Path planning method, system and device for new energy vehicle, medium and vehicle

Abstract

The invention discloses a path planning method, system and device for a new energy automobile, a medium and a vehicle, and the method comprises the steps: obtaining current influence factors, and predicting the operation state of a charging pile in a future preset time period according to the current influence factors and a trained first prediction model, the current influence factors including a current time factor and a current environment factor; acquiring a starting point and an ending point, acquiring a recommended route according to the starting point and the ending point, and identifying the operation state of the charging pile on the recommended route according to the predicted operation state of the charging pile in the future preset time period and the arrival time; acquiring a target charging pile, and acquiring a driving route according to the starting point and the target charging pile; the target charging pile is determined according to the operation state of the identified charging pile. The embodiment of the invention can reduce charging waiting time and travel time, and can be widely applied to the technical field of path planning.

Description

Technical Field

[0001] The present invention relates to the technical field of path planning, and in particular to a path planning method, system, device, medium and vehicle for a new energy vehicle. Background Art

[0002] In recent years, due to the increasingly serious environmental pollution problem, the world has vigorously developed new energy vehicles, and the resulting route planning issues related to new energy vehicles have also received widespread attention. Unlike the route planning of traditional fuel vehicles, new energy vehicles need to consider the charging situation, and the choice of charging piles is added to the route planning, which is more complicated than the route planning of traditional fuel vehicles. At present, most of the relevant information on route planning only considers fuel vehicles, and does not consider the charging problem of new energy vehicles. A small number of relevant materials have proposed route planning for new energy vehicles, but only combined the current mileage, vehicle power data and destination location to plan the driving route, and did not consider the change in the operating status of the charging pile, resulting in increased charging waiting time and travel time. Summary of the invention

[0003] In view of this, the purpose of the embodiments of the present invention is to provide a path planning method, system, device, medium and vehicle for a new energy vehicle, which can reduce the charging waiting time and travel time.

[0004] On the one hand, an embodiment of the present invention provides a path planning method for a new energy vehicle, comprising:

[0005] Obtaining current influencing factors, and predicting the operating status of the charging pile in a future preset time period according to the current influencing factors and the trained first prediction model, wherein the current influencing factors include current time factors and current environmental factors;

[0006] Obtaining a starting point and an end point, obtaining a recommended route according to the starting point and the end point, and marking the operating status of the charging piles on the recommended route according to the predicted operating status and arrival time of the charging piles in a future preset time period;

[0007] A destination charging pile is obtained, and a driving route is obtained according to the starting point and the destination charging pile; the destination charging pile is determined according to the operating status of the marked charging pile.

[0008] Optionally, the path planning method further includes:

[0009] A driving route is obtained according to the starting point, the end point and the trained second prediction model.

[0010] Optionally, the first prediction model includes a bidirectional long short-term memory network layer, a double-layer attention mechanism layer, a Dropout layer and a fully connected layer, and predicting the operating status of the charging pile in a future preset time period according to the current influencing factors and the trained first prediction model includes:

[0011] Normalizing the current influencing factors to obtain normalized data;

[0012] Inputting the normalized data into the bidirectional long short-term memory network layer to extract first feature information;

[0013] Input the first feature information into a double-layer attention mechanism layer to obtain second feature information;

[0014] The second feature information is passed through the dropout layer and the fully connected layer in sequence to obtain the operating status of the charging pile in a preset time period in the future.

[0015] Optionally, the second prediction model includes a convolutional neural network, and the training process of the second prediction model is as follows:

[0016] Build a road network environment, initialize the vehicle position and the convolutional neural network;

[0017] Acquire the current vehicle position according to the road network environment, determine the current reward value and the next vehicle position according to the current vehicle position and the current action selected, store the current vehicle position, the current action, the current reward value and the next vehicle position into an experience pool, and update the next vehicle position to the current vehicle position until the amount of experience in the experience pool reaches a preset value;

[0018] Update the parameters of the convolutional neural network according to the experience values ​​in the experience pool.

[0019] Optionally, the path planning method further includes:

[0020] The map interface is displayed, and the operating status of the charging piles on the map interface is marked.

[0021] Optionally, the operating status of the charging pile includes busy and idle, and marking the operating status of the charging pile on the map interface includes:

[0022] If the arrived operation status is busy, the charging pile is marked with a first color;

[0023] If the arrived operating state is idle, the charging pile is marked with a second color.

[0024] On the other hand, an embodiment of the present invention provides a path planning system for a new energy vehicle, including:

[0025] The first module is used to obtain current influencing factors and predict the operating status of the charging pile in a future preset time period according to the current influencing factors and the trained first prediction model, wherein the current influencing factors include current time factors and current environmental factors;

[0026] The second module is used to obtain the starting point and the end point, obtain the recommended route according to the starting point and the end point, and identify the operating status of the charging piles on the recommended route according to the predicted operating status and arrival time of the charging piles in the future preset time period;

[0027] The third module is used to obtain the destination charging pile, and obtain the driving route according to the starting point and the destination charging pile; the destination charging pile is determined according to the operating status of the marked charging pile.

[0028] On the other hand, an embodiment of the present invention provides a path planning device for a new energy vehicle, comprising:

[0029] at least one processor;

[0030] at least one memory for storing at least one program;

[0031] When the at least one program is executed by the at least one processor, the at least one processor implements the above-mentioned path planning method.

[0032] On the other hand, an embodiment of the present invention provides a computer-readable storage medium, in which a program executable by a processor is stored. When the program executable by the processor is executed by the processor, it is used to perform the above method.

[0033] On the other hand, an embodiment of the present invention provides a vehicle, which includes the above-mentioned path planning system or the above-mentioned path planning device.

[0034] Implementation of the embodiments of the present invention includes the following beneficial effects: the path planning method for new energy vehicles provided in the embodiment, first, obtains current influencing factors, and predicts the operating status of the charging pile in a future preset time period based on the current influencing factors and a trained first prediction model, the current influencing factors include current time factors and current environmental factors, and predicts the operating status of the charging pile in a future preset time period based on the current influencing factors, thereby improving the real-time and accuracy of the prediction of the operating status of the charging pile, then, obtains the starting point and the end point, obtains a recommended route based on the starting point and the end point, identifies the operating status of the charging pile on the recommended route based on the predicted operating status of the charging pile in the future preset time period and the arrival time, and selects available charging piles, and finally, obtains the destination charging pile, obtains the driving route based on the starting point and the destination charging pile, and the destination charging pile is determined based on the operating status of the identified charging pile, thereby selecting the most suitable destination charging pile and driving route, and reducing charging waiting time and travel time. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a flowchart of the steps of a path planning method for a new energy vehicle provided by an embodiment of the present invention;

[0036] Figure 2 A flowchart of the steps of predicting the operation status of a charging pile in a preset time period in the future by a first prediction model provided by an embodiment of the present invention;

[0037] Figure 3 It is a schematic diagram of the structure of a bidirectional long short-term memory network layer provided by an embodiment of the present invention;

[0038] Figure 4 is a structural schematic diagram of a channel attention module provided by an embodiment of the present invention;

[0039] Figure 5 is a schematic diagram of the structure of a spatial attention module provided by an embodiment of the present invention;

[0040] Figure 6 is a flowchart of the steps of training a second prediction model provided by an embodiment of the present invention;

[0041] Figure 7 is a step flow chart of a DQN algorithm provided by an embodiment of the present invention;

[0042] Figure 8 is a schematic diagram of a driving path provided by an embodiment of the present invention;

[0043] Fig. 9 is a flowchart of another path planning method for a new energy vehicle provided by an embodiment of the present invention;

[0044] Fig.10 It is a structural block diagram of a path planning system for a new energy vehicle provided by an embodiment of the present invention;

[0045] Fig.11 It is a structural block diagram of a path planning device for a new energy vehicle provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0046] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The step numbers in the following embodiments are only provided for the convenience of explanation and description, and the order between the steps is not limited in any way. The execution order of each step in the embodiment can be adaptively adjusted according to the understanding of those skilled in the art.

[0047] It should be noted that, although the functional module division is performed in the device schematic diagram and the logical order is shown in the flow chart, in some cases, the steps shown or described can be performed in a different order from the module division in the device or the order in the flow chart. The terms "first", "second", etc. in the specification and claims and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0048] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0049] The DQN (Deep Q-Network) algorithm is a reinforcement learning algorithm based on deep learning. Its core idea is to combine the Q-learning algorithm with a deep network. The process of the DQN algorithm mainly includes the following steps: initialization and training until the preset training rounds are reached or other stopping conditions are met. DQN has three important improvements over traditional reinforcement learning algorithms: 1. Introducing neural networks in deep learning, using neural networks to fit the Q table in Q-learning, solving the "dimensionality curse" problem in Q-learning when the state dimension is too high; 2. Fixing the Q target network, using the delayed updated target network to calculate the target Q value, greatly improving the stability and convergence of network training; 3. Introducing the experience replay mechanism, so that the input data conforms to independent and identical distribution when the network is updated, breaking the correlation between data.

[0050] See also Figure 1 , an embodiment of the present invention provides a path planning method for a new energy vehicle, comprising:

[0051] S100, obtaining current influencing factors, and predicting the operation status of the charging pile in a future preset time period according to the current influencing factors and the trained first prediction model, where the current influencing factors include current time factors and current environmental factors.

[0052] Current influencing factors refer to factors that affect the predicted future operating status (idle / busy) of charging piles. Current influencing factors include but are not limited to current time factors and current environmental factors. Current time factors include but are not limited to holidays, morning and evening peak time periods, etc. Current environmental factors include but are not limited to weather factors, traffic conditions, etc. For example, when the weather is bad, users are often reluctant to travel, and the number of idle charging piles will increase; for weekdays, users usually charge in the morning and evening, while non-weekday users charge more evenly, without obvious trends, and traffic congestion and other conditions will also affect users' charging choices.

[0053] It should be noted that the specific structure of the first prediction model is determined according to the actual application, and this embodiment does not make any specific restrictions. The future preset time period is determined according to the actual application, and this embodiment does not make any specific restrictions, such as within the next 1 hour or 2 hours. The operating status includes the number of busy, idle or free charging piles, etc.

[0054] Before prediction, the first prediction model needs to be trained with sample data. The sample data includes data from different data sources, such as different weather conditions, day types (working days / non-working days), traffic conditions, charging pile locations and historical operation data within a month. The first prediction model is trained with sample data such as weather, day types, traffic and charging pile historical records. After the training is completed, the operating status of the charging pile within the next hour is predicted. Compared with a single prediction algorithm, the prediction algorithm of the embodiment of the present invention has higher accuracy and saves users’ waiting time at the charging pile.

[0055] S200, obtaining a starting point and an end point, obtaining a recommended route according to the starting point and the end point, and identifying the operating status of the charging piles on the recommended route according to the predicted operating status and arrival time of the charging piles in a future preset time period.

[0056] The starting point indicates the starting point of the trip, and the end point indicates the destination of the trip. The method of obtaining the recommended route according to the starting point and the end point is determined according to the actual application, and this embodiment does not impose any specific restrictions. For example, the starting point and the end point are input into the map software, and the recommended route and the arrival time of each waypoint fed back by the map software are obtained. Then, the arrival time of the charging pile is determined according to the arrival time of each waypoint, and the operation status of the charging pile on the recommended route is determined according to the predicted operation status of the charging pile in the future preset time period and the arrival time of the charging pile, and it is marked according to the operation status.

[0057] S300: Obtain a destination charging pile, and obtain a driving route according to the starting point and the destination charging pile; the destination charging pile is determined according to the operating status of the marked charging pile.

[0058] When a new energy vehicle needs to be charged, the user can select the destination charging pile according to the actual situation, for example, according to the operating status of each charging pile on the recommended route. After obtaining the destination charging pile input by the user, the starting point and the destination charging pile can be input into the map software to obtain the driving route recommended by the map software.

[0059] In a specific embodiment, when the car battery is low and the user needs to go to a charging pile for charging, the user can select the charging pile route in the map APP software, and combine the time required for the current vehicle to reach the charging pile predicted by the map software and the prediction algorithm of the charging pile operation status to predict the number of idle charging piles when the vehicle arrives in the future (only charging piles within 3km are marked), and mark them in the driving route. The user can choose to go to an idle charging pile on the way to charge, which gives the user greater flexibility. If other situations occur in the middle of the way and the route needs to be changed, the user can also choose a suitable route independently.

[0060] Optionally, the first prediction model includes a bidirectional long short-term memory network layer, a double-layer attention mechanism layer, a Dropout layer and a fully connected layer, and predicts the operating status of the charging pile in a preset time period in the future according to the current influencing factors and the trained first prediction model, see Figure 2 ,include:

[0061] S110, normalizing the current influencing factors to obtain normalized data.

[0062] Normalize the current influencing factors and build a complete data set. Since the data format and size are inconsistent, this difference will lead to inaccurate model prediction, so the data is first normalized. The embodiment of the present invention adopts Min-Max normalization, that is, linear transformation of the original data is performed to map the result value to [0-1]. The conversion function is as follows:

[0063] X * =(x-min) / (max-min)

[0064] Among them, X * represents normalized data, x represents the current influencing factor, max represents the maximum value among the current influencing factors, and min represents the minimum value among the current influencing factors.

[0065] S120, input the normalized data into the bidirectional long short-term memory network layer to extract the first feature information.

[0066] The normalized data is passed to the Bidirectional Long Short-Term Memory Network layer (BiLSTM layer) for training. The input sequence is processed in forward and reverse order using two independent LSTMs, so that the contextual information can be accessed at each time step. Figure 3 , Figure 3 A diagram showing the structure of the BiLSTM layer unfolded along the time axis, which has information in both positive and negative directions, which enables it to better extract information from the input data.

[0067] The calculation formula is as follows:

[0068] h t =LSTM(x t ,h t-1 )

[0069] h t '=LSTM(x t ,h t ' +1 )

[0070] y t =W1h t +W2h t '+b t

[0071] The input signal passes through the forward LSTM layer output ht and the reverse LSTM layer output h t, Together they determine the value passed into the hidden layer and obtain the output y of BiLSTM t LSTM refers to the LSTM network calculation process, x t It is represented as the input at time t, W1 and W2 are the weights of the forward input and backward input respectively, b t Optimize the parameters for bias.

[0072] Compared with the conventional LSTM network, the BiLSTM network has the characteristics of bidirectional propagation. The output at the current moment is not only related to the past state, but also affected by the future state. The BiLSTM network mainly processes the input data through two LSTM networks, one in forward order and the other in reverse order. This method can access the previous and next information at the same time, which can improve the performance of the model. However, it still has the problem of easy feature information loss, so the embodiment of the present invention combines the BiLSTM network with the dual attention mechanism. The attention mechanism can preferentially extract important feature information and prevent feature information loss. By calculating the importance of multiple elements and assigning different weights to different elements, the model pays attention to important information and improves the accuracy of the model. The dual attention mechanism adopted in the embodiment of the present invention is different from the self-attention mechanism. The dual attention mechanism simultaneously establishes global context information from the channel and spatial dimensions, improves the expression ability of the model, and extracts features more accurately, reducing data loss.

[0073] S130. Input the first feature information into the double-layer attention mechanism layer to obtain the second feature information.

[0074] The output of the BiLSTM layer is passed to the dual attention mechanism layer. The core idea of ​​the dual attention mechanism is to refine the input feature map in two stages, through the channel attention module and the spatial attention module respectively. The channel attention module is responsible for paying attention to the key information, and the spatial attention module is responsible for paying attention to the location of the key information, so as to capture the key information more comprehensively.

[0075] This attention mechanism replaces the traditional global average pooling. Figure 4 First, the maximum pooling and average pooling operations are applied to each channel of the input feature map to obtain features of the same size as the input feature map, and the sigmoid activation function is used to calculate the weights to obtain the channel attention map. The weights in the feature map are adjusted through the neural network, and the weighted formula is used to fuse the obtained weights with the original features to extract the key information in the image. Figure 5 , the spatial dimension attention also uses the maximum pooling and average pooling operations on the input features in the spatial position to obtain two spatial attention feature maps, and the average pooling and maximum pooling results are spliced ​​along the channel axis. The sigmoid function is also used to obtain the weight information of different spatial positions to obtain the spatial attention map, and it is weighted fused, and the global context information is established from the channel and spatial dimensions. The formula is as follows.

[0076] Mc(F)=σ(MLP(AvgPool(F)+MaxPool(F)))

[0077] Ms(F)=σ(f 3 × 3 [AvgPool(F);MaxPool(F)])

[0078] Among them, M C (F), M S (F) represents the channel attention parameter matrix and the spatial attention parameter matrix, σ represents the activation function, f represents the convolutional layer, [AvgPool(F); MaxPool(F)] represents the pooled feature map, and MLP (MultilayerPerceptron) represents a neural network algorithm.

[0079] S140: Pass the second feature information through the dropout layer and the fully connected layer in sequence to obtain the operating status of the charging pile in a future preset time period.

[0080] A random Dropout layer is used to avoid overfitting problems. Finally, a fully connected layer with a ReLU activation function is added, and the softmax function is used to predict the operating status of the charging pile in the preset time period in the future.

[0081] Optionally, the path planning method further includes:

[0082] S400: Obtain a driving route according to the starting point, the end point and the trained second prediction model.

[0083] The second prediction model is determined according to actual application, and this embodiment does not impose any specific restrictions. The second prediction model predicts the driving route of the new energy vehicle according to the starting point, the end point, the traffic status, etc.

[0084] In a specific embodiment, the map software APP can also combine the deep reinforcement learning DQN algorithm to recommend a reasonable driving route for users. That is, the road image is converted into a road network environment that is easier to identify, and the intersection nodes and related road information are used to express the current traffic status. This is used as the input of the model so that the model can obtain complete information about the road. This introduces a deep convolutional neural network, which extracts features from the image through the convolutional neural network, fits the state-action function, and outputs a state value function based on the current state and after taking action. In addition, a target network is introduced, and the output of the network is used as the target value estimate of the state value function network. Its parameters will not be updated in real time. Using the fixed parameters of the target network as the target value estimate can reduce the instability of the algorithm and improve the performance of the algorithm.

[0085] Optionally, see Figure 6 , the second prediction model includes a convolutional neural network, and the training process of the second prediction model is as follows:

[0086] S401, build a road network environment, initialize vehicle position and convolutional neural network;

[0087] S402, obtaining the current vehicle position according to the road network environment, determining the current reward value and the next vehicle position according to the current vehicle position and the current action selected, storing the current vehicle position, the current action, the current reward value and the next vehicle position in the experience pool, and updating the next vehicle position to the current vehicle position, until the amount of experience in the experience pool reaches a preset value;

[0088] S403. Update the parameters of the convolutional neural network according to the experience values ​​in the experience pool.

[0089] In a specific embodiment, see Figure 7First, build the road network environment, initialize the vehicle position and convolutional neural network, and then start model training. During the model training process, the model obtains the current vehicle state s, that is, the current vehicle position, through the environment, and selects an action a through the state-action function, that is, the decision made on the road selection, and chooses where to go. Then observe the current reward value r and the next decision new state s' after making the decision, that is, the position of the next state of the vehicle, and store the four-tuple (s, a, r, s') in the experience pool. After that, when the number of experiences in the experience pool reaches the preset value, the model will extract experiences with a certain probability and use these experiences to update the parameters of the neural network to obtain a better strategy. Finally, it is necessary to determine whether the vehicle has reached the end point at the next state s' and calculate the maximum estimated Q value of the current state. If the end point has not been reached, continue to allocate actions, iterate continuously, and accumulate experience and store it in the experience pool until the target vehicle reaches the end point. When the vehicle reaches the end point, the training ends and the Q value is output. When updating, the action with the largest Q value is selected for update. By continuously iterating and updating the Q value, the Q value eventually converges to the optimal value, thereby planning the optimal path.

[0090] The calculation formula is as follows.

[0091]

[0092] Among them, y t represents the value of Q at time t, R t+1 represents the reward value at time t+1, γ represents the discount factor, S t+1 represents the position of the vehicle at time t+1, a represents the decision action, θ t - Represents the parameters of the target network.

[0093] Optionally, the path planning method further includes:

[0094] S500: Display a map interface, and mark the operating status of the charging pile on the map interface.

[0095] The path planning method also includes displaying a map interface on a display and marking the operating status of the charging piles on the map interface. The marking method is determined according to the actual application and is not specifically limited in this embodiment. For example, different colors or shapes are used to distinguish them.

[0096] Optionally, the operating status of the charging pile includes busy and idle, and the operating status of the charging pile on the map interface is marked, including:

[0097] S501, if the operation status is busy, the charging pile is marked with the first color;

[0098] S502: If the operating state reaches idle, the charging pile is marked with a second color.

[0099] The first color and the second color are determined according to actual applications, and are not specifically limited in this embodiment. Figure 8 , the first color is red, the second color is green, S represents the starting point, P represents the end point, M1 represents the charging pile in the idle state, and M2 represents the charging pile in the busy state. After path planning, the driving path of the new energy vehicle is: S->M1->P.

[0100] In a specific embodiment, see Fig. 9 ,The path planning method for new energy vehicles is as follows: collect the current influencing factor data of the charging pile operation status, and then input the current influencing factor data into the first prediction model to obtain the operation status of the charging pile in the future preset time period, and display the operation status of the charging pile on the map. If the user chooses to charge at an idle charging pile by himself, the route recommended by Amap is used as the driving path. If the user charges at the charging pile recommended by default, the traffic situation and the time required to reach the destination are obtained through Amap, and the travel path is planned according to the second model (path planning model) and the DQN algorithm.

[0101] Implementation of the embodiments of the present invention includes the following beneficial effects: the path planning method for new energy vehicles provided in the embodiment, first, obtains current influencing factors, and predicts the operating status of the charging pile in a future preset time period based on the current influencing factors and a trained first prediction model, the current influencing factors include current time factors and current environmental factors, and predicts the operating status of the charging pile in a future preset time period based on the current influencing factors, thereby improving the real-time and accuracy of the prediction of the operating status of the charging pile, then, obtains the starting point and the end point, obtains a recommended route based on the starting point and the end point, identifies the operating status of the charging pile on the recommended route based on the predicted operating status of the charging pile in the future preset time period and the arrival time, and selects available charging piles, and finally, obtains the destination charging pile, obtains the driving route based on the starting point and the destination charging pile, and the destination charging pile is determined based on the operating status of the identified charging pile, thereby selecting the most suitable destination charging pile and driving route, and reducing charging waiting time and travel time.

[0102] See also Fig.10 , an embodiment of the present invention provides a path planning system for a new energy vehicle, comprising:

[0103] The first module is used to obtain current influencing factors and predict the operating status of the charging pile in a preset time period in the future based on the current influencing factors and the trained first prediction model. The current influencing factors include current time factors and current environmental factors.

[0104] The second module is used to obtain the starting point and the end point, obtain the recommended route according to the starting point and the end point, and identify the operating status of the charging piles on the recommended route according to the predicted operating status and arrival time of the charging piles in the future preset time period;

[0105] The third module is used to obtain the destination charging pile and obtain the driving route according to the starting point and the destination charging pile; the destination charging pile is determined according to the operating status of the marked charging pile.

[0106] It can be seen that the contents of the above method embodiments are all applicable to the present system embodiments, the functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0107] See also Fig.11 , an embodiment of the present invention provides a path planning device for a new energy vehicle, comprising:

[0108] at least one processor;

[0109] at least one memory for storing at least one program;

[0110] When the at least one program is executed by the at least one processor, the at least one processor implements the above-mentioned path planning method.

[0111] Among them, the memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. The memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a remote memory remotely arranged relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0112] It can be seen that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0113] In addition, the embodiment of the present application also discloses a computer program product or a computer program, which is stored in a computer-readable storage medium. The processor of the computer device can read the computer program from the computer-readable storage medium, and the processor executes the computer program, so that the computer device performs the above method.

[0114] The embodiment of the present invention further provides a computer-readable storage medium, which stores a program executable by a processor, and the program executable by the processor is used to implement the above method when executed by the processor. Similarly, the contents of the above method embodiment are all applicable to the present storage medium embodiment, and the functions specifically implemented by the present storage medium embodiment are the same as those of the above method embodiment, and the beneficial effects achieved are also the same as those achieved by the above method embodiment.

[0115] It is understood that all or some steps and systems in the disclosed method above can be implemented as software, firmware, hardware and appropriate combinations thereof. Some physical components or all physical components can be implemented as software executed by a processor, such as a central processing unit, a digital signal processor or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, and the computer-readable medium can include a computer storage medium (or a non-transitory medium) and a communication medium (or a temporary medium). As known to those of ordinary skill in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic cassette, magnetic tape, disk storage or other magnetic storage device, or any other medium that can be used to store desired information and can be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically embodies computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0116] The embodiment further provides a vehicle, the vehicle comprising the electric drive assembly of the above-mentioned path planning system (or device) or computer device (or electronic device). Specifically, the vehicle can be a private car, such as a sedan, SUV, MPV or pickup truck. The vehicle can also be an operating vehicle, such as a van, bus, small truck or large trailer. The vehicle can be a fuel vehicle or a new energy vehicle. When the vehicle is a new energy vehicle, it can be a hybrid vehicle or a pure electric vehicle.

[0117] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0118] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0119] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0120] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0121] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store program codes.

[0122] The above is a specific description of the preferred implementation of the present invention, but the invention is not limited to the embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A path planning method for a new energy vehicle, characterized in that: include: Obtaining current influencing factors, and predicting the operating status of the charging pile in a future preset time period according to the current influencing factors and the trained first prediction model, wherein the current influencing factors include current time factors and current environmental factors; Obtaining a starting point and an end point, obtaining a recommended route according to the starting point and the end point, and marking the operating status of the charging piles on the recommended route according to the predicted operating status and arrival time of the charging piles in a future preset time period; Obtain a destination charging pile, and obtain a driving route according to the starting point and the destination charging pile; The destination charging pile is determined according to the operating status of the identified charging pile.

2. The path planning method according to claim 1, characterized in that: The path planning method further includes: A driving route is obtained according to the starting point, the end point and the trained second prediction model.

3. The path planning method according to claim 1, characterized in that: The first prediction model includes a bidirectional long short-term memory network layer, a double-layer attention mechanism layer, a Dropout layer and a fully connected layer. The operation status of the charging pile in a preset time period in the future is predicted according to the current influencing factors and the trained first prediction model, including: Normalizing the current influencing factors to obtain normalized data; Inputting the normalized data into the bidirectional long short-term memory network layer to extract first feature information; Input the first feature information into a double-layer attention mechanism layer to obtain second feature information; The second feature information is passed through the dropout layer and the fully connected layer in sequence to obtain the operating status of the charging pile in a preset time period in the future.

4. The path planning method according to claim 2, characterized in that: The second prediction model includes a convolutional neural network, and the training process of the second prediction model is as follows: Build a road network environment, initialize the vehicle position and the convolutional neural network; Acquire the current vehicle position according to the road network environment, determine the current reward value and the next vehicle position according to the current vehicle position and the current action selected, store the current vehicle position, the current action, the current reward value and the next vehicle position into an experience pool, and update the next vehicle position to the current vehicle position until the amount of experience in the experience pool reaches a preset value; Update the parameters of the convolutional neural network according to the experience values ​​in the experience pool.

5. The path planning method according to claim 1, characterized in that: The path planning method further includes: The map interface is displayed, and the operating status of the charging piles on the map interface is marked.

6. The path planning method according to claim 5, characterized in that: The operating status of the charging pile includes busy and idle. The marking of the operating status of the charging pile on the map interface includes: If the arrived operation status is busy, the charging pile is marked with a first color; If the arrived operating state is idle, the charging pile is marked with a second color.

7. A path planning system for new energy vehicles, characterized in that: include: The first module is used to obtain current influencing factors and predict the operating status of the charging pile in a future preset time period according to the current influencing factors and the trained first prediction model, wherein the current influencing factors include current time factors and current environmental factors; The second module is used to obtain the starting point and the end point, obtain the recommended route according to the starting point and the end point, and identify the operating status of the charging piles on the recommended route according to the predicted operating status and arrival time of the charging piles in the future preset time period; The third module is used to obtain a destination charging pile and obtain a driving route according to the starting point and the destination charging pile; The destination charging pile is determined according to the operating status of the identified charging pile.

8. A path planning device for a new energy vehicle, characterized in that: include: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a program executable by a processor, characterized in that: The processor-executable program is used to perform the method according to any one of claims 1 to 6 when executed by the processor.

10. A vehicle, characterized in that: The vehicle comprises the path planning system according to claim 7 or the path planning device according to claim 8.

Images