TWI870046B - Wireless charging system and method for unmanned vehicles - Google Patents

Abstract

A wireless charging system and method for unmanned vehicles. The system comprises at least one unmanned aerial vehicle (UAV) and a power supply device. UAVs supply power for a plurality of unmanned vehicles (UVs). The power supply device monitors the power information and real-time locations of the UVs, and judges whether to send status confirmation messages to the UAV according to the power information. After receiving the status confirmation messages, the UAVs send the remaining power information to the power supply device. The power supply device judges that the remaining power of the UAVs can be used to charge the UVs, and performs weighted calculations based on multiple charging condition parameters to generate a charging schedule. Then, according to the charging schedule, a charging command is sent to the UAVs can charge the UVs, so that they can charge the UVs according to the order of the charging schedule. With the present invention, the UVs can be charged in the most power-saving condition.

Description

Wireless charging system and method for electric-driven vehicle

The present invention relates to a wireless charging method, and more particularly to a wireless charging system and method for an electric-driven vehicle.

Smart factories can save a lot of manpower by using automated guided vehicles (AGVs). Traditional automated guided vehicle systems mostly use plug-in charging systems. The disadvantage is that if the automated guided vehicle runs out of power, it must stop working and go to the charging pile to charge, which will significantly reduce the utilization rate of the automated guided vehicle. Generally speaking, it takes 4 hours to fully charge an automated guided vehicle. If there are other automated guided vehicles charging at the charging station, they must queue up to wait for charging, resulting in more time wasted. In addition, due to frequent plugging and unplugging actions, static electricity is easily generated and the connectors are damaged, resulting in high maintenance costs.

If wireless charging is used, since AGVs usually need to operate continuously, in order to avoid the waste of utilization rate caused by the AGVs needing to frequently go to charging stations for charging, it is best to perform wireless charging during the movement process, that is, embed a set of power transmitters under the workplace, and install a receiver at the bottom of the AGV, so that the AGV can be charged while working. However, the RF signal will seriously degrade the signal strength as the distance increases, that is, the charging efficiency of the AGV will decrease exponentially as the distance increases, so the charging efficiency of the wireless charging device is greatly affected by the distance, and the AGV needs to be within a certain distance from the wireless charging device to be effectively charged. Therefore, a large number of wireless charging devices must be deployed, which will greatly increase the cost and thus hinder the implementation of the wireless charging system.

In view of this, the present invention proposes a wireless charging system and method for an electric-driven vehicle in view of the above-mentioned deficiencies in the prior art and future needs. The specific structure and implementation method thereof are described in detail below:

The main purpose of the present invention is to provide a wireless charging system and method for an electric-powered vehicle, which utilizes a drone to provide wireless charging for the electric-powered vehicle without plugging and unplugging the charging plug, thereby avoiding the problem of socket damage and eliminating the need to configure a large number of wireless charging stations.

Another object of the present invention is to provide a wireless charging system and method for an electric-powered vehicle, which utilizes a drone to automatically go to the location of the electric-powered vehicle to charge the electric-powered vehicle, thereby saving the time of the electric-powered vehicle going back and forth to a charging station.

Another object of the present invention is to provide a wireless charging system and method for an electric-powered vehicle, which schedules each electric-powered vehicle that needs to be charged, and then allows a drone to charge the electric-powered vehicle according to the schedule, thereby solving the problem of wasting time when the electric-powered vehicle queues at a charging station to wait for charging.

To achieve the above-mentioned purpose, the present invention provides a wireless charging system for an electric-powered vehicle, comprising: at least one drone for supplying power to a plurality of electric-powered vehicles; and a power supply device, comprising a power module and a scheduling processor, for monitoring power information and a real-time position of the electric-powered vehicle, and determining whether to send a status confirmation message to the drone according to the power information, wherein the power information indicates the current power of each electric-powered vehicle; In the process, after receiving the status confirmation message, the drone transmits a remaining power information to the power supply device. The scheduling processor in the power supply device determines whether the remaining power of the drone meets a charging requirement based on the remaining power information, and performs weighted calculation based on multiple charging parameters to generate a charging schedule. According to the charging schedule, a charging command is sent to the drone that meets the charging requirement, so that it can charge the power-driven vehicle according to the order of the charging schedule.

According to an embodiment of the present invention, when the remaining power of the drone is less than a preset value, the drone automatically goes to a power supply device for charging.

According to an embodiment of the present invention, when the distance between the drone and the power supply device is less than a preset distance, the drone automatically goes to the power supply device for charging.

According to an embodiment of the present invention, the charging parameters include the distance between the electric-powered vehicle and the power supply device, the remaining power of the electric-powered vehicle, the remaining power of the drone, and the floating electricity price.

According to an embodiment of the present invention, the power-driven vehicle continuously sends power information to the power supply device. When the power information is lower than a threshold value, the power supply device automatically sends a status confirmation message to the at least one drone.

According to an embodiment of the present invention, the charging requirement is that the remaining power of the drone reaches more than 80%.

According to an embodiment of the present invention, the scheduling processor uses a feature-based expected SARSA (Feature-based Expected SARSA, Feature-based Expected State-Action-Reward-State-Action) algorithm with reinforcement learning to perform charging optimization management of the drone.

A wireless charging method for a power-driven vehicle is applied to a power supply device. The wireless charging method for a power-driven vehicle includes the following steps: monitoring power information and a real-time position of a plurality of power-driven vehicles, and transmitting a status confirmation message to at least one drone according to the power information, wherein the power information indicates the current power of each power-driven vehicle; receiving a response from the drone; A remaining power information is outputted in response to a status confirmation message; judging whether a remaining power of the drone meets a charging requirement based on the remaining power information, and performing weighted calculation based on a plurality of charging parameters to generate a charging schedule; and sending a charging command to the drone that meets the charging requirement based on the charging schedule, so that the drone charges the power-driven vehicle according to the order of the charging schedule.

The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

It should be understood that when used in this specification and the appended patent applications, the terms "include" and "comprising" indicate the presence of described features, wholes, steps, operations, elements and/or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and/or combinations thereof.

It should also be understood that the terms used in this specification are for the purpose of describing specific embodiments only and are not intended to limit the present invention. As used in this specification and the appended patent applications, the singular forms "a", "an" and "the" are intended to include plural forms unless the context clearly indicates otherwise.

It should be further understood that the term "and/or" used in this specification and the appended patent application refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

The present invention provides a wireless charging system and method for an electric-powered vehicle. Please refer to FIG. 1 and FIG. 2, which are respectively a block diagram and a schematic diagram of an embodiment of the wireless charging system 10 for an electric-powered vehicle of the present invention. The wireless charging system 10 for an electric-powered vehicle of the present invention includes a power supply device 12, a plurality of electric-powered vehicles 14, and at least one drone 16. Among them, the electric-powered vehicle 14 is an automated guided vehicle (AGV), an electric car, an electric boat, or other mobile vehicles driven by electricity. In one embodiment, the present invention can be applied to the working environment of a smart factory 20, in which case the electric-powered vehicle 14 is an automatic working robot that is not operated by humans, such as an automatic guided vehicle. The power-driven vehicle 14 works on the machines 22a and 22b, and reciprocates along a fixed route in the mobile area 24. The machines 22a and 22b, the power-driven vehicle 14, the drone 16, and the power supply device 12 are all Industrial Internet of Things (IIoT) devices, and can communicate with each other. The drone 16 uses wireless charging to supply power to the power-driven vehicle 14. When the number of drones 16 is plural, all drones 16 can form a fleet. Generally speaking, the drone 16 will be on standby near the power supply device 12. The power supply device 12 is a charging station that provides power and has the ability to send and receive messages and calculate scheduling. When the power-driven vehicle 14 is working in the smart factory 20, if the power is insufficient and needs to be charged, the power-driven vehicle 14 will transmit power information and real-time location information to the power supply device 12. The power supply device 12 is equivalent to the decision center of the system. After receiving the power information, it will determine when to dispatch the fleet of drones 16 to perform the mission based on the real-time power information. Then the fleet of drones 16 will move to the location of the power-driven vehicle 14, and transmit power to the power-driven vehicle 14 using wireless power transfer (WPT) technology. Before the power of the drone 14 is exhausted, it will return to the power supply device 12 for charging. Finally, the drone 16 returns to the fleet of drones 16.

In the embodiment of FIG. 2 , the power supply device 12 is set in a corner of the smart factory 20, but this is not a limitation. In fact, the power supply device 12 can be set at any position of the smart factory 20, and the best position is in the middle of the smart factory 20, so that the distance of the drone 16 to any power-driven vehicle 14 will not exceed half of the diagonal line, which saves the power of the drone. The power supply device 12 is at least one of a wireless charging station, an automatic guided vehicle charging station, and a UAV-mounted wireless power transmission device (WPT).

The power supply device 12 includes a power module 122 and a scheduling processor 124, wherein the power module 122 includes at least one lithium battery or other power storage device, and can also be connected to the mains via a wire to replenish the power in the power module 122 at any time. The scheduling processor 124 is used to monitor the power information of the power-driven vehicle 14, and determine whether to send a status confirmation message to the drone based on the power information. The power information indicates the current power of each power-driven vehicle 14.

Please refer to FIG. 3, which is a flow chart of the wireless charging method of the power-driven vehicle of the present invention. In steps S10-S12, the power supply device 12 monitors the power information of the power-driven vehicle 14. The specific monitoring method is that the power-driven vehicle 14 continuously sends the power information and real-time position to the power supply device 12, and the scheduling processor 124 confirms whether the power information of each power-driven vehicle 14 is still at a threshold value. If the power is at the threshold value, it means that the power-driven vehicle 14 does not need to be charged yet. However, if the power information of a certain power-driven vehicle 14 is determined in step S12 to show that its power is less than or equal to the threshold value, for example, only 20% of the power is left, the power supply device 12 activates the charging mechanism. In step S14, the power supply device 12 sends a status confirmation message to all drones 16 to check whether the power of each drone 16 is sufficient to supply power to the power-driven vehicle 14. Therefore, in step S16, after receiving the status confirmation message, the drone 16 responds to the status confirmation message and replies with a remaining power information to the power supply device 12. Then, as described in step S18, the scheduling processor 124 determines whether the remaining power of the drone 16 meets a charging requirement based on the remaining power information, and performs weighted calculations based on multiple charging parameters to generate a charging schedule. In step S20, the scheduling processor 124 sends a charging command to the drone 16 that meets the charging requirement, that is, the drone 16 that has sufficient power to power the power-driven vehicle 14. If there are multiple drones 16 that meet the conditions, the scheduling processor 124 can select the drone 16 with the most remaining power. Finally, as described in step S22, the drone 16 that receives the charging command charges the power-driven vehicle 14 in the order of the charging schedule.

In the present invention, the remaining power of the electric-powered vehicle 14 and the drone 16 are expressed in stages, for example, 80% or more represents sufficient power, 60-80% represents acceptable charging, 20-60% represents dangerous power, and below 20% represents insufficient power and stops operation. Therefore, the threshold value of the electric-powered vehicle 14 in step S12 can be set to 80% of the total power, and the charging requirement of the drone 16 in step S18 is that the remaining power is greater than 80%.

The scheduling of the present invention is based on the principle of the lowest electricity cost. How to charge while saving the most electricity and having the least impact on work, that is, combining the electricity cost with the utilization rate loss cost, is the result that the scheduling processor 124 wants to obtain when calculating the charging schedule. Therefore, in addition to the remaining power of the drone 16, the charging parameters also include the distance between the power-driven vehicle 14 and the power supply device 12, the remaining power of the power-driven vehicle 14, and the floating electricity price. For example, the power-driven vehicle 14 with the least remaining power is charged first, the power-driven vehicle 14 closest to the power supply device 12 is charged first, or when the floating electricity price is lower than a certain price, all power-driven vehicles 14 are fully charged first. When the power of the electric vehicle 14 is less than 80%, it is not necessary to immediately dispatch the drone 16 to replace the battery, because the current location of the electric vehicle 14 may be very far from the power supply device 12. It is possible that it will be closer to the power supply device 12 before the power is less than 20%. In this way, the power consumption of the drone 16 during the transportation process will be relatively low. However, it is also necessary to pay attention to the electric vehicle 14 as much as possible not to stop operating due to the power being less than 20%, resulting in a loss of utilization rate. If the electric vehicle 14 stops operating due to insufficient power, the scheduling processor 124 also sets a penalty value and calculates the charging schedule in the charging parameters.

The above is a schedule for the drone 16 to charge the power-driven vehicle 14. In addition, the drone 16 also has its own charging schedule. For example, when the remaining power of the drone 16 is less than a preset value, it will automatically go to the power supply device 12 for charging. The preset value can be 80% of the total power, so that it can receive a charging command at any time to go for power supply. Or when the distance between the drone 16 and the power supply device 12 is less than a preset distance, it means that the drone 16 is very close to the power supply device 12. At this time, the drone 16 can also automatically go to the power supply device 12 for charging.

In the present invention, the scheduling processor uses a feature-based expected SARSA (Feature-based Expected State-Action-Reward-State-Action) algorithm based on reinforcement learning to perform charging optimization management of the drone to solve the scheduling problem of wireless charging of the power-driven vehicle 14 through the drone 16 in the smart factory 20. The main concept of the algorithm is to reduce the state and action space by weighting multiple feature functions. The number of features can be set and adjusted according to the requirements of accuracy and efficiency. The reinforcement learning algorithm based on Expected SARSA is used to solve the problem. This algorithm is calculated as the data changes and does not require information about the future work schedule of the power-driven vehicle 14 and the distribution of electricity prices.

Traditional reinforcement learning algorithms cannot be directly applied to this problem. This is because the dimension of the state space _St is proportional to the number of electric vehicles 14 and drones | _Kt | that are dispatched to work in the system, and changes with the operation of the smart factory 20. To solve this problem, the present invention proposes a reinforcement learning algorithm of Expected SARSA with binary linear characteristic function approximation.

Since the distribution of the next AGV and UAV requests is unknown, the state-value function is Replaced with a state-action value function ,in, is the status value, is the action value, given by the reward function and the state space for transitioning to the next state Specifically, it is updated in the following way: ← ) [ + γ ( , a )] , where is the learning rate. In addition, in order to balance the relationship between exploration and exploitation, a - Greedy strategy, in which the power supply device 12 has a probability Select Maximize The best action of , and randomly selects an action with probability ϵ. ,0＜ ＜ 1.

Traditional reinforcement learning methods require that all Learn and save Q However, since the problem state and action space of the present invention are continuous, if the quantization level is relatively small, discretizing it will lead to an extremely large Q table, which is unacceptable. Therefore, the present invention uses the characteristic function , y =1,…, Y is linearly combined to approximate Q .Q Approximate value of is the weighted sum of the eigenvalues multiplied by their individual weights. Specifically, ,in, := is the weight of the eigenvalue. Therefore, the algorithm proposed by the present invention only needs to learn and record , rather than for all The Q table is stored. The characteristic function will search the space from a Q that changes over time. The table is reduced to Y values.

Then construct the characteristic function based on the objective function and constraint function of the problem ,…, However, the defined eigenfunction may not be convergent. In addition, the eigenfunction can take any real number, which may increase the computation and storage cost. In order to force convergence and reduce the computation and storage cost, the present invention converts the previously defined eigenfunction into a binary eigenfunction. Specifically, The moving averages of If the current reward value is not less than the moving average , then is set to 1; otherwise, it is set to 0. When the binary feature approximation function is well defined, each iteration updates only the weights because they control the relationship between each eigenfunction and contribution.

Based on the above-mentioned enhanced learning algorithm of Expected SARSA, since the goal of the present invention is to minimize the electricity cost plus the utilization rate loss cost, the formula is as follows: in, is the real-time electricity price, is the charging rate of the drone 16, is whether the electric-driven vehicle 14 stops or not, P is the unit penalty cost, represents a collection of drones 16 in the waiting area, Represents a collection of power-driven vehicles 14 working in the machine 22a, 22b or the mobile area 24. The charging parameters (i.e., characteristic functions) of the present invention are as follows: 1. Calculate the total charging cost: ; 2. Calculate the penalty value when the electric-driven vehicle 14 stops due to insufficient power: ; 3. Use geometric progression to weight, sort all drones 16 in the waiting area in ascending order of battery charge state, and classify drones 16 with the same battery charge state into the same category. The drones 16 with the larger battery charge state in the waiting area are given priority to be charged by the power supply device 12, where c _it is the battery charge state of the drone 16, θ ₂ is the common ratio of the geometric series, is the total number of categories, is the battery charge status category index of the drone 16 i at time t : 4. Use an arithmetic progression to weight the electric-powered vehicle 14 that is closer to the electric power supply device 12 to be charged first, where c _jt is the battery charge state of the electric-powered vehicle 14, θ ₁ is the tolerance of the arithmetic progression, is the total number of categories, is the category index of the distance between the electric-powered vehicle 14 i and the charging station at time t : 5. Use geometric progression to weight the battery of the electric-powered vehicle 14 with the smaller state of charge to be charged first, where is the total number of categories, is the battery charge status category index of the electric-powered vehicle 14 j at time t : 6. Will Convert to binary values of 0 and 1 , among which, greater than The average of the last 20 times is 1, otherwise it is 0.

After calculating each charging parameter according to the above charging parameter formula, the weight value of reinforcement learning is updated by weighted calculation of expected value to obtain the optimal charging schedule.

In summary, the present invention provides a wireless charging system and method for an electric-driven vehicle, wherein the power supply device provides power to the drone, and the drone then goes to the electric-driven vehicle to charge the electric-driven vehicle. The advantage of this is that wireless charging can avoid static damage caused by repeated plugging and unplugging of the connector. In addition, the electric-driven vehicle with a slower speed does not need to move to the power supply device, and needs to move back to the working route after being fully charged, which can save a lot of time; while the drone with a faster flying speed can quickly fly to the location of the electric-driven vehicle to charge, allowing the electric-driven vehicle to be charged earlier. The advantages of using drone charging instead of electric vehicles charging themselves include: electric vehicles do not need to reserve the power required along the way to the power supply device, and can work for a longer time before charging.

However, the above is only the preferred embodiment of the present invention and is not intended to limit the scope of the present invention. Therefore, all equivalent changes or modifications based on the features and spirit described in the scope of the present invention should be included in the scope of the patent application of the present invention.

10: Wireless charging system for power-driven vehicles 12: Power supply device 122: Power module 124: Scheduling processor 14: Power-driven vehicles 16: Drones 20: Smart factories 22a, 22b: Machines 24: Mobile areas

FIG. 1 is a block diagram of the wireless charging system of the electric-powered vehicle of the present invention. FIG. 2 is a schematic diagram of an embodiment of the wireless charging system of the electric-powered vehicle of the present invention. FIG. 3 is a flow chart of the wireless charging method of the electric-powered vehicle of the present invention.

10: Wireless charging system for electric-powered vehicles

12: Power supply device

122: Power module

124: Scheduler Processor

14: Electric powered vehicles

16: Drones

Claims (14)

A wireless charging system for an electric-powered vehicle comprises: at least one drone for supplying power to a plurality of electric-powered vehicles; and a power supply device comprising a power module and a scheduling processor, wherein the scheduling processor monitors power information and a real-time position of the electric-powered vehicles, and determines whether to send a status confirmation message to the at least one drone according to the power information, wherein the power information indicates the current power of each of the electric-powered vehicles; wherein the at least one drone receives the power information and sends a status confirmation message to the at least one drone; After the status confirmation message is sent, a remaining power information is transmitted to the power supply device. The scheduling processor in the power supply device determines whether the remaining power of the at least one drone meets a charging requirement based on the remaining power information. If so, a weighted calculation is performed based on multiple charging parameters to generate a charging schedule, and a charging command is sent to the at least one drone that meets the charging requirement based on the charging schedule, so that it can charge the power-driven vehicles according to the order of the charging schedule. A wireless charging system for a power-driven vehicle as described in claim 1, wherein the at least one drone automatically charges at the power supply device when the remaining power is less than a preset value. A wireless charging system for a power-driven vehicle as described in claim 1, wherein when the distance between the at least one drone and the power supply device is less than a preset distance, the at least one drone automatically goes to the power supply device for charging. A wireless charging system for an electric-powered vehicle as described in claim 1, wherein the charging parameters include the distance between each of the electric-powered vehicles and the power supply device, the remaining power of each of the electric-powered vehicles, the remaining power of the at least one drone, and the floating electricity price. A wireless charging system for a power-driven vehicle as described in claim 1, wherein each of the power-driven vehicles continuously sends the power information to the power supply device, and when any of the power information is lower than a threshold value, the power supply device automatically sends the status confirmation message to the at least one drone. A wireless charging system for an electric-powered vehicle as described in claim 1, wherein the charging requirement is that the remaining power of at least one drone reaches more than 80%. A wireless charging system for an electric vehicle as described in claim 1, wherein the scheduling processor uses a feature-based expected SARSA (Feature-based Expected SARSA, Feature-based Expected State-Action-Reward-State-Action) algorithm based on reinforcement learning to perform charging optimization management of at least one drone. A wireless charging method for a power-driven vehicle is applied to a power supply device. The wireless charging method for a power-driven vehicle comprises the following steps: monitoring power information and a real-time position of a plurality of power-driven vehicles, and transmitting a status confirmation message to at least one drone according to the power information, wherein the power information indicates the current power of each of the power-driven vehicles; receiving the status confirmation message from the at least one drone in response to the status confirmation message; A remaining power information outputted by the confirmation message; Based on the remaining power information, determine whether the remaining power of the at least one drone meets a charging requirement, and perform weighted calculation based on multiple charging parameters to generate a charging schedule; and send a charging command to the at least one drone that meets the charging requirement according to the charging schedule, so that it can charge the power-driven vehicles according to the order of the charging schedule. A wireless charging method for a power-driven vehicle as described in claim 8, wherein the at least one drone automatically charges at the power supply device when the remaining power is less than a preset value. A wireless charging method for a power-driven vehicle as described in claim 8, wherein when the distance between the at least one drone and the power supply device is less than a preset distance, the at least one drone automatically goes to the power supply device for charging. A wireless charging method for a power-driven vehicle as described in claim 8, wherein the charging parameters include the distance between each of the power-driven vehicles and the power supply device, the remaining power of each of the power-driven vehicles, the remaining power of the at least one drone, and the floating electricity price. The wireless charging method of a power-driven vehicle as described in claim 8, wherein each of the power-driven vehicles continuously sends the power information to the power supply device, and when the power information is lower than a threshold value, the power supply device automatically sends the status confirmation message to the at least one drone. A wireless charging method for a power-driven vehicle as described in claim 8, wherein the charging requirement is that the remaining power of at least one drone reaches more than 80%. A wireless charging method for a power-driven vehicle as described in claim 8, wherein a scheduling processor in the power supply device uses a feature-based prediction SARSA algorithm with enhanced learning to execute the charging method for at least one drone.