described in FIG. 4, and since this has been described previously, a detailed description thereof will be omitted.

[0096] In addition, the location of the charging tower (100) in the charger call step (S100) and the charger return step (S400) and the travel path (400) along which the mobile charger (200) travels can be limited to a certain area of ​​an existing parking lot designated in advance as a charging service available area (300) as described in FIG. 3, and since this has been described previously, a detailed description thereof will be omitted.

[0097] Meanwhile, in the unplugging step (S300), as mentioned in FIG. 2, when electric vehicle charging is completed, the charging plug (230) plugged into the charging port (510) of the electric vehicle (500) by the user is automatically unplugged by the charging plug retrieval unit (220) of the mobile charger (200).

[0098] In this way, the present invention has the advantage of implementing a portable charger (200) with a simple structure and economical cost compared to existing ones by having the user perform the plug-in operation of the charging plug (230), and further, after electric vehicle charging is completed, the charging plug retrieval unit (220) of the portable charger (200) automatically separates the charging plug (230), so that the user does not have to wait until charging is completed.

[0099]

[0100] As described above, the electric vehicle charging system and its operating method, which include a reinforcement learning-based autonomous driving mobile charger according to the present invention, can conveniently and safely charge multiple vehicles by operating the mobile charger through autonomous driving using artificial intelligence. In addition, unlike fixed charging stands, a portion of the parking space is not restricted to electric vehicles, thereby minimizing the occupation of parking spaces within the parking area. Furthermore, since mobile chargers can be flexibly added according to the increasing demand for electric vehicle charging, the system can address the growing demand for charging infrastructure at a lower cost than existing charging stands.

[0101]

[0102] Those skilled in the art will appreciate that the various illustrative logical blocks, modules, processors, means, circuits, and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, various forms of programs or design code (referred to herein, for convenience, as software), or a combination of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Those skilled in the art may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.

[0103] The various embodiments presented herein can be implemented as a method, apparatus, or article of manufacture using standard programming and / or engineering techniques. The term article of manufacture includes a computer program, carrier, or media accessible from any computer-readable storage device. For example, computer-readable storage media include, but are not limited to, magnetic storage devices (e.g., hard disks, floppy disks, magnetic strips, etc.), optical disks (e.g., CDs, DVDs, etc.), smart cards, and flash memory devices (e.g., EEPROMs, cards, sticks, key drives, etc.). Furthermore, various storage media presented herein include one or more devices and / or other machine-readable media for storing information.

[0104] It should be understood that the specific order or hierarchy of steps in the presented processes is merely an example of exemplary approaches. It should be understood that the specific order or hierarchy of steps in the processes may be rearranged within the scope of the present invention based on design priorities. The appended method claims provide elements of various steps in a sample order, but are not intended to be limited to the specific order or hierarchy presented.

[0105] The description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other embodiments without departing from the scope of the present invention. Therefore, the present invention is not intended to be limited to the embodiments disclosed herein, but is to be construed in the broadest scope consistent with the principles and novel features disclosed herein.

[0106] The present invention relates to an electric vehicle charging system equipped with a portable charger and an operating method thereof, and can be used in the field of electric vehicle chargers.

Claims

1. A charging tower that performs battery charging; and An electric vehicle charging system having a reinforcement learning-based autonomous driving mobile charger, including a mobile charger that moves to a charging service area where an electric vehicle is parked and charges the electric vehicle using the battery charged at the charging tower.

2. In paragraph 1, The above portable charger, Charger body that enables movement; A charging plug that is plugged into the electric vehicle by a user to charge the electric vehicle; and An electric vehicle charging system equipped with a reinforcement learning-based autonomous driving mobile charger, characterized in that it includes a charging plug retrieval unit that retrieves the charging plug when the electric vehicle charging is completed.

3. In paragraph 2, The above portable charger, An electric vehicle charging system equipped with a reinforcement learning-based autonomous mobile charger, characterized in that the electric vehicle moves autonomously to a charging service area where the electric vehicle designated by the user is parked when requested by the user.

4. In paragraph 3, The above portable charger, An electric vehicle charging system equipped with a reinforcement learning-based autonomous driving mobile charger, characterized in that when the above electric vehicle charging is completed, the charger autonomously drives and returns to the charging tower.

5. In paragraph 1, An electric vehicle charging system equipped with a reinforcement learning-based autonomous mobile charger, characterized in that when the mobile charger starts moving, a safety warning light is activated at the starting and ending points of the moving path of the mobile charger to alert the surroundings.

6. In paragraph 1, An electric vehicle charging system equipped with a reinforcement learning-based autonomous driving mobile charger, characterized in that the driver of an arriving vehicle can reserve charging using a user app (APP) or an HMI (Human Machine Interface) at the charging station while the mobile charger is charging a preceding vehicle.

7. In paragraph 6, When the charging of a vehicle that was in the process of charging service is completed, a notification is sent sequentially to the driver who made the charging reservation. An electric vehicle charging system equipped with a reinforcement learning-based autonomous driving mobile charger, characterized in that if the driver above encounters a situation where he or she cannot continue charging in his or her turn, he or she can cancel the reservation, or automatically assign the charging turn to the next vehicle after waiting for a certain period of time.

8. In paragraph 6, An electric vehicle charging system equipped with a reinforcement learning-based autonomous driving mobile charger, characterized in that it provides the driver with the location of an electric vehicle charging station near the area searched by the driver, as well as whether the electric vehicle charger is currently in use and, if so, the estimated time until charging is completed, and also provides the status of vehicles reserved for charging and the estimated travel time from the driver's current location to each charging station.

9. In paragraph 2, The above charging plug recovery part, An electric vehicle charging system having a reinforcement learning-based autonomous driving mobile charger, characterized in that when the electric vehicle charging is completed, the charging plug plugged into the charging port of the electric vehicle by the user is unplugged.

10. In paragraph 1, The above charging service area is: An electric vehicle charging system equipped with a reinforcement learning-based autonomous mobile charger that features pre-designated and limited use of a certain area of ​​an existing parking lot.

11. In paragraph 4, The above autonomous driving is, An electric vehicle charging system having a reinforcement learning-based autonomous driving mobile charger, characterized in that reinforcement learning is performed by a learning module equipped in the mobile charger and performing reinforcement learning.

12. In paragraph 11, The above reinforcement learning is, An electric vehicle charging system equipped with a reinforcement learning-based autonomous driving mobile charger, characterized in that it is performed by training a DDQN (Dual Deep Q-Network) based on images recognized by a camera equipped in the charger body, surrounding information recognized by a proximity sensor equipped in the charger body, and driving information (output) of an actuator that moves the charger body.

13. In paragraph 12, The above learning module, Q-network that performs learning within an episode; and An electric vehicle charging system having a reinforcement learning-based autonomous driving mobile charger, characterized in that it includes a target Q-network having the same structure as the above Q-network but copying and using the parameters of the Q-network after the end of one episode without performing learning within the episode.

14. In paragraph 13, The above Q-network is, An electric vehicle charging system having a reinforcement learning-based autonomous driving mobile charger, characterized in that the parameters of the Q-network are updated so that the difference between the results of the Q-network and the results of the target Q-network during the above episode is minimized.

15. In paragraph 14, The above episode, An electric vehicle charging system having a reinforcement learning-based autonomous mobile charger, characterized in that it includes at least one of driving, obstacle avoidance, and stopping of the mobile charger.

16. Charger movement stage where a mobile charger moves autonomously from a charging tower to an electric vehicle parked in a charging service area based on reinforcement learning at the user's charging request; An electric vehicle charging step in which electric vehicle charging is performed through a charging plug of the mobile charger plugged into the charging port of the electric vehicle by the user; An unplugging step of unplugging the charging plug of the mobile charger plugged into the charging port of the electric vehicle by the charging plug recovery unit of the mobile charger; and An operating method of an electric vehicle charging system having a reinforcement learning-based autonomous mobile charger, comprising a charger returning step in which the mobile charger returns to the charging tower by autonomous driving based on reinforcement learning after the unplugging step is completed.

Images