Scheduling method, device and system of unmanned battery transport vehicle

Abstract

The invention relates to an unmanned battery transport vehicle scheduling method, device and system, and the method comprises the steps: taking the minimization of the total distribution distance and the total distribution demand of a battery replacement system as an optimization target, and determining the corresponding relation between each storage station and each battery replacement station; wherein the total distribution distance of the battery swap system is the sum of path distances between each pair of storage stations and battery swap stations, the total distribution demand of the battery swap system is the sum of distribution demands between each pair of storage stations and battery swap stations, and the distribution demand between any pair of storage stations i and battery swap stations j is the product of a first numerical value and a second numerical value, the first value is an electricity ratio of the number of the current to-be-distributed batteries of the battery swap station j to the total battery capacity, and the second value is a path distance normalization value between the storage station i and the battery swap station j. The problem that the battery transport vehicles cannot be effectively and reasonably dispatched by the battery transport vehicle dispatching method of the current battery replacement system is solved.

Description

Technical Field

[0001] The present application relates to the field of electric vehicles, and in particular to a scheduling method, device and system for an unmanned battery transport vehicle. Background Art

[0002] The battery swap service can replace exhausted batteries in a short period of time and charge them during the off-peak period. Usually, battery storage stations, battery recycling stations and battery swap stations form a battery swap system, and a transportation network is formed between the stations in the battery swap system. The operation process of the above-mentioned battery swap system is as follows: the vehicle exchanges batteries at the battery swap station, batteries with poor health need to be transported to the battery recycling station, batteries with better health are transported to the battery storage station and charged during the off-peak period, and the battery storage station transports fully charged batteries to the battery swap station. Among them, the efficient distribution of batteries in the transportation network is more critical.

[0003] At present, the transportation and distribution of batteries are basically achieved by drivers, which requires a lot of manpower. More importantly, the current battery transport vehicle scheduling method of the battery swap system fails to effectively and reasonably schedule the battery transport vehicles, that is, it fails to reasonably plan the transportation relationship between the various stations in the battery swap system, which may cause the battery storage station to provide fully charged batteries to the distant battery swap station or the waste batteries in the battery swap station to be transported to the distant battery recycling station for recycling, resulting in the battery transport vehicles in the battery swap system traveling a long distance, thereby reducing the overall transportation efficiency of the battery.

[0004] There is currently no effective solution to the problem that the current battery transport vehicle scheduling method of the battery swap system fails to effectively and reasonably schedule the battery transport vehicles. Summary of the invention

[0005] The present invention provides a method, device and system for scheduling an unmanned battery transport vehicle to solve the problem that the current battery transport vehicle scheduling method of the battery replacement system fails to effectively and reasonably schedule the battery transport vehicles.

[0006] In a first aspect, the present invention provides a scheduling method for an unmanned battery transport vehicle, which is applied to a battery swap system, wherein the battery swap system includes a plurality of battery swap stations and a plurality of storage stations, and the scheduling method includes: Taking minimizing the total distribution distance and total distribution demand of the battery swap system as the optimization goal, determine the corresponding relationship between each storage station and each battery swap station; among which, the total distribution distance of the battery swap system is the sum of the path distances between each pair of storage stations and battery swap stations, and the total distribution demand of the battery swap system is the sum of the distribution demands between each pair of storage stations and battery swap stations. i and battery swap stations j The distribution demand between is the product of the first value and the second value, where the first value is the exchange station jThe current ratio of the number of batteries to be delivered to the total battery capacity, and the second value is the storage station i With battery swap station j The normalized value of the path distance between them; Dispatch unmanned battery transport vehicles to achieve battery distribution between corresponding battery swap stations and storage stations.

[0007] In a second aspect, the present invention provides a dispatching device for an unmanned battery transport vehicle, which is applied to a battery swapping system, wherein the battery swapping system includes a plurality of battery swapping stations and a plurality of storage stations, and the dispatching device includes: The optimization module is used to determine the corresponding relationship between each storage station and each battery swap station with the optimization goal of minimizing the total distribution distance and total distribution demand of the battery swap system; wherein the total distribution distance of the battery swap system is the sum of the path distances between each pair of storage stations and the battery swap station, and the total distribution demand of the battery swap system is the sum of the distribution demands between each pair of storage stations and the battery swap station. i and battery swap stations j The distribution demand between is the product of the first value and the second value, where the first value is the exchange station j The ratio of the number of batteries to be delivered to the total battery capacity, and the second value is the storage station i With battery swap station j The normalized value of the path distance between them; The dispatching module is used to dispatch unmanned battery transport vehicles to realize battery distribution between corresponding battery swap stations and storage stations.

[0008] In a third aspect, the present invention provides a scheduling system for an unmanned battery transport vehicle, comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the scheduling method for the unmanned battery transport vehicle described in the first aspect.

[0009] Compared with related technologies, the dispatching method of unmanned battery transport vehicles provided by the present invention takes the total delivery distance and total delivery demand of the battery swap system as optimization targets, and can obtain a set of corresponding relationships between battery swap stations and storage stations, and dispatch unmanned battery transport vehicles to achieve battery delivery between the corresponding battery swap stations and storage stations. This can minimize the total delivery distance and total delivery demand of the battery swap system, achieve a more reasonable dispatching of unmanned battery transport vehicles, and solve the problem that the current dispatching method of battery transport vehicles for the battery swap system fails to effectively and reasonably dispatch battery transport vehicles.

[0010] Details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1is a flow chart of a method for dispatching an unmanned battery transport vehicle provided in an embodiment of the present invention; Figure 2 Comparison chart of simulation results of PID-Stanley control compared with other various controls. DETAILED DESCRIPTION

[0012] In order to more clearly understand the purpose, technical solutions and advantages of the present application, the present application is described and illustrated below in conjunction with the accompanying drawings and embodiments.

[0013] Unless otherwise defined, the technical terms or scientific terms involved in this application shall have the general meaning understood by people with general skills in the technical field to which this application belongs. The words "one", "a", "the", "these" and the like in this application do not indicate a quantitative limitation, and they may be singular or plural. The terms "include", "comprise", "have" and any variants thereof involved in this application are intended to cover non-exclusive inclusions; for example, a process, method and system, product or device comprising a series of steps or modules (units) is not limited to the listed steps or modules (units), but may include unlisted steps or modules (units), or may include other steps or modules (units) inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like involved in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The "multiple" involved in this application refers to two or more. "And / or" describes the association relationship of associated objects, indicating that there may be three relationships, for example, "A and / or B" may mean: A exists alone, A and B exist at the same time, and B exists alone. Generally, the character " / " indicates that the objects associated with each other are in an "or" relationship. The terms "first", "second", "third", etc. in this application are only used to distinguish similar objects and do not represent a specific ordering of the objects.

[0014] In an embodiment of the present invention, a dispatching method for an unmanned battery transport vehicle is provided, which is applied to a battery swapping system. The battery swapping system includes at least a plurality of battery swapping stations and a plurality of storage stations. The storage stations are used to store and charge batteries that need to be charged in the battery swapping stations. The battery swapping system may also include a plurality of recycling stations. The recycling stations are used to recycle batteries that need to be recycled in the battery swapping stations. In an embodiment that does not include a recycling station, the battery recycling problem is not considered, and it is assumed that the replaced batteries in the battery swapping station are charged and reused.

[0015] Figure 1 1 is a flow chart of a method for dispatching an unmanned battery transport vehicle provided in an embodiment of the present invention. Figure 1 , the process includes step S110 and step S120.

[0016] Step S110, with minimizing the total delivery distance and total delivery demand of the battery swap system as the optimization goal, determine the corresponding relationship between each storage station and each battery swap station; wherein the total delivery distance of the battery swap system is the sum of the path distances between each pair of storage stations and the battery swap station, and the total delivery demand of the battery swap system is the sum of the delivery demands between each pair of storage stations and the battery swap station. i and battery swap stations j The distribution demand between is the product of the first value and the second value, where the first value is the exchange station j The current ratio of the number of batteries to be delivered to the total battery capacity, and the second value is the storage station i With battery swap station j The path distance normalization value between . i and j They represent the serial numbers of the storage station and the battery swap station respectively.

[0017] Step S120, dispatching unmanned battery transport vehicles to achieve battery distribution between corresponding battery swap stations and storage stations.

[0018] The above is mainly used to determine the corresponding relationship between each storage station and each battery swap station. The corresponding storage station and battery swap station constitute a pair of storage station and battery swap station. For any pair of storage station and battery swap station, an unmanned vehicle needs to be dispatched to transport the batteries in the storage station to the corresponding battery swap station or to transport the batteries in the battery swap station to the corresponding storage station.

[0019] In an embodiment where the battery swap system does not include a recycling station, the battery swap station j The current number of batteries to be delivered is the number of batteries to be replaced in the current delivery cycle. The batteries replaced from the electric vehicle need to be transported to the storage station for storage and charging, and the same number of batteries are obtained from the storage station for replenishment. In this embodiment, the number of batteries that need to be charged and replenished at a single battery swap station is the same, so the same unmanned battery transport vehicle is used to transport the batteries from the battery swap station to the storage station and from the storage station to the battery swap station.

[0020] In an embodiment where the battery swap system includes a recycling station, any battery swap station j The current number of batteries to be delivered is the first number or the second number; the first number is the number of batteries that can be delivered to the battery swap station within the current delivery cycle. j The second number is the sum of the first number and the number of battery swap stations in the current delivery cycle. jThe sum of the number of batteries that need to be recycled. Part of the batteries replaced from the electric car need to be transported to the recycling station for recycling, and the other part needs to be transported to the storage station for storage and charging, and the number of batteries that need to be charged in the battery swap station is the sum of the two parts. In this embodiment, the number of batteries that need to be charged, recycled, and replenished at a single battery swap station is different, so different unmanned battery transport vehicles are used to dispatch the batteries from the battery swap station to the storage station, the batteries from the storage station to the battery swap station, and the batteries from the battery swap station to the recycling station, that is, the dispatching method of the unmanned battery transport vehicle provided by the present invention is used to implement them respectively.

[0021] Among them, any battery swap station j The number of batteries that need to be charged and recycled is determined according to the health of each battery. The health of the battery is the weighted result of the battery's capacity health state, internal resistance health state, and charge and discharge efficiency state.

[0022] The following is an example to illustrate the determination and classification of battery health.

[0023] When the battery swap vehicle arrives at the battery swap station, the battery health detection system of the battery swap station conducts a comprehensive test on the battery currently on the vehicle, including the battery number, power level, and battery health. SOH And the battery detection time. According to the battery health SOH The size of the battery will SOH Divide into 4 intervals for subsequent processing: When the battery health is good, When the battery health is good, When the battery health is medium, When the battery health is poor, it is specified that when the battery health is excellent, good, or medium, the battery will be stored and charged during the grid off-peak period for continued use. When the battery health is poor, the battery will be recycled and uploaded to the recycling system.

[0024] In order to comprehensively evaluate the health of the battery, a weighted scoring model is used here to combine multiple battery performance parameters to judge its health. The goal of the weighted scoring model is to obtain a comprehensive health score by weighting and summarizing multiple health indicators (such as internal resistance, capacity, charge and discharge efficiency, etc.). The specific formula is as follows:

[0025] in, H Indicates the battery health level (between 0 and 100%). , and are weight coefficients, representing the influence of capacity, internal resistance, and charge and discharge efficiency on the health of the battery. . H cap is the capacity health status, H res The internal resistance is healthy. H eff It is a healthy status based on charge and discharge efficiency. C current is the current capacity of the battery, C design is the design capacity of the battery. R min is the minimum internal resistance of the battery in a normal healthy state. R current is the actual internal resistance of the battery at present. and They represent the efficiency of charge and discharge respectively. E charge , E input , E discharge and E output They represent the energy received by the battery during charging and the energy provided to the battery externally, as well as the energy output by the battery during discharging and the energy extracted from the battery by the external load.

[0026] Through the above calculation formula, the health status of each battery in the battery swap station can be calculated and divided, so as to determine the batteries that need to be recycled and the batteries that need to be charged.

[0027] Among them, if the battery health status is poor, the battery will be transported to the battery recycling station by an unmanned battery transport vehicle, and the battery information will be uploaded to the background management system to register the battery recycling record to facilitate the subsequent processing of such batteries.

[0028] After the unmanned battery transport vehicle arrives at the battery recycling station, the recycling station will further inspect and classify the batteries, evaluate their recyclability, and record relevant information for each battery, including the battery model, health status, transportation time, transportation method and other data. All this information will be uploaded to the backend management system in real time to form a complete battery recycling record. The system will automatically generate a unique identification code for each battery and associate the corresponding recycling data for subsequent tracking and analysis. These recycling records not only help ensure the compliance of each battery, but also provide important data support for future battery management and help optimize the battery recycling process. By analyzing the battery recycling data, the system can identify which battery models or batteries with poor health are more common, so as to provide manufacturers with improvement suggestions to further improve the quality and safety of the batteries. In addition, the system will share relevant data with environmental protection departments or relevant regulatory agencies to ensure compliance with relevant laws and policies on environmental protection. In the subsequent processing stage, the batteries will be disassembled, repaired or destroyed according to the battery condition and recycling standards to maximize the recycling of resources. This process not only improves the recycling rate of batteries, but also effectively reduces the potential harm of discarded batteries to the environment and promotes the development of a green circular economy.

[0029] Regardless of whether the battery swap system includes a recycling station, the scheduling method includes scheduling unmanned battery transport vehicles to realize the battery distribution between the corresponding battery swap stations and storage stations. Step S110 is used to determine the correspondence between each storage station and each battery swap station. It constructs vehicle scheduling as an optimization problem, and determines the correspondence between the battery swap station and the storage station by minimizing the total distribution distance and total distribution demand of the battery swap system. Preferably, the correspondence between each storage station and each battery swap station is one-to-one. For example, a storage station just provides batteries for one battery swap station, and a battery swap station just provides batteries from one storage station. Using a one-to-one correspondence to establish a battery distribution relationship can simplify the scheduling of unmanned vehicles.

[0030] The step S110 is specifically described below using an example.

[0031] First, the relevant information of the battery swap system includes: the number of stations in the battery swap system n (battery swap stations and storage stations) and coordinate information; demand information matrix of each battery swap station compared to different storage stations R ; Distance information matrix between various sites D ; Demand weights and distance weights customized by managers.

[0032] Secondly, scheduling optimization is achieved through an objective function. The objective function consists of two parts: distance and demand. The connection relationship between the battery swap station and the battery storage station is represented by a binary variable. The mixed integer programming (MIP) algorithm is used to solve this optimal scheduling problem.

[0033] The objective function takes into account two aspects: minimizing the delivery distance, that is, the shortest delivery path; and minimizing the delivery demand, that is, meeting the needs of each battery storage station. x ij Indicates that from the battery storage station i To the battery swap station j Whether to select this path. .definition r ij Demand Information Matrix R The elements in represent the battery storage station i To the battery swap station j The demand value of .definition d ij is the distance information matrix D The elements in represent the battery storage station i To the battery swap station j The demand weight and distance weight range between [0, 1]. The expression of the objective function is defined as:

[0034] in, n represents the number of storage stations and battery swap stations, i and j Respectively represent the serial numbers of the storage station and the battery swap station, w d and w r denote the distance weight and demand weight respectively, and Respectively represent the maximum distance and minimum distance between the battery swap station and the storage station, Indicates battery replacement station j The number of batteries that need to be replenished, Indicates battery replacement station j The maximum number of batteries that can be accommodated.

[0035] It is understandable that the ratio of the number of batteries that need to be replenished to the maximum number of batteries that can be accommodated at the battery swap station determines the urgency of the need for the battery swap station. The higher the ratio, the fewer remaining batteries can be used at the battery swap station, and the faster the replenishment should be delivered, so the battery swap station should be matched with a storage station with a shorter distance. Storage station i and battery swap stations j Delivery demand between r ij is the first value (characterizing the battery swap station j urgency) and a second value (for storage stations i and battery swap stationsj By establishing different corresponding relationships between storage stations and battery swap stations, the distribution demand between different storage stations and different battery swap stations is minimized, and the more urgent battery swap stations can establish a corresponding relationship with the closer storage stations.

[0036] like x ij = 0, it means battery storage station i Not connected to a battery swap station j ;like x ij = 1, it means battery storage station i Connect to a battery swap station j . That is, battery storage station i Connect to a battery swap station j Once a one-to-one relationship is established, it is necessary to dispatch an unmanned battery transport vehicle to transport batteries from the battery storage station. i Transport to the battery swap station j Finally, through the objective function, a series of distribution routes can be determined, and the starting point and end point of each distribution route are the corresponding battery storage station and battery swap station.

[0037] It should be noted that, for a pair of storage stations and battery swap stations, there may be multiple connection paths between the two. The present invention selects the shortest connection path as the distribution path between the two by default.

[0038] To ensure that each battery storage station has only one battery swap station and each battery swap station must also be connected to a battery storage station, set the following constraints:

[0039] For each battery swap station, it must be routed by exactly one battery storage station.

[0040]

[0041] For the above objective function, the goal is to connect multiple battery storage stations and multiple battery swap stations by optimizing path selection to minimize the weighted objective function. For this MIP problem, the intlinprog function in MATLAB is used to solve it. Among them, the intlinprog function is a solver in MATLAB for solving integer linear programming, which can handle linear problems containing integer constraints.

[0042] In order to adapt the solution function, the objective function is specified . The objective function fExpanded into a one-dimensional column vector, representing the weighted value of the path selection between each pair of battery storage stations and battery swap stations. This step converts the two-dimensional matrix (containing the distance and demand value of each pair of stations) into a one-dimensional vector for easy processing by the solver.

[0043] For this problem, the complete form of the solution process is:

[0044] Among them, intcon is the index of the integer variable, representing all x ij variable. A eq and b eq is the matrix and vector of path connectivity constraints, ensuring that each battery storage station and each battery swap station are properly connected. lb and ub represent the lower and upper bounds of the decision variables respectively.

[0045] After the solution is completed, the solution vector x Generate the final path configuration. These paths are the optimal solutions that meet the constraints, that is, minimize the objective function. x ij = 1, that is, the battery swap station corresponding to each battery storage station can be found.

[0046] The above example is a specific description of the scenario of a storage station delivering batteries to a battery swap station. The scenarios of a battery swap station delivering batteries to a storage station and a battery swap station delivering batteries to a recycling station are similar. For example, it is necessary to change Adapt to changes. In the scenario where storage stations deliver batteries to battery swap stations, Indicates battery replacement station j The number of batteries that need to be replenished; in the scenario where the battery is delivered from the battery swap station to the storage station, Indicates battery replacement station j The number of batteries that need to be charged; in the scenario where the battery swap station delivers batteries to the recycling station, Indicates battery replacement station j The number of batteries that need to be recycled is calculated, and the distance between the battery swap station and the storage station is replaced by the distance between the battery swap station and the recycling station.

[0047] In summary, in step S110: taking the total delivery distance and total delivery demand of the battery swap system as the optimization target, a set of corresponding relationships between battery swap stations and storage stations can be obtained, and unmanned battery transport vehicles can be dispatched to achieve battery delivery between the corresponding battery swap stations and storage stations. This can minimize the total delivery distance and total delivery demand of the battery swap system, achieve a more reasonable dispatch of unmanned battery transport vehicles, and solve the problem that the current dispatch method of battery transport vehicles for the battery swap system fails to effectively dispatch battery transport vehicles.

[0048] Step S120 specifically includes: determining the delivery route of the unmanned battery transport vehicle and performing trajectory tracking control on the unmanned battery transport vehicle; wherein the trajectory tracking control step includes step S121, step S122 and step S123.

[0049] Step S121, determining a first control variable of the unmanned battery transport vehicle through the Stanley control algorithm and providing a trajectory error of the unmanned battery transport vehicle to the PID control algorithm.

[0050] Step S122, determining a second control amount of the unmanned battery transport vehicle through a PID control algorithm based on the trajectory error amount of the unmanned battery transport vehicle provided by the Stanley control algorithm.

[0051] Step S123, controlling the unmanned battery transport vehicle based on the weighted result of the first control amount and the second control amount.

[0052] In some embodiments, the track tracking is to let the unmanned battery transport vehicle track the target lane line in the delivery path, and the target lane line can be the lane line closest to the vehicle. That is, the track tracking of the unmanned battery transport vehicle includes: obtaining a lane image through a vehicle-mounted camera; identifying the lane line in the lane image and taking the lane line closest to the vehicle-mounted camera as the target lane line; controlling the unmanned battery transport vehicle to track the target lane line in the delivery path, and the track error of the unmanned battery transport vehicle is the lateral error and heading error of the unmanned battery transport vehicle compared to the target lane line.

[0053] In order to ensure the recognition accuracy of lane images, a series of processing is required for lane images. Identifying lane lines in lane images and taking the lane line closest to the vehicle camera as the target lane line specifically includes: performing grayscale transformation, Gaussian filtering, Canny edge detection and Reichhoff transform processing on the lane image in sequence; extracting at least one straight line from the processed lane image, calculating the Euclidean distance between each straight line and the vehicle camera, and taking the straight line with the smallest Euclidean distance to the vehicle camera as the target lane line.

[0054] The step S120 is specifically described below using an example.

[0055] In this example, step S120 mainly includes two parts.

[0056] In the first part, the unmanned battery transport vehicle detects the lane line closest to the camera through the on-board camera. Specifically, it involves grayscale transformation, Gaussian filtering, Canny edge detection, Hough transformation and Euclidean distance calculation.

[0057] (1) The lane image captured by the vehicle camera is converted into a single-channel intensity value by grayscale transformation. The conversion formula is as follows:

[0058] in, I R , I G and I B Represented as the pixel values ​​of the image in the red, green, and blue channels respectively. I gray is the pixel value of the converted grayscale image.

[0059] (2) Next, the grayscale processed image is subjected to Gaussian filtering to reduce image noise and make the edges smoother. Gaussian filtering uses a Gaussian function as a weight to perform a convolution operation on the image, thereby blurring the image. The mathematical formula of the Gaussian filter is as follows:

[0060] in,( x , y ) is the offset of the pixel position relative to the center of the filter. is the standard deviation of the Gaussian distribution, which controls the strength of the filter. A larger standard deviation will make the image blurred more smoothly.

[0061] In Gaussian filtering, each pixel in the image captured by the on-board camera is replaced by the weighted average of the surrounding pixels, where the weights of the surrounding pixels are determined by the Gaussian function. This can effectively reduce noise and maintain the main structural information in the image.

[0062] (3) Perform Canny edge detection on the image after Gaussian filtering. The basic process of this step includes the following steps: Gaussian smoothing, gradient calculation, dual threshold detection and edge tracking.

[0063] First, the image is Gaussian smoothed to reduce noise, and then the gradient values ​​of the image in the horizontal and vertical directions are calculated, and the Sobel operator is used to calculate the gradient. Then the calculated gradient intensity image is subjected to non-maximum suppression. The purpose of non-maximum suppression is to accurately retain edge pixels in the gradient direction, remove weak pixels in the edge direction, and avoid edge breaks. Finally, two thresholds are used to classify edge pixels: if the gradient value of a pixel is greater than the high threshold, it is considered a strong edge; if the gradient value of a pixel is between the low threshold and the high threshold, it is considered a weak edge; if the gradient value of a pixel is less than the low threshold, it is considered not an edge. The algorithm then connects the weak edge with the strong edge. If the weak edge is adjacent to the strong edge, the weak edge is considered to be part of the edge and is retained.

[0064] The Gaussian smoothing in the Canny edge detection step is similar to the Gaussian filtering method mentioned above. The image is convolved with a Gaussian function to smooth the image.

[0065] The second step of the Canny algorithm is to use the Sobel operator to calculate the gradient of the image in the horizontal and vertical directions. Specifically, two convolution kernels are used for calculation. The Sobel operators in the horizontal and vertical directions are defined as:

[0066] By convolving these convolution kernels with the image, the gradient of the image in the horizontal and vertical directions can be obtained. G x and G y They represent the image in x and y Directional rate of change of brightness: , . where represents the gray value of the image. Next, calculate the intensity of the gradient G and direction .

[0067] The gradient strength is the square root of the sum of the squares of the gradients in the horizontal and vertical directions, which represents the strength of the image transformation. The gradient direction indicates the direction of the edge, which can be calculated by the horizontal and vertical gradients: This direction is the main direction of the edge in the image, and determines whether the pixel is on the edge.

[0068] The third step of the Canny algorithm is to perform non-maximum suppression. For each pixel, according to its gradient direction, determine the two neighboring pixels in its neighborhood along the gradient direction, and retain the pixel with the local maximum value in the gradient direction. x , y),if G ( x , y ) is not the maximum value of the two neighboring pixels in the gradient direction, it is suppressed to zero, which can be expressed as: .

[0069] in, G left and G right Refers to pixels ( x , y ) is the intensity of the left and right neighboring pixels in the gradient direction, G top and G bottom Refers to pixels ( x , y ) is the intensity of the upper and lower neighboring pixels in the gradient direction. The above formula shows that when G ( x , y ) is greater than the intensity of all its neighboring pixels, it is retained; if it is less than the intensity of one of its neighboring pixels, it is set to 0.

[0070] The fourth step of the Canny algorithm is to perform dual threshold detection on edge pixels and perform edge tracking. In dual threshold classification, two thresholds are used to determine the category of edge pixels. The high threshold and the low threshold are defined as T H and T L .Regulation:

[0071] After double threshold detection, for each weak edge pixel, the algorithm checks whether its neighborhood contains a strong edge pixel. If there is a strong edge pixel, the weak edge pixel is retained; if not, the weak edge pixel is suppressed.

[0072] (4) Perform Hough transform on the processed image obtained in the above steps, map the straight line from the image space to the parameter space, and use polar coordinates to represent the straight line. The equation of the straight line in polar coordinate form is:

[0073] in, is the shortest distance from the straight line to the origin, is the angle between the straight line and the horizontal direction. By dividing each edge point in the image ( x , y ) is converted to The Hough transform generates an accumulated graph (Hough space) for each point in the parameter space. Corresponding to a peak. By accumulating these peaks, straight lines in the image can be detected.

[0074] (5) After extracting all the straight lines detected by the camera through the above steps, use the Euclidean distance to calculate the distance between each straight line and the camera, and select the straight line with the smallest distance as the target lane line detected by the camera. Specifically, define the position of the camera as , for each straight line obtained by Hough transform and , we can calculate the Euclidean distance between this line and the camera. d The calculation formula is:

[0075] By calculating the Euclidean distance between each straight line and the camera, the straight line closest to the vehicle-mounted camera is selected as the target lane line.

[0076] In the second part, the trajectory tracking of the unmanned battery transport vehicle is realized based on the PID control algorithm and the Stanley control algorithm.

[0077] First, the PID control algorithm is responsible for adjusting the system output according to the current error size (the lateral error and heading error of the unmanned battery transport vehicle compared to the target lane line provided by the Stanley control algorithm). The proportional controller directly adjusts the output signal according to the error size, so that the output signal is proportional to the error.

[0078] Secondly, Stanley lateral control controls the vehicle steering wheel angle based on the relative geometric relationship between the vehicle position and the target lane line on the delivery path. Specifically, the Stanley lateral control algorithm uses the vehicle lateral tracking error and heading tracking error as feedback signals, calculates the front wheel steering angle through a nonlinear proportional function, reduces the lateral tracking error and improves the vehicle's lateral tracking performance.

[0079] Not only the current dynamic state and error of the vehicle is controlled by the PID control algorithm, but also the lateral error and heading error of the vehicle are controlled by the Stanley control algorithm, thus providing a PID-Stanley control algorithm.

[0080] According to the PID-Stanley control algorithm, the unmanned transport vehicle tracks the trajectory of the white lane line, enabling the vehicle to be transported between battery storage stations, battery swap stations, and battery recycling stations, thereby improving transportation efficiency.

[0081] Specifically, after considering proportional regulation, integral regulation, and differential regulation at the same time, the PID control formula is expressed as follows:

[0082] in, u ( t ) is the proportional controller output, the control signal that needs to be applied to the system; e ( t ) is the error at the current moment, defined as the difference between the expected value and the actual value; K p is the proportional gain coefficient, which controls the multiple between the output signal and the error; K i is the integral gain coefficient; K d is the differential gain coefficient, C is a variable matrix used to dynamically adjust the error weights according to the relative importance of lateral error and heading error. v ( t ) is a vehicle t The linear speed at any given moment, It is a vehicle t The curvature of the path at that moment. k 1 and k 2 is the dynamic coefficient related to the current state. Both coefficients are proportional to the linear velocity and curvature. They can be randomly generated according to the proportional relationship in the simulation, such as , . d e is the lateral error, is the heading error.

[0083] In the trajectory tracking task, PID control is used to calculate the steering angle or speed of the vehicle so that the vehicle can minimize the deviation from the desired path. Specifically, the input errors of PID control are usually lateral error and heading error. Therefore, in order to make these two errors controllable, the Stanley control algorithm is introduced to reduce the lateral error and heading error.

[0084] For Stanley control, the steering angle is calculated based on the current vehicle position and heading angle. To reduce the lateral error and heading error. The control formula of Stanley algorithm is:

[0085]

[0086] in, k e is an empirically adjusted gain factor used to control the sensitivity of the steering angle. x , y ) is the current coordinate of the vehicle. is the closest point of the target path (target lane line). is the tangent direction of the nearest point, that is, the desired heading. is the current heading angle of the vehicle.

[0087] The lateral error and heading error calculated in the above Stanley control are provided to the PID control.

[0088] For the PID-Stanley control method, the output steering angles of PID control and Stanley control are calculated respectively. and The final control signal is a weighted combination of the two control methods:

[0089] in, It is a tuning parameter used to control the weight of PID and Stanley control in the final control signal.

[0090] In summary, in step S120: first, the lane image captured by the vehicle-mounted camera is processed in sequence by grayscale transformation, Gaussian filtering, Canny edge detection and Reichhoff transform, at least one straight line is extracted from the processed lane image, the Euclidean distance between each straight line and the vehicle-mounted camera is calculated, and the straight line with the smallest Euclidean distance to the vehicle-mounted camera is used as the target lane line. Then, the target lane line is used as a reference for trajectory tracking, and the PID-Stanley control method is used to make the unmanned battery transport vehicle follow the target lane line. It should be noted that the PID-Stanley control method is a new trajectory tracking control method provided by the present invention, and its key point is that Stanley control provides lateral error and heading error for PID control, and Stanley control is used as the final control quantity of the unmanned battery transport vehicle after the weighted combination of the output of PID control, so as to achieve high-precision vehicle tracking control.

[0091] As shown below, some simulation experimental data show that the above PID-Stanley control has certain advantages in control accuracy compared with other controls.

[0092] Figure 2 The following is a comparison chart of the simulation results of PID-Stanley control compared with other various controls. The comparison results are shown in Table 1.

[0093] Table 1 Error comparison of various controls

[0094] According to Table 1, the improved PID-Stanley control adopted by the present invention is superior to the traditional fuzzy logic control, the traditional Kalman filter control and the traditional PID control in all aspects of error. As for the traditional Stanley control with higher accuracy, except that the average error is slightly larger, the minimum error, maximum error, standard deviation and error range are all better than the traditional Stanley control, which shows that the PID-Stanley control significantly improves the control stability while maintaining high-precision control.

[0095] In an embodiment of the present invention, a dispatching device for an unmanned battery transport vehicle is also provided, which is applied to a battery exchange system, and the battery exchange system includes multiple battery exchange stations and multiple storage stations. The device is used to implement the above-mentioned embodiments and preferred embodiments, and those that have been explained will not be repeated. The terms "module", "unit", "sub-unit", etc. used below can implement a combination of software and / or hardware for predetermined functions. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceived.

[0096] The dispatching device of the unmanned battery transport vehicle includes: The optimization module is used to determine the corresponding relationship between each storage station and each battery swap station with the optimization goal of minimizing the total distribution distance and total distribution demand of the battery swap system; wherein the total distribution distance of the battery swap system is the sum of the path distances between each pair of storage stations and the battery swap station, and the total distribution demand of the battery swap system is the sum of the distribution demands between each pair of storage stations and the battery swap station. i and battery swap stations j The distribution demand between is the product of the first value and the second value, where the first value is the exchange station j The ratio of the number of batteries to be delivered to the total battery capacity, and the second value is the storage station i With battery swap station j The normalized value of the path distance between them; The dispatching module is used to dispatch unmanned battery transport vehicles to realize battery distribution between corresponding battery swap stations and storage stations.

[0097] It should be noted that the above modules can be functional modules or program modules, and can be implemented by software or hardware. For modules implemented by hardware, the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.

[0098] In an embodiment of the present invention, a scheduling system for an unmanned battery transport vehicle is also provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to implement the scheduling method for the unmanned battery transport vehicle provided by the present invention.

[0099] It should be understood that the specific embodiments described herein are only used to explain the application, rather than to limit it. Based on the embodiments provided in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the protection scope of this application.

[0100] Obviously, the drawings are only some examples or embodiments of the present application. For ordinary technicians in the field, the present application can also be applied to other similar situations based on these drawings without creative work. In addition, it is understandable that although the work done in this development process may be complicated and lengthy, for ordinary technicians in the field, certain changes in design, manufacturing or production based on the technical content disclosed in this application are only conventional technical means and should not be regarded as insufficient content disclosed in this application.

Claims

1. A dispatching method for an unmanned battery transport vehicle, applied to a battery swapping system, wherein the battery swapping system includes a plurality of battery swapping stations and a plurality of storage stations, characterized in that: Scheduling methods include: Taking minimizing the total distribution distance and total distribution demand of the battery swap system as the optimization goal, determine the corresponding relationship between each storage station and each battery swap station; the total distribution distance of the battery swap system is the sum of the path distances between each pair of storage stations and the battery swap station, the total distribution demand of the battery swap system is the sum of the distribution demands between each pair of storage stations and the battery swap station, and the storage station i and battery swap stations j The distribution demand between is the product of the first value and the second value, where the first value is the exchange station j The current ratio of the number of batteries to be delivered to the total battery capacity, and the second value is the storage station i With battery swap station j The normalized value of the path distance between them; Dispatch unmanned battery transport vehicles to achieve battery distribution between corresponding battery swap stations and storage stations.

2. The method for dispatching an unmanned battery transport vehicle according to claim 1, characterized in that: The correspondence between each storage station and each battery swap station is one-to-one.

3. The dispatching method of unmanned battery transport vehicle according to claim 1, characterized in that: Dispatching unmanned battery transport vehicles to achieve battery distribution between corresponding battery swap stations and storage stations includes: Determine the delivery route of the unmanned battery transport vehicle and perform trajectory tracking control on the unmanned battery transport vehicle; The steps of trajectory tracking control include: Determine a first control variable of the unmanned battery transport vehicle by using a Stanley control algorithm and provide a trajectory error of the unmanned battery transport vehicle to a PID control algorithm; Determine a second control amount of the unmanned battery transport vehicle by using a PID control algorithm based on a trajectory error amount of the unmanned battery transport vehicle provided by a Stanley control algorithm; The unmanned battery transport vehicle is controlled by weighting the first control amount and the second control amount.

4. The method for dispatching an unmanned battery transport vehicle according to claim 3, characterized in that: Tracking of unmanned battery transport vehicles includes: The unmanned battery transport vehicle is controlled to track the target lane line in the delivery path. The trajectory error of the unmanned battery transport vehicle is the lateral error and heading error of the unmanned battery transport vehicle compared to the target lane line.

5. The method for dispatching an unmanned battery transport vehicle according to claim 4, characterized in that: Tracking of unmanned battery transport vehicles also includes: Acquire lane images through a vehicle-mounted camera; Identify the lane lines in the lane image and use the lane line closest to the vehicle camera as the target lane line.

6. The method for dispatching an unmanned battery transport vehicle according to claim 5, characterized in that: Identifying lane lines in a lane image and using the lane line closest to the vehicle camera as the target lane line includes: The lane image is processed in sequence by grayscale transformation, Gaussian filtering, Canny edge detection and Reichhof transform; At least one straight line is extracted from the processed lane image, the Euclidean distance between each straight line and the vehicle-mounted camera is calculated, and the straight line with the smallest Euclidean distance to the vehicle-mounted camera is taken as the target lane line.

7. According to the dispatching method of unmanned battery transport vehicles of claim 1, the battery replacement system further comprises a recycling station, characterized in that: Any battery swap station j The current number of batteries to be delivered is the first number or the second number; The first quantity is the battery swap station within the current delivery cycle j The second number is the sum of the first number and the number of battery swap stations in the current delivery cycle. j The total number of batteries that need to be recycled.

8. The method for dispatching an unmanned battery transport vehicle according to claim 7, characterized in that: Any battery swap station j The number of batteries that need to be charged and recycled is determined according to the health of each battery. The health of the battery is the weighted result of the battery's capacity health state, internal resistance health state, and charge and discharge efficiency health state.

9. A dispatching device for an unmanned battery transport vehicle, applied to a battery swapping system, the battery swapping system comprising a plurality of battery swapping stations and a plurality of storage stations, characterized in that: The dispatching device includes: The optimization module is used to determine the corresponding relationship between each storage station and each battery swap station with the optimization goal of minimizing the total distribution distance and total distribution demand of the battery swap system; wherein the total distribution distance of the battery swap system is the sum of the path distances between each pair of storage stations and the battery swap station, and the total distribution demand of the battery swap system is the sum of the distribution demands between each pair of storage stations and the battery swap station. i and battery swap stations j The distribution demand between is the product of the first value and the second value, where the first value is the exchange station j The ratio of the number of batteries to be delivered to the total battery capacity, and the second value is the storage station i With battery swap station j The normalized value of the path distance between them; The dispatching module is used to dispatch unmanned battery transport vehicles to realize battery distribution between corresponding battery swap stations and storage stations.

10. A dispatching system for an unmanned battery transport vehicle, comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to run the computer program to execute the method for dispatching an unmanned battery transport vehicle according to any one of claims 1 to 8.

Images