Vehicle speed prediction model construction method and vehicle speed prediction method
By constructing a vehicle speed prediction model with multiple algorithms, the problem of insufficient vehicle speed prediction in the existing technology is solved, the prediction accuracy is improved, and the vehicle's advance feedback on future working conditions is achieved, ensuring driving safety and energy saving.
Patent Information
- Application Number
- CN202411811140.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-10
- Publication Date
- 2025-03-28
AI Technical Summary
The lack of technology in the prior art to predict vehicle speed based on vehicle speed prediction fusion algorithms, resulting in insufficient prediction in vehicle control.
By constructing a vehicle speed prediction model, using multiple algorithms (LSM curve fitting, WMA moving average window, gated cycle unit and time prediction model) to predict the historical vehicle speed, multiple predicted vehicle speeds are obtained, and then the weight coefficient is calculated based on these predicted vehicle speeds to generate the final vehicle speed prediction model.
The accuracy of vehicle speed prediction is improved, allowing the vehicle to provide feedback on possible working conditions in the future in advance, thereby achieving stable and safe driving and reducing fuel consumption.
Smart Images
Figure CN119720774A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of engine system diagnosis, and in particular to a vehicle speed prediction model construction method and a vehicle speed prediction method. Background Art
[0002] The development of electronic control technology has made vehicles move towards mobile terminals. Based on the prediction results, the ECU is controlled to open or close related functions in advance, making the vehicle more intelligent, and predicting the possible operating conditions of the vehicle in the short term in the future, so that the vehicle can give feedback in advance on possible changes in the future, thereby achieving smooth control and operating in a more efficient and energy-saving manner. However, the prediction algorithm is only used in the fields of machine vision, intelligent AI, etc.
[0003] In the prior art, there is no vehicle speed prediction fusion algorithm for predicting vehicle speed, but there is a related demand in actual vehicle control. At present, the relevant prediction technology is only involved in related directions such as machine vision, and there is no relevant involvement in the field of vehicle speed prediction. Therefore, the prior art lacks the technical problem of predicting vehicle speed based on the vehicle speed prediction fusion algorithm. Summary of the invention
[0004] In view of this, it is necessary to provide a vehicle speed prediction model construction method and a vehicle speed prediction method to solve the technical problem that the prior art lacks a vehicle speed prediction based on a vehicle speed prediction fusion algorithm.
[0005] In order to solve the above technical problems, on the one hand, the present invention provides a method for constructing a vehicle speed prediction model, comprising: Obtain historical vehicle speeds, and input the historical vehicle speeds into a preset neural network, predict the historical vehicle speeds based on LSM curve fitting, and obtain a first predicted vehicle speed; Predict the historical vehicle speed based on the WMA moving average window to obtain a second predicted vehicle speed; Predicting the historical vehicle speed based on the gated recurrent unit to obtain a third predicted vehicle speed; Predicting the historical vehicle speed based on the time prediction model to obtain a fourth predicted vehicle speed; Obtaining a weight coefficient based on the first predicted vehicle speed, the second predicted vehicle speed, the third predicted vehicle speed, and the fourth predicted vehicle speed; Substitute the weight coefficient into the preset neural network to obtain the vehicle speed prediction model.
[0006] In a possible implementation, predicting the historical vehicle speed based on LSM curve fitting to obtain the first predicted vehicle speed includes: Normalizing the historical vehicle speed to obtain normalized data; The normalized data is predicted based on LSM curve fitting to obtain a first predicted vehicle speed.
[0007] In a possible implementation, predicting the historical vehicle speed based on the WMA moving average window to obtain the second predicted vehicle speed includes: Normalizing the historical vehicle speed to obtain normalized data; The window weight is defined and predicted based on the WMA moving average window to obtain the second predicted vehicle speed.
[0008] In a possible implementation, predicting the historical vehicle speed based on the gated loop unit to obtain the third predicted vehicle speed includes: Normalize and preprocess the historical vehicle speed to obtain preprocessed data; The preprocessed data is predicted based on the gated recurrent unit to obtain a third predicted vehicle speed.
[0009] In a possible implementation, predicting the historical vehicle speed based on the time prediction model to obtain the fourth predicted vehicle speed includes: Embedding the historical vehicle speed to obtain embedded processing data; The embedded processed data is predicted based on the time prediction model to obtain a fourth predicted vehicle speed.
[0010] In a possible implementation, the weight coefficient is obtained based on the first predicted vehicle speed, the second predicted vehicle speed, the third predicted vehicle speed and the fourth predicted vehicle speed, including: Based on the first predicted vehicle speed, the second predicted vehicle speed, the third predicted vehicle speed and the fourth predicted vehicle speed, they are respectively matched with the preset standard vehicle speed to obtain the first weight coefficient corresponding to the first predicted vehicle speed, the second weight coefficient corresponding to the second predicted vehicle speed, the third weight coefficient corresponding to the third predicted vehicle speed and the fourth weight coefficient corresponding to the fourth predicted vehicle speed.
[0011] On the other hand, the present invention also provides a vehicle speed prediction method, comprising: Get the real-time speed of the current vehicle; Inputting the current real-time speed of the vehicle into a speed prediction model, wherein the speed prediction model is determined according to any of the above-mentioned speed prediction model construction methods; The vehicle speed prediction model outputs the predicted vehicle speed in the future.
[0012] On the other hand, the present invention also provides a vehicle speed prediction model construction device, comprising: A first vehicle speed prediction unit is used to obtain a historical vehicle speed, input the historical vehicle speed into a preset neural network, and predict the historical vehicle speed based on LSM curve fitting to obtain a first predicted vehicle speed; A second vehicle speed prediction unit, used for predicting the historical vehicle speed based on the WMA moving average window to obtain a second predicted vehicle speed; A third vehicle speed prediction unit, used to predict the historical vehicle speed based on the gated loop unit to obtain a third predicted vehicle speed; A fourth vehicle speed prediction unit, used for predicting the historical vehicle speed based on the time prediction model to obtain a fourth predicted vehicle speed; A weight coefficient unit, used to obtain a weight coefficient based on the first predicted vehicle speed, the second predicted vehicle speed, the third predicted vehicle speed and the fourth predicted vehicle speed; The vehicle speed prediction model unit is used to substitute the weight coefficient into a preset neural network to obtain a vehicle speed prediction model.
[0013] On the other hand, the present invention also provides an electronic device, including a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps in the vehicle speed prediction model construction method in any of the above implementations.
[0014] On the other hand, the present invention also provides a computer-readable storage medium for storing computer-readable programs or instructions, which, when executed by a processor, can implement the steps in the vehicle speed prediction model construction method described in any of the above-mentioned implementation methods.
[0015] The beneficial effect of the present invention is as follows: the vehicle speed prediction model construction method provided by the present invention includes: obtaining the historical vehicle speed, and inputting the historical vehicle speed into a preset neural network, predicting the historical vehicle speed based on LSM curve fitting to obtain a first predicted vehicle speed; predicting the historical vehicle speed based on the WMA moving average window to obtain a second predicted vehicle speed; predicting the historical vehicle speed based on the gated loop unit to obtain a third predicted vehicle speed; predicting the historical vehicle speed based on the time prediction model to obtain a fourth predicted vehicle speed; obtaining a weight coefficient based on the first predicted vehicle speed, the second predicted vehicle speed, the third predicted vehicle speed and the fourth predicted vehicle speed; substituting the weight coefficient into the preset neural network to obtain a vehicle speed prediction model. The present invention mainly constructs a vehicle speed prediction model, and the vehicle speed prediction model is obtained by training with multiple algorithms (LSM curve fitting, WMA moving average window, gated loop unit and time prediction model). Predicting the future vehicle speed based on the constructed vehicle speed prediction model can improve the vehicle speed prediction accuracy. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.
[0017] Figure 1 A schematic diagram of a flow chart of an embodiment of a method for constructing a vehicle speed prediction model provided by the present invention; Figure 2 A structural diagram of a fully trained vehicle speed prediction model of an embodiment of the vehicle speed prediction model construction method provided by the present invention; Figure 3 A structural diagram of a gated recurrent unit in a neural network of an embodiment of a method for constructing a vehicle speed prediction model provided by the present invention; Figure 4 A schematic flow chart of an embodiment of a vehicle speed prediction method provided by the present invention; Figure 5 A schematic diagram of the structure of an embodiment of a vehicle speed prediction model building device provided by the present invention; Figure 6 A schematic structural diagram of an embodiment of an electronic device provided by the present invention. DETAILED DESCRIPTION
[0018] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.
[0019] In the description of the embodiments of the present application, unless otherwise specified, “plurality” means two or more than two.
[0020] The terms "including" and "having" and any variations thereof in the embodiments of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, apparatus, product or equipment comprising a series of steps or modules is not necessarily limited to those steps or modules clearly listed, but may include other steps or modules not clearly listed or inherent to these processes, methods, products or equipment.
[0021] The naming or numbering of the steps in the embodiments of the present invention does not mean that the steps in the method flow must be executed in the time / logical sequence indicated by the naming or numbering. The execution order of the named or numbered process steps can be changed according to the technical purpose to be achieved, as long as the same or similar technical effects can be achieved.
[0022] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.
[0023] The present invention provides a vehicle speed prediction model construction method, device, electronic device and storage medium, which are respectively described below.
[0024] Figure 1 A schematic diagram of an embodiment of a vehicle speed prediction model construction method provided by the present invention and Figure 2 A structural diagram of a fully trained vehicle speed prediction model of an embodiment of the vehicle speed prediction model construction method provided by the present invention includes: S101, obtaining historical vehicle speeds, and inputting the historical vehicle speeds into a preset neural network, predicting the historical vehicle speeds based on LSM (Least Squares Method) curve fitting, and obtaining a first predicted vehicle speed; S102, predicting the historical vehicle speed based on a WMA (weighted moving average) window to obtain a second predicted vehicle speed; S103, predicting the historical vehicle speed based on the gated loop unit to obtain a third predicted vehicle speed; S104, predicting the historical vehicle speed based on the time prediction model to obtain a fourth predicted vehicle speed; S105, obtaining a weight coefficient based on the first predicted vehicle speed, the second predicted vehicle speed, the third predicted vehicle speed and the fourth predicted vehicle speed; S106: Substitute the weight coefficient into a preset neural network to obtain a vehicle speed prediction model.
[0025] It is understandable that the present invention mainly predicts the vehicle speed for a short period of time in the future, and turns on certain components (such as thermostats) in advance through the ECU (Electronic Control Unit) controller, so that the vehicle can make early feedback on the working conditions that may occur in the future, so as to achieve the purpose of stable and safe driving and reduced fuel consumption. For example, when the vehicle is about to go up a big slope, the engine water temperature may exceed the limit. In order to prevent the vehicle water temperature from overheating. It is necessary to predict the possible working condition changes of the vehicle in the next 30 seconds based on the past 45 seconds of data, and give feedback in advance. If it is predicted that the vehicle will go up a big slope in the next 30 seconds, the water temperature controller is turned on in advance at this time to reduce the engine cooling water temperature. Thereby, the water temperature will not exceed the limit when the vehicle goes up the big slope, and finally the vehicle runs smoothly without safety problems.
[0026] It can be further understood that the present invention mainly takes the first 75s after the vehicle starts running as the first stage. After obtaining the vehicle speed information of the first 45s after the vehicle starts running, the vehicle speed information of the next 30s (i.e., the 46th to 75th second) is predicted according to different algorithms (such as the traditional LSM fitting curve method, the WMA moving average window algorithm, and the relatively new gated recurrent unit, LSTM algorithm, GRU (Gated Recurrent Unit, gated recurrent unit) algorithm, iTransformer (time series) algorithm, etc.), and the actual vehicle speed data in the 30s is compared with the predicted data, and the weights of each algorithm are divided according to the amount of deviation, and the final result after the weights output by each algorithm are normalized is output, and the output information is filtered and submitted to the ECU. It serves as the basis for whether the relevant functions inside the vehicle need to be turned on in advance. The subsequent second stage is after 75s. Because the output results of several models are continuously output, the deviation between the continuously output predicted values and the actual values is continuously compared, and the weights of each model are continuously output and corrected until they are stable, so that the prediction accuracy of the model is continuously improved.
[0027] In some embodiments of the present invention, predicting the historical vehicle speed based on LSM curve fitting to obtain a first predicted vehicle speed includes: Normalizing the historical vehicle speed to obtain normalized data; The normalized data is predicted based on LSM curve fitting to obtain a first predicted vehicle speed.
[0028] It can be understood that after the vehicle speed information is imported into the LSM curve fitting algorithm, the prediction information of the next 30 seconds output by the model is compared with the actual running vehicle speed, and a deviation value is output.
[0029] It can be further understood that the present invention mainly takes the first 75s after the vehicle starts running as the first stage. After obtaining the vehicle speed information of the first 45s after the vehicle starts running, the vehicle speed information of the next 30s (i.e., the 46th to 75th second) is predicted according to different algorithms (such as the traditional LSM fitting curve method, the WMA moving average window algorithm, and the relatively new gated recurrent unit, LSTM algorithm, GRU algorithm, iTransformer algorithm, etc.), and the actual vehicle speed data in the 30s is compared with the predicted data, and the weights of each algorithm are divided according to the amount of deviation, and the final result after the weights output by each algorithm are normalized is output, and the output information is filtered and submitted to the ECU. It serves as the basis for whether the relevant functions inside the vehicle need to be turned on in advance. The subsequent second stage is after 75s. Because the output results of several models are continuously output, the deviation between the continuously output predicted values and the actual values is continuously compared, and the weights of each model are continuously output and corrected until they are stable, so that the prediction accuracy of the model is continuously improved.
[0030] It should be further explained that the present invention mainly predicts the vehicle speed for a short period of time in the future, and turns on certain components (such as thermostats) in advance through the ECU controller, so that the vehicle can make early feedback on the working conditions that may occur in the future, so as to achieve the purpose of stable and safe driving and reducing fuel consumption. For example, when the vehicle is about to go up a big slope, the engine water temperature may exceed the limit. In order to prevent the vehicle water temperature from overheating. It is necessary to predict the possible working condition changes of the vehicle in the next 30 seconds based on the past 45 seconds of data, and give feedback in advance. If it is predicted that the vehicle will go up a big slope in the next 30 seconds, the water temperature controller is turned on in advance at this time to reduce the engine cooling water temperature. Thereby, the water temperature will not exceed the limit when the vehicle goes up the big slope, and finally the vehicle runs smoothly without safety problems.
[0031] In some embodiments of the present invention, predicting the historical vehicle speed based on the WMA moving average window to obtain the second predicted vehicle speed includes: Normalizing the historical vehicle speed to obtain normalized data; The window weight is defined and predicted based on the WMA moving average window to obtain the second predicted vehicle speed.
[0032] It can be understood that after the vehicle speed information is normalized, it is imported into the WMA moving window, the weight of each window is defined (the closer to the prediction point position, the heavier the weight), and after comparing the prediction information of the last 30 seconds after the model output with the actual running vehicle speed information, a deviation value is output.
[0033] It can be further understood that the present invention mainly takes the first 75s after the vehicle starts running as the first stage. After obtaining the vehicle speed information of the first 45s after the vehicle starts running, the vehicle speed information of the next 30s (i.e., the 46th to 75th second) is predicted according to different algorithms (such as the traditional LSM fitting curve method, the WMA moving average window algorithm, and the relatively new gated recurrent unit, LSTM algorithm, GRU algorithm, iTransformer algorithm, etc.), and the actual vehicle speed data in the 30s is compared with the predicted data, and the weights of each algorithm are divided according to the amount of deviation, and the final result after the weights output by each algorithm are normalized is output, and the output information is filtered and submitted to the ECU. It serves as the basis for whether the relevant functions inside the vehicle need to be turned on in advance. The subsequent second stage is after 75s. Because the output results of several models are continuously output, the deviation between the continuously output predicted values and the actual values is continuously compared, and the weights of each model are continuously output and corrected until they are stable, so that the prediction accuracy of the model is continuously improved.
[0034] It should be further explained that the present invention mainly predicts the vehicle speed for a short period of time in the future, and turns on certain components (such as thermostats) in advance through the ECU controller, so that the vehicle can make early feedback on the working conditions that may occur in the future, so as to achieve the purpose of stable and safe driving and reducing fuel consumption. For example, when the vehicle is about to go up a big slope, the engine water temperature may exceed the limit. In order to prevent the vehicle water temperature from overheating. It is necessary to predict the possible working condition changes of the vehicle in the next 30 seconds based on the past 45 seconds of data, and give feedback in advance. If it is predicted that the vehicle will go up a big slope in the next 30 seconds, the water temperature controller is turned on in advance at this time to reduce the engine cooling water temperature. Thereby, the water temperature will not exceed the limit when the vehicle goes up the big slope, and finally the vehicle runs smoothly without safety problems.
[0035] Figure 3 A structural diagram of a gated recurrent unit in a neural network of an embodiment of a vehicle speed prediction model construction method provided by the present invention includes: Normalize and preprocess the historical vehicle speed to obtain preprocessed data; The preprocessed data is predicted based on the gated recurrent unit to obtain a third predicted vehicle speed.
[0036] It can be understood that after the vehicle speed information is normalized and preprocessed, it is imported into the GRU (gated recurrent unit in the neural network) model algorithm of the AI model, and the update gate (zt), reset gate (rt), and candidate hidden state (ht) are defined. After the parameters are defined, the GRU algorithm will output the predicted vehicle speed information 30 seconds after the input conditions of the model according to the preset parameters, and after comparing the information with the actual running speed information, it will output a deviation value.
[0037] It can be further understood that the present invention mainly takes the first 75s after the vehicle starts running as the first stage. After obtaining the vehicle speed information of the first 45s after the vehicle starts running, the vehicle speed information of the next 30s (i.e., the 46th to 75th second) is predicted according to different algorithms (such as the traditional LSM fitting curve method, the WMA moving average window algorithm, and the relatively new gated recurrent unit, LSTM algorithm, GRU algorithm, iTransformer algorithm, etc.), and the actual vehicle speed data in the 30s is compared with the predicted data, and the weights of each algorithm are divided according to the amount of deviation, and the final result after the weights output by each algorithm are normalized is output, and the output information is filtered and submitted to the ECU. It serves as the basis for whether the relevant functions inside the vehicle need to be turned on in advance. The subsequent second stage is after 75s. Because the output results of several models are continuously output, the deviation between the continuously output predicted values and the actual values is continuously compared, and the weights of each model are continuously output and corrected until they are stable, so that the prediction accuracy of the model is continuously improved.
[0038] It should be further explained that the present invention mainly predicts the vehicle speed for a short period of time in the future, and turns on certain components (such as thermostats) in advance through the ECU controller, so that the vehicle can make early feedback on the working conditions that may occur in the future, so as to achieve the purpose of stable and safe driving and reducing fuel consumption. For example, when the vehicle is about to go up a big slope, the engine water temperature may exceed the limit. In order to prevent the vehicle water temperature from overheating. It is necessary to predict the possible working condition changes of the vehicle in the next 30 seconds based on the past 45 seconds of data, and give feedback in advance. If it is predicted that the vehicle will go up a big slope in the next 30 seconds, the water temperature controller is turned on in advance at this time to reduce the engine cooling water temperature. Thereby, the water temperature will not exceed the limit when the vehicle goes up the big slope, and finally the vehicle runs smoothly without safety problems.
[0039] In some embodiments of the present invention, predicting the historical vehicle speed based on the time prediction model to obtain a fourth predicted vehicle speed includes: Embedding the historical vehicle speed to obtain embedded processing data; The embedded processed data is predicted based on the time prediction model to obtain a fourth predicted vehicle speed.
[0040] It can be understood that after the vehicle speed data information is embedded and processed (discrete data is mapped into continuous vectors), it is imported into the Transformer time prediction model algorithm of the AI model, and the calculation dimension and the number of weight matrix columns in the Transformer model are defined; on this basis, the data is initialized, and then add (residual connection) and Norm (layer normalization) processing are performed (a method to improve model training efficiency and performance), and the obtained data is then processed by the feedforward neural network (Feedforward Neural Network) and directly transmitted to the output layer, and the model outputs the predicted information of the next 30 seconds and compares it with the actual vehicle speed information, and then outputs a deviation value.
[0041] It can be further understood that the present invention mainly takes the first 75s after the vehicle starts running as the first stage. After obtaining the vehicle speed information of the first 45s after the vehicle starts running, the vehicle speed information of the next 30s (i.e., the 46th to 75th second) is predicted according to different algorithms (such as the traditional LSM fitting curve method, the WMA moving average window algorithm, and the relatively new gated recurrent unit, LSTM algorithm, GRU algorithm, iTransformer algorithm, etc.), and the actual vehicle speed data in the 30s is compared with the predicted data, and the weights of each algorithm are divided according to the amount of deviation, and the final result after the weights output by each algorithm are normalized is output, and the output information is filtered and submitted to the ECU. It serves as the basis for whether the relevant functions inside the vehicle need to be turned on in advance. The subsequent second stage is after 75s. Because the output results of several models are continuously output, the deviation between the continuously output predicted values and the actual values is continuously compared, and the weights of each model are continuously output and corrected until they are stable, so that the prediction accuracy of the model is continuously improved.
[0042] It should be further explained that the present invention mainly predicts the vehicle speed for a short period of time in the future, and turns on certain components (such as thermostats) in advance through the ECU controller, so that the vehicle can make early feedback on the working conditions that may occur in the future, so as to achieve the purpose of stable and safe driving and reducing fuel consumption. For example, when the vehicle is about to go up a big slope, the engine water temperature may exceed the limit. In order to prevent the vehicle water temperature from overheating. It is necessary to predict the possible working condition changes of the vehicle in the next 30 seconds based on the past 45 seconds of data, and give feedback in advance. If it is predicted that the vehicle will go up a big slope in the next 30 seconds, the water temperature controller is turned on in advance at this time to reduce the engine cooling water temperature. Thereby, the water temperature will not exceed the limit when the vehicle goes up the big slope, and finally the vehicle runs smoothly without safety problems.
[0043] In some embodiments of the present invention, Based on the first predicted vehicle speed, the second predicted vehicle speed, the third predicted vehicle speed and the fourth predicted vehicle speed, a weight coefficient is obtained, including: Based on the first predicted vehicle speed, the second predicted vehicle speed, the third predicted vehicle speed and the fourth predicted vehicle speed, they are respectively matched with the preset standard vehicle speed to obtain the first weight coefficient corresponding to the first predicted vehicle speed, the second weight coefficient corresponding to the second predicted vehicle speed, the third weight coefficient corresponding to the third predicted vehicle speed and the fourth weight coefficient corresponding to the fourth predicted vehicle speed.
[0044] It can be understood that after the output deviation values obtained by each model are normalized, they are divided into different weight proportional coefficients (the greater the deviation from the actual deviation value, the smaller the weight, and the overall sum is 1), and the output results of different models multiplied by the weight proportional coefficients are added as the output, and after filtering, they are used as the actual output of the predicted vehicle speed.
[0045] It can be further understood that the present invention mainly takes the first 75s after the vehicle starts running as the first stage. After obtaining the vehicle speed information of the first 45s after the vehicle starts running, the vehicle speed information of the next 30s (i.e., the 46th to 75th second) is predicted according to different algorithms (such as the traditional LSM fitting curve method, the WMA moving average window algorithm, and the relatively new gated recurrent unit, LSTM algorithm, GRU algorithm, iTransformer algorithm, etc.), and the actual vehicle speed data in the 30s is compared with the predicted data, and the weights of each algorithm are divided according to the amount of deviation, and the final result after the weights output by each algorithm are normalized is output, and the output information is filtered and submitted to the ECU. It serves as the basis for whether the relevant functions inside the vehicle need to be turned on in advance. The subsequent second stage is after 75s. Because the output results of several models are continuously output, the deviation between the continuously output predicted values and the actual values is continuously compared, and the weights of each model are continuously output and corrected until they are stable, so that the prediction accuracy of the model is continuously improved.
[0046] It should be further explained that the present invention mainly predicts the vehicle speed for a short period of time in the future, and turns on certain components (such as thermostats) in advance through the ECU controller, so that the vehicle can make early feedback on the working conditions that may occur in the future, so as to achieve the purpose of stable and safe driving and reducing fuel consumption. For example, when the vehicle is about to go up a big slope, the engine water temperature may exceed the limit. In order to prevent the vehicle water temperature from overheating. It is necessary to predict the possible working condition changes of the vehicle in the next 30 seconds based on the past 45 seconds of data, and give feedback in advance. If it is predicted that the vehicle will go up a big slope in the next 30 seconds, the water temperature controller is turned on in advance at this time to reduce the engine cooling water temperature. Thereby, the water temperature will not exceed the limit when the vehicle goes up the big slope, and finally the vehicle runs smoothly without safety problems.
[0047] Figure 4 A schematic flow chart of an embodiment of a vehicle speed prediction method provided by the present invention includes: S401, obtaining the real-time speed of the current vehicle; S402, inputting the current real-time speed of the vehicle into a speed prediction model, wherein the speed prediction model is determined by the speed prediction model construction method described above; S403: The vehicle speed prediction model outputs the predicted vehicle speed in the future.
[0048] It can be understood that the output values of each model obtained 75 seconds after the vehicle starts running are continuously compared with the actual values, and the output deviation values obtained are normalized. Then, the weights of each model algorithm are continuously corrected according to the size of the deviation, so that the prediction model accuracy becomes higher and higher until it stabilizes.
[0049] It can be further understood that the present invention mainly takes the first 75s after the vehicle starts running as the first stage. After obtaining the vehicle speed information of the first 45s after the vehicle starts running, the vehicle speed information of the next 30s (i.e., the 46th to 75th second) is predicted according to different algorithms (such as the traditional LSM fitting curve method, the WMA moving average window algorithm, and the relatively new gated recurrent unit, LSTM algorithm, GRU algorithm, iTransformer algorithm, etc.), and the actual vehicle speed data in the 30s is compared with the predicted data, and the weights of each algorithm are divided according to the amount of deviation, and the final result after the weights output by each algorithm are normalized is output, and the output information is filtered and submitted to the ECU. It serves as the basis for whether the relevant functions inside the vehicle need to be turned on in advance. The subsequent second stage is after 75s. Because the output results of several models are continuously output, the deviation between the continuously output predicted values and the actual values is continuously compared, and the weights of each model are continuously output and corrected until they are stable, so that the prediction accuracy of the model is continuously improved.
[0050] It should be further explained that the present invention mainly predicts the vehicle speed for a short period of time in the future, and turns on certain components (such as thermostats) in advance through the ECU controller, so that the vehicle can make early feedback on the working conditions that may occur in the future, so as to achieve the purpose of stable and safe driving and reducing fuel consumption. For example, when the vehicle is about to go up a big slope, the engine water temperature may exceed the limit. In order to prevent the vehicle water temperature from overheating. It is necessary to predict the possible working condition changes of the vehicle in the next 30 seconds based on the past 45 seconds of data, and give feedback in advance. If it is predicted that the vehicle will go up a big slope in the next 30 seconds, the water temperature controller is turned on in advance at this time to reduce the engine cooling water temperature. Thereby, the water temperature will not exceed the limit when the vehicle goes up the big slope, and finally the vehicle runs smoothly without safety problems.
[0051] Figure 5 A schematic diagram of the structure of an embodiment of a vehicle speed prediction model building device provided by the present invention includes: The first vehicle speed prediction unit 501 is used to obtain the historical vehicle speed, and input the historical vehicle speed into a preset neural network, and predict the historical vehicle speed based on LSM curve fitting to obtain a first predicted vehicle speed; A second vehicle speed prediction unit 502 is used to predict the historical vehicle speed based on the WMA moving average window to obtain a second predicted vehicle speed; A third vehicle speed prediction unit 503, configured to predict the historical vehicle speed based on the gated loop unit to obtain a third predicted vehicle speed; A fourth vehicle speed prediction unit 504 is used to predict the historical vehicle speed based on the time prediction model to obtain a fourth predicted vehicle speed; A weight coefficient unit 505, used to obtain a weight coefficient based on the first predicted vehicle speed, the second predicted vehicle speed, the third predicted vehicle speed and the fourth predicted vehicle speed; The vehicle speed prediction model unit 506 is used to substitute the weight coefficient into a preset neural network to obtain a vehicle speed prediction model.
[0052] The vehicle speed prediction device 500 provided in the above embodiment can implement the technical solution described in the above vehicle speed prediction model construction method embodiment. The specific implementation principles of the above modules or units can refer to the corresponding contents in the above vehicle speed prediction method embodiment, which will not be repeated here.
[0053] like Figure 6 As shown, the present invention also provides an electronic device 600. The electronic device 600 includes a processor 601, a memory 602 and a display 603. Figure 6 Only some components of the electronic device 600 are shown, but it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.
[0054] In some embodiments, the processor 601 may be a central processing unit (CPU), a microprocessor or other data processing chip, which is used to run the program code or process data stored in the memory 602, such as the vehicle speed prediction model construction method in the present invention.
[0055] In some embodiments, the processor 601 may be a single server or a server group. The server group may be centralized or distributed. In some embodiments, the processor 601 may be local or remote. In some embodiments, the processor 601 may be implemented in a cloud platform. In one embodiment, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-cloud, etc., or any combination thereof.
[0056] In some embodiments, the memory 602 may be an internal storage unit of the electronic device 600, such as a hard disk or memory of the electronic device 600. In other embodiments, the memory 602 may also be an external storage device of the electronic device 600, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc., equipped on the electronic device 600.
[0057] Furthermore, the memory 602 may include both an internal storage unit of the electronic device 600 and an external storage device. The memory 602 is used to store application software installed in the electronic device 600 and various data.
[0058] In some embodiments, the display 603 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, etc. The display 603 is used to display information on the electronic device 600 and to display a visual user interface. The components 601-603 of the electronic device 600 communicate with each other via a system bus.
[0059] In one embodiment, when the processor 601 executes the vehicle speed prediction model building program in the memory 602, the following steps may be implemented: Obtain historical vehicle speeds, and input the historical vehicle speeds into a preset neural network, predict the historical vehicle speeds based on LSM curve fitting, and obtain a first predicted vehicle speed; Predict the historical vehicle speed based on the WMA moving average window to obtain a second predicted vehicle speed; Predicting the historical vehicle speed based on the gated recurrent unit to obtain a third predicted vehicle speed; Predicting the historical vehicle speed based on the time prediction model to obtain a fourth predicted vehicle speed; Obtaining a weight coefficient based on the first predicted vehicle speed, the second predicted vehicle speed, the third predicted vehicle speed, and the fourth predicted vehicle speed; Substitute the weight coefficient into the preset neural network to obtain the vehicle speed prediction model.
[0060] It should be understood that: when the processor 601 executes the vehicle speed prediction model building program in the memory 602, in addition to the above functions, other functions can also be implemented. For details, please refer to the description of the corresponding method embodiment above.
[0061] Furthermore, the embodiment of the present invention does not specifically limit the type of the electronic device 600 mentioned, and the electronic device 600 may be a portable electronic device such as a mobile phone, a tablet computer, a personal digital assistant (PDA), a wearable device, a laptop computer, etc. Exemplary embodiments of portable electronic devices include but are not limited to portable electronic devices equipped with IOS, Android, Microsoft or other operating systems. The above-mentioned portable electronic devices may also be other portable electronic devices, such as a laptop computer with a touch-sensitive surface (e.g., a touch panel). It should also be understood that in some other embodiments of the present invention, the electronic device 600 may not be a portable electronic device, but a desktop computer with a touch-sensitive surface (e.g., a touch panel).
[0062] Accordingly, an embodiment of the present application also provides a computer-readable storage medium, which is used to store computer-readable programs or instructions. When the program or instructions are executed by a processor, it can implement the steps or functions in the vehicle speed prediction model construction method provided in the above-mentioned method embodiments.
[0063] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware (such as a processor, a controller, etc.) through a computer program, and the computer program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.
[0064] The vehicle speed prediction model construction method, device, electronic device and storage medium provided by the present invention are introduced in detail above. The principles and implementation methods of the present invention are explained in detail using specific examples in this article. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.
Claims
1. A method for constructing a vehicle speed prediction model, characterized in that: include: Obtain historical vehicle speed, input the historical vehicle speed into a preset neural network, predict the historical vehicle speed based on LSM curve fitting, and obtain a first predicted vehicle speed; Predict the historical vehicle speed based on the WMA moving average window to obtain a second predicted vehicle speed; Predicting the historical vehicle speed based on the gated recurrent unit to obtain a third predicted vehicle speed; Predicting the historical vehicle speed based on the time prediction model to obtain a fourth predicted vehicle speed; Obtaining a weight coefficient based on the first predicted vehicle speed, the second predicted vehicle speed, the third predicted vehicle speed, and the fourth predicted vehicle speed; Substitute the weight coefficient into the preset neural network to obtain the vehicle speed prediction model.
2. The vehicle speed prediction model construction method according to claim 1, characterized in that: The method of predicting the historical vehicle speed based on LSM curve fitting to obtain a first predicted vehicle speed includes: Normalizing the historical vehicle speed to obtain normalized data; The normalized data is predicted based on LSM curve fitting to obtain a first predicted vehicle speed.
3. The vehicle speed prediction model construction method according to claim 1, characterized in that: The method of predicting the historical vehicle speed based on the WMA moving average window to obtain a second predicted vehicle speed includes: Normalizing the historical vehicle speed to obtain normalized data; The window weight is defined and predicted based on the WMA moving average window to obtain the second predicted vehicle speed.
4. The vehicle speed prediction model construction method according to claim 1, characterized in that: The method of predicting the historical vehicle speed based on the gated loop unit to obtain a third predicted vehicle speed includes: Normalize and preprocess the historical vehicle speed to obtain preprocessed data; The preprocessed data is predicted based on the gated recurrent unit to obtain a third predicted vehicle speed.
5. The vehicle speed prediction model construction method according to claim 1, characterized in that: The predicting of the historical vehicle speed based on the time prediction model to obtain a fourth predicted vehicle speed includes: Embedding the historical vehicle speed to obtain embedded processing data; The embedded processed data is predicted based on the time prediction model to obtain a fourth predicted vehicle speed.
6. The vehicle speed prediction model construction method according to claim 1, characterized in that: The weight coefficient is obtained based on the first predicted vehicle speed, the second predicted vehicle speed, the third predicted vehicle speed and the fourth predicted vehicle speed, including: Based on the first predicted vehicle speed, the second predicted vehicle speed, the third predicted vehicle speed and the fourth predicted vehicle speed, they are respectively matched with the preset standard vehicle speed to obtain the first weight coefficient corresponding to the first predicted vehicle speed, the second weight coefficient corresponding to the second predicted vehicle speed, the third weight coefficient corresponding to the third predicted vehicle speed and the fourth weight coefficient corresponding to the fourth predicted vehicle speed.
7. A vehicle speed prediction method, characterized in that: include: Get the real-time speed of the current vehicle; Inputting the current real-time vehicle speed into a vehicle speed prediction model, wherein the vehicle speed prediction model is determined according to the vehicle speed prediction model construction method according to any one of claims 1 to 6; The vehicle speed prediction model outputs the predicted vehicle speed in the future.
8. A vehicle speed prediction model construction device, characterized in that: include: A first vehicle speed prediction unit is used to obtain a historical vehicle speed, input the historical vehicle speed into a preset neural network, and predict the historical vehicle speed based on LSM curve fitting to obtain a first predicted vehicle speed; A second vehicle speed prediction unit, used for predicting the historical vehicle speed based on the WMA moving average window to obtain a second predicted vehicle speed; A third vehicle speed prediction unit, used to predict the historical vehicle speed based on the gated loop unit to obtain a third predicted vehicle speed; A fourth vehicle speed prediction unit, used for predicting the historical vehicle speed based on the time prediction model to obtain a fourth predicted vehicle speed; A weight coefficient unit, used to obtain a weight coefficient based on the first predicted vehicle speed, the second predicted vehicle speed, the third predicted vehicle speed and the fourth predicted vehicle speed; The vehicle speed prediction model unit is used to substitute the weight coefficient into a preset neural network to obtain a vehicle speed prediction model.
9. An electronic device, characterized in that: comprising a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps in the vehicle speed prediction model construction method as described in any one of claims 1 to 6 above.
10. A computer-readable storage medium, characterized in that: Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps in the vehicle speed prediction model construction method described in any one of claims 1 to 6 above.